diff --git a/README.md b/README.md index 878a383..578c3df 100644 --- a/README.md +++ b/README.md @@ -59,6 +59,8 @@ $ pip install -r requirements.txt ``` ### Using Conda +⚠️ If you're using Apple with M1 Chip, please follow these [instructions](#note-for-conda-on-apple-m1-chip) + You can create an `pydata-global-2022-ml-repro` conda environment executing: ``` @@ -77,7 +79,7 @@ You might also only update your current environment using: $ conda env update --prefix ./env --file environment.yml --prune ``` -#### Note for Conda nn Apple M1 Chip +#### Note for Conda on Apple M1 Chip If you're using a Mac with the latest M1 chip, it is highly recommended to install the packages in your conda environment specifically tailored for your hardware architecture (i.e. `arm64`). @@ -124,7 +126,7 @@ So how do we actually go about obtaining these goals? ## Data -[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) +[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/0%20-%20Basic%20Data%20Prep%20and%20Model.ipynb) This tutorial uses the [Palmer Penguins dataset](https://allisonhorst.github.io/palmerpenguins/). @@ -137,7 +139,7 @@ Data were collected and made available by [Dr. Kristen Gorman](https://www.uaf.e ## Model Evaluation -[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/1%20-%20Model%20Evaluation.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/1%20-%20Model%20Evaluation.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/1%20-%20Model%20Evaluation.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/1%20-%20Model%20Evaluation.ipynb) +[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/1%20-%20Model%20Evaluation.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/1%20-%20Model%20Evaluation.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/1%20-%20Model%20Evaluation.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/1%20-%20Model%20Evaluation.ipynb) Applying machine learning in an applied science context is often method work. We build a prototype model and expect want to show that this method can be applied to our specific problem. This means that we have to guarantee that the insights we glean from this application generalize to new data from the same problem set. @@ -153,7 +155,7 @@ So we’ll go into some methods to properly evaluate machine learning models eve ## Benchmarking -[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/2%20-%20Benchmarking.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/2%20-%20Benchmarking.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/2%20-%20Benchmarking.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/2%20-%20Benchmarking.ipynb) +[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/2%20-%20Benchmarking.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/2%20-%20Benchmarking.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/2%20-%20Benchmarking.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/2%20-%20Benchmarking.ipynb) Another common reason for rejections of machine learning papers in applied science is the lack of proper benchmarks. This section will be fairly short, as it differs from discipline to discipline. @@ -165,7 +167,7 @@ However, any time we apply a superfancy deep neural network, we need to supply a ## Model Sharing -[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/3%20-%20Model%20Sharing.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/3%20-%20Model%20Sharing.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/3%20-%20Model%20Sharing.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/3%20-%20Model%20Sharing.ipynb) +[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/3%20-%20Model%20Sharing.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/3%20-%20Model%20Sharing.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/3%20-%20Model%20Sharing.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/3%20-%20Model%20Sharing.ipynb) Some journals will require the sharing of code or models, but even if they don’t we might benefit from it. @@ -184,7 +186,7 @@ In this section, we explore how we can export models and make our training codes ## Testing -[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/4%20-%20Testing.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/4%20-%20Testing.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/4%20-%20Testing.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/4%20-%20Testing.ipynb) +[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/4%20-%20Testing.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/4%20-%20Testing.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/4%20-%20Testing.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/4%20-%20Testing.ipynb) Machine learning is very hard to test. Due to the nature of the our models, we often have soft failures in the model that are difficult to test against. @@ -200,7 +202,7 @@ Writing software tests in science, is already incredibly hard, so in this sectio ## Interpretability -[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/5%20-%20Interpretability.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/5%20-%20Interpretability.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/5%20-%20Interpretability.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/5%20-%20Interpretability.ipynb) +[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/5%20-%20Interpretability.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/5%20-%20Interpretability.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/5%20-%20Interpretability.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/5%20-%20Interpretability.ipynb) One way to probe the models we build is to test them against the established knowledge of domain experts. In this final section, we’ll explore how to build intuitions about our machine learning model and avoid pitfalls like spurious correlations. These methods for model interpretability increase our trust into models, but they can also serve as an additional level of reproducibility in our research and a valuable research artefact that can be discussed in a publication. @@ -214,7 +216,7 @@ This section will introduce tools like `shap`, discuss feature importance, and m ## Ablation Studies -[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/6%20-%20Ablation%20Study.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/6%20-%20Ablation%20Study.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/6%20-%20Ablation%20Study.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/master/notebooks/6%20-%20Ablation%20Study.ipynb) +[![](https://img.shields.io/badge/view-notebook-orange)](notebooks/6%20-%20Ablation%20Study.ipynb) [![](https://img.shields.io/badge/open-colab-yellow)](https://colab.research.google.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/6%20-%20Ablation%20Study.ipynb) [![Gradient](https://assets.paperspace.io/img/gradient-badge.svg)](https://console.paperspace.com/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/6%20-%20Ablation%20Study.ipynb) [![Open%20In%20SageMaker%20Studio%20Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/jesperdramsch/ml-for-science-reproducibility-tutorial/blob/main/notebooks/6%20-%20Ablation%20Study.ipynb) Finally, the gold standard in building complex machine learning models is proving that each constituent part of the model contributes something to the proposed solution. diff --git a/.jupytext.toml b/jupytext.toml similarity index 100% rename from .jupytext.toml rename to jupytext.toml diff --git a/notebooks/0 - Basic Data Prep and Model.ipynb b/notebooks/0 - Basic Data Prep and Model.ipynb index f184f41..31d00fa 100644 --- a/notebooks/0 - Basic Data Prep and Model.ipynb +++ b/notebooks/0 - Basic Data Prep and Model.ipynb @@ -20,10 +20,10 @@ "id": "54158e1d", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:36.930901Z", - "iopub.status.busy": "2022-12-01T10:51:36.930738Z", - "iopub.status.idle": "2022-12-01T10:51:36.936141Z", - "shell.execute_reply": "2022-12-01T10:51:36.935600Z" + "iopub.execute_input": "2022-12-02T12:08:36.427016Z", + "iopub.status.busy": "2022-12-02T12:08:36.426566Z", + "iopub.status.idle": "2022-12-02T12:08:36.435106Z", + "shell.execute_reply": "2022-12-02T12:08:36.434696Z" } }, "outputs": [], @@ -40,10 +40,10 @@ "id": "36b24fd4", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:36.937997Z", - "iopub.status.busy": "2022-12-01T10:51:36.937897Z", - "iopub.status.idle": "2022-12-01T10:51:40.292581Z", - "shell.execute_reply": "2022-12-01T10:51:40.292295Z" + "iopub.execute_input": "2022-12-02T12:08:36.437449Z", + "iopub.status.busy": "2022-12-02T12:08:36.437298Z", + "iopub.status.idle": "2022-12-02T12:08:36.752040Z", + "shell.execute_reply": "2022-12-02T12:08:36.751780Z" } }, "outputs": [], @@ -57,10 +57,10 @@ "id": "01e133b7", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:40.294318Z", - "iopub.status.busy": "2022-12-01T10:51:40.294218Z", - "iopub.status.idle": "2022-12-01T10:51:40.307229Z", - "shell.execute_reply": "2022-12-01T10:51:40.306947Z" + "iopub.execute_input": "2022-12-02T12:08:36.753558Z", + "iopub.status.busy": "2022-12-02T12:08:36.753477Z", + "iopub.status.idle": "2022-12-02T12:08:36.767488Z", + "shell.execute_reply": "2022-12-02T12:08:36.767137Z" }, "scrolled": true }, @@ -271,10 +271,10 @@ "id": "93eedeb8", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:40.308822Z", - "iopub.status.busy": "2022-12-01T10:51:40.308746Z", - "iopub.status.idle": "2022-12-01T10:51:40.315078Z", - "shell.execute_reply": "2022-12-01T10:51:40.314824Z" + "iopub.execute_input": "2022-12-02T12:08:36.768880Z", + "iopub.status.busy": "2022-12-02T12:08:36.768799Z", + "iopub.status.idle": "2022-12-02T12:08:36.774601Z", + "shell.execute_reply": "2022-12-02T12:08:36.774395Z" } }, "outputs": [ @@ -460,10 +460,10 @@ "id": "8378dc03", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:40.316459Z", - "iopub.status.busy": "2022-12-01T10:51:40.316386Z", - "iopub.status.idle": "2022-12-01T10:51:47.226742Z", - "shell.execute_reply": "2022-12-01T10:51:47.226405Z" + "iopub.execute_input": "2022-12-02T12:08:36.775953Z", + "iopub.status.busy": "2022-12-02T12:08:36.775892Z", + "iopub.status.idle": "2022-12-02T12:08:38.226669Z", + "shell.execute_reply": "2022-12-02T12:08:38.226416Z" } }, "outputs": [ @@ -502,10 +502,10 @@ "id": "791232d7", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:47.228421Z", - "iopub.status.busy": "2022-12-01T10:51:47.228337Z", - "iopub.status.idle": "2022-12-01T10:51:47.234278Z", - "shell.execute_reply": "2022-12-01T10:51:47.234048Z" + "iopub.execute_input": "2022-12-02T12:08:38.228486Z", + "iopub.status.busy": "2022-12-02T12:08:38.228405Z", + "iopub.status.idle": "2022-12-02T12:08:38.233604Z", + "shell.execute_reply": "2022-12-02T12:08:38.233388Z" } }, "outputs": [ @@ -677,10 +677,10 @@ "id": "44aaf953", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:47.235803Z", - "iopub.status.busy": "2022-12-01T10:51:47.235732Z", - "iopub.status.idle": "2022-12-01T10:51:47.239270Z", - "shell.execute_reply": "2022-12-01T10:51:47.239016Z" + "iopub.execute_input": "2022-12-02T12:08:38.234920Z", + "iopub.status.busy": "2022-12-02T12:08:38.234865Z", + "iopub.status.idle": "2022-12-02T12:08:38.238379Z", + "shell.execute_reply": "2022-12-02T12:08:38.238172Z" } }, "outputs": [], @@ -712,10 +712,10 @@ "id": "210ae85e", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:47.240766Z", - "iopub.status.busy": "2022-12-01T10:51:47.240683Z", - "iopub.status.idle": "2022-12-01T10:51:48.765086Z", - "shell.execute_reply": "2022-12-01T10:51:48.760434Z" + "iopub.execute_input": "2022-12-02T12:08:38.239692Z", + "iopub.status.busy": "2022-12-02T12:08:38.239625Z", + "iopub.status.idle": "2022-12-02T12:08:38.325291Z", + "shell.execute_reply": "2022-12-02T12:08:38.325047Z" } }, "outputs": [ @@ -748,39 +748,39 @@ " \n", " \n", " \n", - " 101\n", - " 41.0\n", - " 20.0\n", - " 203.0\n", - " MALE\n", + " 166\n", + " 45.8\n", + " 14.6\n", + " 210.0\n", + " FEMALE\n", " \n", " \n", - " 82\n", - " 36.7\n", - " 18.8\n", - " 187.0\n", - " FEMALE\n", + " 0\n", + " 39.1\n", + " 18.7\n", + " 181.0\n", + " MALE\n", " \n", " \n", - " 300\n", - " 46.7\n", - " 17.9\n", - " 195.0\n", + " 272\n", + " 46.8\n", + " 14.3\n", + " 215.0\n", " FEMALE\n", " \n", " \n", - " 126\n", - " 38.8\n", - " 17.6\n", - " 191.0\n", + " 252\n", + " 48.5\n", + " 15.0\n", + " 219.0\n", " FEMALE\n", " \n", " \n", - " 144\n", - " 37.3\n", - " 16.8\n", - " 192.0\n", - " FEMALE\n", + " 336\n", + " 51.9\n", + " 19.5\n", + " 206.0\n", + " MALE\n", " \n", " \n", " ...\n", @@ -790,25 +790,25 @@ " ...\n", " \n", " \n", - " 200\n", - " 44.9\n", - " 13.3\n", - " 213.0\n", + " 246\n", + " 44.5\n", + " 14.7\n", + " 214.0\n", " FEMALE\n", " \n", " \n", - " 309\n", - " 51.0\n", - " 18.8\n", - " 203.0\n", + " 143\n", + " 40.7\n", + " 17.0\n", + " 190.0\n", " MALE\n", " \n", " \n", - " 95\n", - " 40.8\n", - " 18.9\n", - " 208.0\n", - " MALE\n", + " 6\n", + " 38.9\n", + " 17.8\n", + " 181.0\n", + " FEMALE\n", " \n", " \n", " 63\n", @@ -818,10 +818,10 @@ " MALE\n", " \n", " \n", - " 220\n", - " 43.5\n", - " 14.2\n", - " 220.0\n", + " 130\n", + " 38.5\n", + " 17.9\n", + " 190.0\n", " FEMALE\n", " \n", " \n", @@ -831,17 +831,17 @@ ], "text/plain": [ " Culmen Length (mm) Culmen Depth (mm) Flipper Length (mm) Sex\n", - "101 41.0 20.0 203.0 MALE\n", - "82 36.7 18.8 187.0 FEMALE\n", - "300 46.7 17.9 195.0 FEMALE\n", - "126 38.8 17.6 191.0 FEMALE\n", - "144 37.3 16.8 192.0 FEMALE\n", + "166 45.8 14.6 210.0 FEMALE\n", + "0 39.1 18.7 181.0 MALE\n", + "272 46.8 14.3 215.0 FEMALE\n", + "252 48.5 15.0 219.0 FEMALE\n", + "336 51.9 19.5 206.0 MALE\n", ".. ... ... ... ...\n", - "200 44.9 13.3 213.0 FEMALE\n", - "309 51.0 18.8 203.0 MALE\n", - "95 40.8 18.9 208.0 MALE\n", + "246 44.5 14.7 214.0 FEMALE\n", + "143 40.7 17.0 190.0 MALE\n", + "6 38.9 17.8 181.0 FEMALE\n", "63 41.1 18.2 192.0 MALE\n", - "220 43.5 14.2 220.0 FEMALE\n", + "130 38.5 17.9 190.0 FEMALE\n", "\n", "[233 rows x 4 columns]" ] @@ -864,10 +864,10 @@ "id": "94df8062", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:48.773221Z", - "iopub.status.busy": "2022-12-01T10:51:48.773034Z", - "iopub.status.idle": "2022-12-01T10:51:48.777923Z", - "shell.execute_reply": "2022-12-01T10:51:48.777550Z" + "iopub.execute_input": "2022-12-02T12:08:38.326717Z", + "iopub.status.busy": "2022-12-02T12:08:38.326644Z", + "iopub.status.idle": "2022-12-02T12:08:38.329761Z", + "shell.execute_reply": "2022-12-02T12:08:38.329542Z" } }, "outputs": [ @@ -897,39 +897,39 @@ " \n", " \n", " \n", - " 101\n", - " Adelie Penguin (Pygoscelis adeliae)\n", + " 166\n", + " Gentoo penguin (Pygoscelis papua)\n", " \n", " \n", - " 82\n", + " 0\n", " Adelie Penguin (Pygoscelis adeliae)\n", " \n", " \n", - " 300\n", - " Chinstrap penguin (Pygoscelis antarctica)\n", + " 272\n", + " Gentoo penguin (Pygoscelis papua)\n", " \n", " \n", - " 126\n", - " Adelie Penguin (Pygoscelis adeliae)\n", + " 252\n", + " Gentoo penguin (Pygoscelis papua)\n", " \n", " \n", - " 144\n", - " Adelie Penguin (Pygoscelis adeliae)\n", + " 336\n", + " Chinstrap penguin (Pygoscelis antarctica)\n", " \n", " \n", " ...\n", " ...\n", " \n", " \n", - " 200\n", + " 246\n", " Gentoo penguin (Pygoscelis papua)\n", " \n", " \n", - " 309\n", - " Chinstrap penguin (Pygoscelis antarctica)\n", + " 143\n", + " Adelie Penguin (Pygoscelis adeliae)\n", " \n", " \n", - " 95\n", + " 6\n", " Adelie Penguin (Pygoscelis adeliae)\n", " \n", " \n", @@ -937,8 +937,8 @@ " Adelie Penguin (Pygoscelis adeliae)\n", " \n", " \n", - " 220\n", - " Gentoo penguin (Pygoscelis papua)\n", + " 130\n", + " Adelie Penguin (Pygoscelis adeliae)\n", " \n", " \n", "\n", @@ -947,17 +947,17 @@ ], "text/plain": [ " Species\n", - "101 Adelie Penguin (Pygoscelis adeliae)\n", - "82 Adelie Penguin (Pygoscelis adeliae)\n", - "300 Chinstrap penguin (Pygoscelis antarctica)\n", - "126 Adelie Penguin (Pygoscelis adeliae)\n", - "144 Adelie Penguin (Pygoscelis adeliae)\n", + "166 Gentoo penguin (Pygoscelis papua)\n", + "0 Adelie Penguin (Pygoscelis adeliae)\n", + "272 Gentoo penguin (Pygoscelis papua)\n", + "252 Gentoo penguin (Pygoscelis papua)\n", + "336 Chinstrap penguin (Pygoscelis antarctica)\n", ".. ...\n", - "200 Gentoo penguin (Pygoscelis papua)\n", - "309 Chinstrap penguin (Pygoscelis antarctica)\n", - "95 Adelie Penguin (Pygoscelis adeliae)\n", + "246 Gentoo penguin (Pygoscelis papua)\n", + "143 Adelie Penguin (Pygoscelis adeliae)\n", + "6 Adelie Penguin (Pygoscelis adeliae)\n", "63 Adelie Penguin (Pygoscelis adeliae)\n", - "220 Gentoo penguin (Pygoscelis papua)\n", + "130 Adelie Penguin (Pygoscelis adeliae)\n", "\n", "[233 rows x 1 columns]" ] @@ -992,10 +992,10 @@ "id": "5c4e2c77", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:48.779452Z", - "iopub.status.busy": "2022-12-01T10:51:48.779368Z", - "iopub.status.idle": "2022-12-01T10:51:48.781561Z", - "shell.execute_reply": "2022-12-01T10:51:48.781193Z" + "iopub.execute_input": "2022-12-02T12:08:38.331077Z", + "iopub.status.busy": "2022-12-02T12:08:38.331004Z", + "iopub.status.idle": "2022-12-02T12:08:38.332602Z", + "shell.execute_reply": "2022-12-02T12:08:38.332409Z" } }, "outputs": [], @@ -1022,10 +1022,10 @@ "id": "1a797998", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:48.783446Z", - "iopub.status.busy": "2022-12-01T10:51:48.783331Z", - "iopub.status.idle": "2022-12-01T10:51:48.788023Z", - "shell.execute_reply": "2022-12-01T10:51:48.787764Z" + "iopub.execute_input": "2022-12-02T12:08:38.333848Z", + "iopub.status.busy": "2022-12-02T12:08:38.333782Z", + "iopub.status.idle": "2022-12-02T12:08:38.337912Z", + "shell.execute_reply": "2022-12-02T12:08:38.337706Z" } }, "outputs": [], @@ -1044,10 +1044,10 @@ "id": "b3efeca5", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:48.789376Z", - "iopub.status.busy": "2022-12-01T10:51:48.789300Z", - "iopub.status.idle": "2022-12-01T10:51:49.968355Z", - "shell.execute_reply": "2022-12-01T10:51:49.968089Z" + "iopub.execute_input": "2022-12-02T12:08:38.339167Z", + "iopub.status.busy": "2022-12-02T12:08:38.339104Z", + "iopub.status.idle": "2022-12-02T12:08:38.377060Z", + "shell.execute_reply": "2022-12-02T12:08:38.376795Z" } }, "outputs": [ @@ -1123,10 +1123,10 @@ "id": "aa8d0a2c", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:49.970182Z", - "iopub.status.busy": "2022-12-01T10:51:49.970092Z", - "iopub.status.idle": "2022-12-01T10:51:50.023470Z", - "shell.execute_reply": "2022-12-01T10:51:50.023168Z" + "iopub.execute_input": "2022-12-02T12:08:38.378434Z", + "iopub.status.busy": "2022-12-02T12:08:38.378361Z", + "iopub.status.idle": "2022-12-02T12:08:38.421713Z", + "shell.execute_reply": "2022-12-02T12:08:38.421466Z" } }, "outputs": [ @@ -1192,17 +1192,17 @@ "id": "fc341232", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:50.025166Z", - "iopub.status.busy": "2022-12-01T10:51:50.025063Z", - "iopub.status.idle": "2022-12-01T10:51:50.029499Z", - "shell.execute_reply": "2022-12-01T10:51:50.029265Z" + "iopub.execute_input": "2022-12-02T12:08:38.423137Z", + "iopub.status.busy": "2022-12-02T12:08:38.423078Z", + "iopub.status.idle": "2022-12-02T12:08:38.426875Z", + "shell.execute_reply": "2022-12-02T12:08:38.426669Z" } }, "outputs": [ { "data": { "text/plain": [ - "0.9957081545064378" + "0.9871244635193133" ] }, "execution_count": 14, @@ -1232,10 +1232,10 @@ "id": "4f28497a", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:50.030942Z", - "iopub.status.busy": "2022-12-01T10:51:50.030855Z", - "iopub.status.idle": "2022-12-01T10:51:50.035654Z", - "shell.execute_reply": "2022-12-01T10:51:50.035252Z" + "iopub.execute_input": "2022-12-02T12:08:38.428179Z", + "iopub.status.busy": "2022-12-02T12:08:38.428124Z", + "iopub.status.idle": "2022-12-02T12:08:38.431532Z", + "shell.execute_reply": "2022-12-02T12:08:38.431314Z" } }, "outputs": [ diff --git a/notebooks/1 - Model Evaluation.ipynb b/notebooks/1 - Model Evaluation.ipynb index eece173..fcff829 100644 --- a/notebooks/1 - Model Evaluation.ipynb +++ b/notebooks/1 - Model Evaluation.ipynb @@ -18,14 +18,14 @@ { "cell_type": "code", "execution_count": 1, - "id": "54158e1d", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:51.424861Z", - "iopub.status.busy": "2022-12-01T10:51:51.424500Z", - "iopub.status.idle": "2022-12-01T10:51:51.450066Z", - "shell.execute_reply": "2022-12-01T10:51:51.449650Z" - } + "iopub.execute_input": "2022-12-02T12:08:39.398672Z", + "iopub.status.busy": "2022-12-02T12:08:39.398408Z", + "iopub.status.idle": "2022-12-02T12:08:39.406601Z", + "shell.execute_reply": "2022-12-02T12:08:39.406147Z" + }, + "tags": [] }, "outputs": [], "source": [ @@ -40,12 +40,13 @@ "execution_count": 2, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:51.451874Z", - "iopub.status.busy": "2022-12-01T10:51:51.451784Z", - "iopub.status.idle": "2022-12-01T10:51:51.723045Z", - "shell.execute_reply": "2022-12-01T10:51:51.722667Z" + "iopub.execute_input": "2022-12-02T12:08:39.409200Z", + "iopub.status.busy": "2022-12-02T12:08:39.409051Z", + "iopub.status.idle": "2022-12-02T12:08:39.574837Z", + "shell.execute_reply": "2022-12-02T12:08:39.574604Z" }, - "lines_to_next_cell": 2 + "lines_to_next_cell": 2, + "tags": [] }, "outputs": [ { @@ -169,11 +170,12 @@ "execution_count": 3, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:51.725570Z", - "iopub.status.busy": "2022-12-01T10:51:51.725401Z", - "iopub.status.idle": "2022-12-01T10:51:52.008292Z", - "shell.execute_reply": "2022-12-01T10:51:52.007966Z" - } + "iopub.execute_input": "2022-12-02T12:08:39.576528Z", + "iopub.status.busy": "2022-12-02T12:08:39.576328Z", + "iopub.status.idle": "2022-12-02T12:08:39.769885Z", + "shell.execute_reply": "2022-12-02T12:08:39.769644Z" + }, + "tags": [] }, "outputs": [ { @@ -332,24 +334,30 @@ "execution_count": 4, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.010253Z", - "iopub.status.busy": "2022-12-01T10:51:52.010136Z", - "iopub.status.idle": "2022-12-01T10:51:52.421465Z", - "shell.execute_reply": "2022-12-01T10:51:52.421062Z" + "iopub.execute_input": "2022-12-02T12:08:39.771431Z", + "iopub.status.busy": "2022-12-02T12:08:39.771374Z", + "iopub.status.idle": "2022-12-02T12:08:40.043245Z", + "shell.execute_reply": "2022-12-02T12:08:40.042956Z" + } + }, + "outputs": [], + "source": [ + "from matplotlib import pyplot as plt" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:40.044922Z", + "iopub.status.busy": "2022-12-02T12:08:40.044858Z", + "iopub.status.idle": "2022-12-02T12:08:40.111764Z", + "shell.execute_reply": "2022-12-02T12:08:40.111512Z" }, - "scrolled": false + "tags": [] }, "outputs": [ - { - "data": { - "text/plain": [ - "" - ] - }, - "execution_count": 4, - "metadata": {}, - "output_type": "execute_result" - }, { "data": { "image/png": "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\n", @@ -362,7 +370,8 @@ } ], "source": [ - "penguins.groupby(\"Species\").Sex.count().plot(kind=\"bar\")" + "penguins.groupby(\"Species\").Sex.count().plot(kind=\"bar\")\n", + "plt.show()" ] }, { @@ -376,14 +385,15 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 6, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.423481Z", - "iopub.status.busy": "2022-12-01T10:51:52.423387Z", - "iopub.status.idle": "2022-12-01T10:51:52.427750Z", - "shell.execute_reply": "2022-12-01T10:51:52.427443Z" - } + "iopub.execute_input": "2022-12-02T12:08:40.113194Z", + "iopub.status.busy": "2022-12-02T12:08:40.113112Z", + "iopub.status.idle": "2022-12-02T12:08:40.116855Z", + "shell.execute_reply": "2022-12-02T12:08:40.116581Z" + }, + "tags": [] }, "outputs": [ { @@ -439,7 +449,7 @@ "Gentoo penguin (Pygoscelis papua) 90" ] }, - "execution_count": 5, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" } @@ -452,37 +462,53 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "We can address this by applying stratification.\n", - "That is simply sampling randomly within a class (or strata) rather than randomly sampling from the entire dataframe." + "We can address this by applying **stratification**.\n", + "\n", + "That is simply achieved by randomly sampling *within a class** (or strata) rather than randomly sampling from the entire dataframe." ] }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 7, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.429440Z", - "iopub.status.busy": "2022-12-01T10:51:52.429356Z", - "iopub.status.idle": "2022-12-01T10:51:52.486563Z", - "shell.execute_reply": "2022-12-01T10:51:52.486257Z" + "iopub.execute_input": "2022-12-02T12:08:40.118437Z", + "iopub.status.busy": "2022-12-02T12:08:40.118374Z", + "iopub.status.idle": "2022-12-02T12:08:40.121490Z", + "shell.execute_reply": "2022-12-02T12:08:40.121187Z" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "X, y = penguins[features], penguins[target[0]]\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=.7, random_state=42, stratify=y)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To qualitatevely assess the effect of stratification, let's plot class distribution in both _training_ and _test_ sets:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:40.123006Z", + "iopub.status.busy": "2022-12-02T12:08:40.122922Z", + "iopub.status.idle": "2022-12-02T12:08:40.220970Z", + "shell.execute_reply": "2022-12-02T12:08:40.220662Z" } }, "outputs": [ { "data": { + "image/png": "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\n", "text/plain": [ - "" - ] - }, - "execution_count": 6, - "metadata": {}, - "output_type": "execute_result" - }, - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" + "
" ] }, "metadata": {}, @@ -490,8 +516,11 @@ } ], "source": [ - "X_train, X_test, y_train, y_test = train_test_split(penguins[features], penguins[target[0]], train_size=.7, random_state=42, stratify=penguins[target[0]])\n", - "y_train.reset_index().groupby(\"Species\").count().plot(kind=\"bar\")" + "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 8))\n", + "\n", + "y_train.reset_index().groupby(\"Species\").count().plot(kind=\"bar\", ax=ax1, ylim=(0, len(y)), title=\"Training\")\n", + "y_test.reset_index().groupby(\"Species\").count().plot(kind=\"bar\", ax=ax2, ylim=(0, len(y)), title=\"Test\")\n", + "plt.show()" ] }, { @@ -503,14 +532,15 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 9, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.488475Z", - "iopub.status.busy": "2022-12-01T10:51:52.488396Z", - "iopub.status.idle": "2022-12-01T10:51:52.506509Z", - "shell.execute_reply": "2022-12-01T10:51:52.506207Z" - } + "iopub.execute_input": "2022-12-02T12:08:40.222749Z", + "iopub.status.busy": "2022-12-02T12:08:40.222661Z", + "iopub.status.idle": "2022-12-02T12:08:40.236282Z", + "shell.execute_reply": "2022-12-02T12:08:40.236063Z" + }, + "tags": [] }, "outputs": [], "source": [ @@ -541,6 +571,7 @@ "That changes however, when we have minority classes with much less data than the majority class.\n", "\n", "Either way it's worth it to keep in mind that stratification exists. The `stratify=` keyword takes any type of vector as long as it matches the dimension of the dataframe.\n", + "\n", "## Cross-Validation\n", "Cross-validation is often considered the gold standard in statistical applications and machine learning.\n", "\n", @@ -554,14 +585,15 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 10, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.508377Z", - "iopub.status.busy": "2022-12-01T10:51:52.508280Z", - "iopub.status.idle": "2022-12-01T10:51:52.531349Z", - "shell.execute_reply": "2022-12-01T10:51:52.531084Z" - } + "iopub.execute_input": "2022-12-02T12:08:40.237781Z", + "iopub.status.busy": "2022-12-02T12:08:40.237688Z", + "iopub.status.idle": "2022-12-02T12:08:40.257932Z", + "shell.execute_reply": "2022-12-02T12:08:40.257680Z" + }, + "tags": [] }, "outputs": [ { @@ -570,27 +602,29 @@ "array([1. , 1. , 0.9787234 , 0.97826087, 0.97826087])" ] }, - "execution_count": 8, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.model_selection import cross_val_score\n", + "\n", "scores = cross_val_score(model, X_train, y_train, cv=5)\n", "scores" ] }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 11, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.532951Z", - "iopub.status.busy": "2022-12-01T10:51:52.532872Z", - "iopub.status.idle": "2022-12-01T10:51:52.534948Z", - "shell.execute_reply": "2022-12-01T10:51:52.534709Z" - } + "iopub.execute_input": "2022-12-02T12:08:40.259381Z", + "iopub.status.busy": "2022-12-02T12:08:40.259297Z", + "iopub.status.idle": "2022-12-02T12:08:40.261276Z", + "shell.execute_reply": "2022-12-02T12:08:40.261069Z" + }, + "tags": [] }, "outputs": [ { @@ -616,78 +650,714 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Time-series Validation\n", + "## Model Evaluation" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Brilliant! So let's recap for a moment what we have done so far, in preparation for our (final) **Model evaluation**.\n", "\n", - "But validation can get tricky if time gets involved.\n", + "We have:\n", "\n", - "Imagine we measured the growth of baby penguin Hank over time and wanted to us machine learning to project the development of Hank. Then our data suddenly isn't i.i.d. anymore, since it is dependent in the time dimension.\n", + "- prepared the model pipeline: `sklearn.pipeline.Pipeline` with `preprocessor + model`\n", + "- generated **train** and **test** data partitions (with stratification): `(X_train, y_train)` and `(X_test, y_test)`, respectively\n", + " - stratification guaranteed that those partitions will retain class distributions\n", + "- assessed model performance via **cross validation** (i.e. `cross_val_score`) on `X_train`(!!)\n", + " - this had the objective of verifying model consistency on multiple data partitioning\n", "\n", - "Were we to split our data randomly for our training and test set, we would test on data points that lie in between training points, where even a simple linear interpolation can do a fairly decent job.\n", + "Now we need the complete our last step, namely \"assess how the model we chose in CV\" (we only had one model, so that was an easy choice :D ) will perform on _future data_!\n", + "And we have a _candidate_ as representative for these data: `X_test`.\n", "\n", - "Therefor, we need to split our measurements along the time axis\n", - "![Scikit-learn time series validation](https://scikit-learn.org/stable/_images/sphx_glr_plot_cv_indices_013.png)\n", - "*Scikit-learn Time Series CV [[Source]](https://scikit-learn.org/stable/modules/cross_validation.html#time-series-split).*" + "Please note that `X_test` has never been used so far (as it should have!). The take away message here is: _generate test partition, and forget about it until the last step!_\n", + "\n", + "\n", + "Thanks to `CV`, We have an indication of how the `SVC` classifier behaves on multiple \"version\" of the training set. We calculated an average score of `0.99` accuracy, therefore we decided this model is to be trusted for predictions on _unseen data_.\n", + "\n", + "Now all we need to do, is to prove this assertion.\n", + "\n", + "To do so we need to: \n", + " - train a new model on the entire **training set**\n", + " - evaluate it's performance on **test set** (using the metric of choice - presumably the same metric we chose in CV!)" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 12, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.536599Z", - "iopub.status.busy": "2022-12-01T10:51:52.536512Z", - "iopub.status.idle": "2022-12-01T10:51:52.539481Z", - "shell.execute_reply": "2022-12-01T10:51:52.539239Z" - }, - "lines_to_next_cell": 2 + "iopub.execute_input": "2022-12-02T12:08:40.262850Z", + "iopub.status.busy": "2022-12-02T12:08:40.262756Z", + "iopub.status.idle": "2022-12-02T12:08:40.267076Z", + "shell.execute_reply": "2022-12-02T12:08:40.266743Z" + } + }, + "outputs": [], + "source": [ + "# training\n", + "model = Pipeline(steps=[\n", + " ('preprocessor', preprocessor),\n", + " ('classifier', SVC()),\n", + "])\n", + "classifier = model.fit(X_train, y_train)" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:40.268513Z", + "iopub.status.busy": "2022-12-02T12:08:40.268455Z", + "iopub.status.idle": "2022-12-02T12:08:40.272042Z", + "shell.execute_reply": "2022-12-02T12:08:40.271816Z" + } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "TimeSeriesSplit(gap=0, max_train_size=None, n_splits=3, test_size=None)\n", - "[0 1 2] [3]\n", - "[0 1 2 3] [4]\n", - "[0 1 2 3 4] [5]\n" + "TEST ACC: 1.0\n" ] } ], "source": [ - "import numpy as np\n", - "from sklearn.model_selection import TimeSeriesSplit\n", - "\n", - "X = np.array([[1, 2], [3, 4], [1, 2], [3, 4], [1, 2], [3, 4]])\n", - "y = np.array([1, 2, 3, 4, 5, 6])\n", - "tscv = TimeSeriesSplit(n_splits=3)\n", - "print(tscv)\n", + "# Model evaluation\n", + "from sklearn.metrics import accuracy_score\n", "\n", - "for train, test in tscv.split(X):\n", - " print(\"%s %s\" % (train, test))" + "y_pred = classifier.predict(X_test)\n", + "print(\"TEST ACC: \", accuracy_score(y_true=y_test, y_pred=y_pred))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Spatial Validation\n", - "\n", - "Spatial data, like maps and satellite data has a similar problem.\n", - "\n", - "Here the data is correlated in the spatial dimension. However, we can mitigate the effect by supplying a group. In this simple example I used continents, but it's possible to group by bins on a lat-lon grid as well. \n", + "Now we can finally say that we have concluded our model evaluation - with a fantastic score of `0.96` Accuracy on the test set." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Choosing the appropriate Evaluation Metric" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Ok, now for the mere sake of considering a more realistic data scenario, let's pretend our reference dataset is composed by only samples from two (out of the three) classes we have. In particular, we will crafting our dataset by choosing the most and the least represented classes, respectively. \n", "\n", - "Here especially, a cross-validation scheme is very important, as it is used to validate against every area on your map at least once." + "The very idea is to explore whether the choice of appropriate metrics could make the difference in our machine learning models evaluation." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's recall class distributions in our dataset:" ] }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 14, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:40.273625Z", + "iopub.status.busy": "2022-12-02T12:08:40.273545Z", + "iopub.status.idle": "2022-12-02T12:08:40.276998Z", + "shell.execute_reply": "2022-12-02T12:08:40.276737Z" + } + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
index
Species
Adelie Penguin (Pygoscelis adeliae)146
Chinstrap penguin (Pygoscelis antarctica)68
Gentoo penguin (Pygoscelis papua)120
\n", + "
" + ], + "text/plain": [ + " index\n", + "Species \n", + "Adelie Penguin (Pygoscelis adeliae) 146\n", + "Chinstrap penguin (Pygoscelis antarctica) 68\n", + "Gentoo penguin (Pygoscelis papua) 120" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "y.reset_index().groupby([\"Species\"]).count()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "So let's select samples from the first two classes, `Adelie Penguin` and `Chinstrap penguin`:" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:40.278480Z", + "iopub.status.busy": "2022-12-02T12:08:40.278410Z", + "iopub.status.idle": "2022-12-02T12:08:40.281719Z", + "shell.execute_reply": "2022-12-02T12:08:40.281327Z" + } + }, + "outputs": [], + "source": [ + "samples = penguins[((penguins[\"Species\"].str.startswith(\"Adelie\")) | (penguins[\"Species\"].str.startswith(\"Chinstrap\")))]" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:40.283136Z", + "iopub.status.busy": "2022-12-02T12:08:40.283076Z", + "iopub.status.idle": "2022-12-02T12:08:40.285491Z", + "shell.execute_reply": "2022-12-02T12:08:40.285212Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "True" + ] + }, + "execution_count": 16, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "samples.shape[0] == 146 + 68 # quick verification" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To make things even harder for our machine learning model, let's also see if we could get rid of _clearly_ separating features in this toy dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:40.287001Z", + "iopub.status.busy": "2022-12-02T12:08:40.286923Z", + "iopub.status.idle": "2022-12-02T12:08:41.009130Z", + "shell.execute_reply": "2022-12-02T12:08:41.008855Z" + } + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAABD0AAALlCAYAAADHSZ0AAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3hUZfbA8e/0JJPeC5DQewfpHRQLKlgQUMHeF11Z9Wdbu+sWV107VkAFOyqoINKU3juEHtL7TMpMkpn5/fGSMslMaAlJyPk8T54l9965952s7yT33POeo3G5XC6EEEIIIYQQQgghLjDahh6AEEIIIYQQQgghRH2QoIcQQgghhBBCCCEuSBL0EEIIIYQQQgghxAVJgh5CCCGEEEIIIYS4IEnQQwghhBBCCCGEEBckCXoIIYQQQgghhBDigiRBDyGEEEIIIYQQQlyQJOghhBBCCCGEEEKIC5IEPQCXy4XFYsHlcjX0UIRoVmTuCdEwZO4JIYQQormQoAdgtVoJCgrCarU29FCEaFZk7gnRMGTuCSGEEKK5kKCHEEIIIYQQQgghLkgS9BBCCCGEEEIIIcQFSYIeQgghhBBCCCGEuCBJ0EMIIYQQQgghhBAXJH1DD0AIIUQ9KrNDYRbgAlMA+AQ19IiEaHi2fLBbQaMFvzDQmxp6REIIIYSoJxL0EEKIC1X+Cfjzf7BtLpQWQdsxMO45COsAekNDj06I86+sBLIOwNKn4fDvYPCD3jfB4AcgqEVDj04IIYQQ9UCCHkIIcSGyJMOcqyD7YOW2g7/B0dVw5yqI7NRwYxOioeQchA9GqwwogJJCWP8uHFoGN/8AgbENOz4hhBBC1Dmp6SGEEBeiE5vdAx7lyuyw/CWwF5z/MQnRkOxWWPZCZcCjqqxESNl6/sckhBBCiHonQQ8hhLjQOJ2w6xvv+w8tA7vl/I1HiMbAZlFLWrzZ/d35G4sQQgghzhtZ3iLcrDmUxbsrDrEn1UJ0oA83D0rgun4t0Gg0DT00IcTp0mrBP9L7fp9AVcBRiOZEqwVTIJQWe95vrmXOCCGEEKLJkr96BQAul4vXf0tk6uz1JOcVM7xDBL5GHY98s4NZX23H6XQ19BCFEGei943e9110l9zgiebHHAkX3el9f68p528sQgghhDhvJNNDAPD2ikP897cDXNe3BVf3jkN7MrPjz4NZvLX8IPFhZv4ypn0Dj1IIcdqC42HUk7D8BfftrQZCz8nqqbcQzYlWB72mQeISSFrvvm/M3yGoZcOMSwghhBD1SoIegpUHMvnXr/uZ1CeOSX3cW/YNaRdOSl4xry9LZHy3aDpEBTTQKIUQZ8Q3WD3V7nS5qlVgy4cuV0FYOwiIaujRCdEwAmPg+rmQnQh7flDzpOvVEBAHvkENPTohhBBC1AMJejRz+UWlzPpqOz1aBHFNtYBHuat7x7HmUDb/+mU/s6f3O88jFEKcNd8g9RXVpaFHIkTjERClvhKGNvRIhBBCCHEeSH5zM/fKr/sosJVx1/C2FUtaqjPotFzdO46le9NJTLee5xEKIYQQQgghhBBnR4Iezdiu5Hy+WH+c6/q1INRsrPXYIW3DCPY1MHfdsfM0OiGEEEIIIYQQ4txI0KMZe/nnvcQG+zKuy6nX9+t1WoZ3iOCH7SmUlDnPw+iEEEIIIYQQQohzI0GPZmrtoWz+PJjN5H4t0Z9mF4ch7cLJKyrlz4NZ9Tw6IYQQQgghhBDi3EnQo5n679IDtIkw0y8h5LRf0zLEl+hAH5bsSa/HkQkhhBBCCCGEEHVDgh7N0IYjOWw4msPE3nFovBQv9USj0dCnVTC/70vH5XLV4wiFEEIIIYQQQohzJ0GPZuidFQdpFepLn1ann+VRrnuLINItdg5lFtbDyIQQQgghhBBCiLojQY9mJjHdyvL9mVzWPdZri9radIoORK/VsOaQ1PUQQtQDmwVyDkPGXsg/AU5HQ49ICHGm7AWQc1TN47wkcJQ19IiEEEI0Y/qGHoA4vz768yghfgaGtA07q9f7GHS0jjCz8WguNw9KqNvBCSGat7zj8PNjcOBncDnBLwzGPA2drwS/0IYenRDidOQnw9KnYM/3KmjpEwwjHoUek8F8dn97CCGEEOdCgh7NSH5RKd9tOcGEnrHodWef5NMxKoCNR3LqcGRCiGbPmgbzroGsA5XbirLhx5mg91E3TGeRnSaEOI8KMmDBjZCypXKbLQ9+/T81f/vfCTpdgw1PCCFE8yTLW5qRrzYnUeZ0MbpT5Dmdp31kAGkWG+kWWx2NTAjR7GUfcg94VLXsWbCmnt/xCCHOXH6ye8CjqpWvQIHMYyGEEOefBD2aCZfLxWfrj9M/IZRgP+M5natthBmAHSfy62JoQgjh/UYJwJICpUXnbyxCiLOTudf7vuJcKJEi6EIIIc4/CXo0ExuO5HAkq5Axnc8tywMg1Gwk2NfAjhN55z4wIYQACI73vs/gBzrT+RuLEOLsBMR436fVg17msRBCiPOvUQc9nnnmGTQajdtXdHR0xf4ZM2bU2D9w4MAGHHHj9dXmE0QFmugSE3jO59JoNMSH+bEn1VIHIxNCCCC2Fxj9Pe/rOwP8zz1gK4SoZ+HtvRcd7nYtmGUeCyGEOP8addADoGvXrqSmplZ87dy5023/+PHj3fYvXry4gUbaeBWVlLFoRyrD2kegqaNCgK1C/diTIkEPIUQdCYiFm74HnyD37W1Hw+C/yBNiIZqCwDi4aWHNwEfLgTD272D0a5hxCSGEaNYaffcWvV7vlt1RnclkqnW/gKV70ikudTC0XXidnTM+zMyPO1LJLyolyM9QZ+cVQjRTOj3E9YG7/4Tsg1CYCZFdVLq8tLkUomnQaCC6O9y1WhUntqZBZCcV1PSPaOjRCSGEaKYafdAjMTGR2NhYTCYTAwYM4KWXXqJNmzYV+1esWEFkZCTBwcGMGDGCF198kchISZ+s6odtKXSI8icq0KfOztkyVD2tScyw0i/BSyqrEEKcCa0OgluqLyFE06TRQFAL9SWEEEI0Ao066DFgwADmzJlDhw4dSE9P54UXXmDw4MHs3r2bsLAwLr30Uq677jri4+M5cuQITz31FKNHj2bz5s2YTN5Toe12O3a7veJ7i+XCXaZhsZWyKjGTG/q3qtPzxgT5oNVAYkaBBD3EaWtOc0+IxkTmnhBCCCGaq0Zd0+PSSy/lmmuuoXv37owdO5ZFixYB8OmnnwIwefJkLr/8crp168aECRP4+eefOXDgQMVx3rz88ssEBQVVfLVseeE+VVy+L4NSh4sBres2MGHQaYkO8iExvaBOzysubM1p7gnRmMjcE0IIIURz1aiDHtWZzWa6d+9OYmKix/0xMTHEx8d73V/u//7v/8jPz6/4SkpKqo/hNgpLdqfTNsJMmH/dFwGMDfIlMcNa5+cVF67mNPdEI1SYBZYUsDW/LAeZe6JWZXY1N6xp4HA09GiEEEKIOtWol7dUZ7fb2bt3L8OGDfO4Pzs7m6SkJGJiaukTjyp+WtvylwtFqcPJygOZjO9WP4VeY4J82HI8r17OLS5MzWXuiUamMAuOroZV/1I3drF9YMxTENYBTOaGHt15IXNPeJV7FNa+Bbu/A50R+kyH3jdCUFxDj0wIIYSoE40602PWrFmsXLmSI0eOsH79eq699losFgvTp0+noKCAWbNmsXbtWo4ePcqKFSuYMGEC4eHhTJw4saGH3ihsPpZLgb2M3i2D6+X8McG+pOQVYyuVp0JCiEaqOB9WvwpfzYD03VCcC4eWwexRkLS+oUcnRMPKPQofjIEN76uOSZZkWPESzJuo/i2EEEJcAOol08Nut7NhwwaOHj1KUVERERER9O7dm9atW5/ReU6cOMGUKVPIysoiIiKCgQMHsm7dOuLj4ykuLmbnzp3MmTOHvLw8YmJiGDVqFAsWLCAgIKA+3laTs+pAJkG+BhLC6+dJZkyQDy7gWHYRHaPlZy6EaIQKM2HdWzW3u1zw04Nw2xIIkLbnohkqK4ENs1UmVHWZ++Hon9Dj+vM/LiGEEKKO1WnQY82aNfzvf//j+++/p6SkhODgYHx9fcnJycFut9OmTRvuvPNO7r777tMKTMyfP9/rPl9fX3799de6HP4FZ3ViFl1jA9FqNPVy/uiTLXCPZBVK0EMI0Til7fS+L+8YFOdJ0EM0T8U5sGeh9/3bv4DOV4DB7/yNSQghhKgHdba85aqrruLaa68lLi6OX3/9FavVSnZ2NidOnKCoqIjExESefPJJli1bRocOHVi6dGldXVp4YLGVsjsln26xQfV2jSBfAz4GLcdzCuvtGkIIcU4MPrXv1+rOzziEaGw02trnh8EXNDI/hBBCNH11lulx8cUX89VXX2E0Gj3ub9OmDW3atGH69Ons3r2blJSUurq08GDT0RycLugcE1hv19BoNEQF+nAsu6jeriGEaIZcLsg/AclbIG0HRPeAuD4Q1ALONHMtsosqzugoqbkvri/4hdXNmIVoaswR0P92+PlRz/sH3A36Worf5p+AlK2Quh0iO0OL/hDYArSNulycEEKIZqjOgh733XffaR/btWtXunbtWleXFh6sP5JDqNlIVGD9VuuPCpCghxCijmXsgU8uV0VHy/mGwIxFEHWGvzv8o2Die/DNrSqYUs4nGK56E/xC62TIQjQ5Gg10vgp2fAXJm9z3dbsWIjp5f23mfvjkMvd6IKZAmPETxPSsn/EKIYQQZ6neW9YWFBTgdDrdtgUG1l/2gVA2H82lQ5Q/mnqq51EuMtDE9qS8er2GEKIZsabBghvdAx6gvl9wI9zyCwREnf75DD7Q4WK4Zy1smQs5h6DNCOh4OQS3qtuxC9HUBMbA5HkqW2PrXND7QL9bIbwD+Ed4fk1BhuqGVL0Aqt0CX0yB23+DwNh6H7oQQghxuuol6HHkyBHuv/9+VqxYgc1mq9jucrnQaDQ4HNLitD6VlDnZcSKfyf1b1vu1IgJMpObbcDhd6LT1G2ARQjQDhVmQc9jzvpzDUJR1ZkEPAKO/Sr+/5EVwlILe8zJMIZqlwBj11W6syv44VZ2bomyVjeWJJVkFRSToIYQQohGpl6DHtGnTAPjoo4+Iioqq92wD4W5fmoUSh5P2kf71fq3IABNlThfpFhuxwb71fj0hxAWuzFb7/tJT7K+NRiMBDyG80Z3mn4SnnKOy5FUIIUTjUi9Bjx07drB582Y6duxYH6cXp7D9RD46rYb4MHO9XyvCX1V+T8opkqCHEOLc+YWDzqAyMqrTGcA3+LwPSQhRhW+o6uxSWlxzn0YrLaCFEEI0OvVSYrt///4kJSXVx6nFadiRlEerUD+M+vqvoB4eoJ6aJuV6+ONHCCHOhNMBWj0MvNfz/n63wu7vIffYeR2WEE2C06k6qhxZBTu/gbRdNetu1AX/aBj2sOd9/e8Ac2TdX1MIIYQ4B/WS6fHBBx9w9913k5ycTLdu3TAYDG77e/ToUR+XFSftTM4n4TxkeQCY9DqCfQ2k5EnQQwhxDpwOSNkCn10HIx6Fsc/C+ndUYdOAaLjoTvV0+Zf/g23zYMZiVYdACKHmT+p2mDfJvQhwm1Fw9Tt1O1dcZRDRGcY9r+aoJeVk+9vbILonOD1kaQkhhBANqF6CHpmZmRw6dIhbbrmlYptGo5FCpueBvczBwYwCBrUNO2/XDPM3StBDCHFuLMkwdyLYrfDLYzBlvgp8GHyhOA+2fQZJ69WxOYdVFxYJegihWFJgzlWqg0pVh5fD6lfhkudVZ5a6UJAOX90MMb1g+COq7bPdAtu+gBX/gPs3qRbTQgghRCNRL0GPW2+9ld69e/PFF19IIdPzLDG9gDKn67xlegCEmU2ckOUtQohzkbJNBTzKOUrgu7tqOX47JAyt92EJ0SSk71KBB40G4vqBTxBkHYC847B1Dgx5oO5aNNvyVWZJ8mb1VV1BBoS3r5trCSGEEHWgXoIex44d44cffqBdu3b1cXpRi/1p6qahZYjfebtmmL+RfWnWUx8ohLgw2K3qyfKeheoGp+OlENUFAs4h8yKvWp0OZxkYzVBS6Pn40NZnfy0hLjR5x6HT5dD3Fji2Bgoz1Pc+wfDr/0GZve6uZTzFQxUpNiyEEKKRqZegx+jRo9m+fbsEPRrA/nQrkQEmfI2683bNcH8TqXkZFcuXhBBNREEGlBSo4qF+EWA8jQ5Mdivs/Ap+eqhy28bZENkVpnwOLidoDaoOh87g/TzVxfZx/37399BzCmz8oOaxPkEQ3f30zltSCIXZ4HKAKQDM4ac/JiGailYDQWeEz69XcxBg6zwIjoeJ7506UFGUozI4NFrVncUnwPuxfhEqy+roHzX3RXZR9T1Ol9MJBWmqDa7OqIqknm7rXCGEEOI01ctvlgkTJvDQQw+xc+dOunfvXqOQ6ZVXXlkflxXAvlTLec3yAAgzG7GVOckrKiXEbDyv1xZCnAV7AZzYCL88Cpn71c1Gj8kw8jEIalH7a62p7gGPchm74Y/XoDgHDi6D/rfDwHs8t6+0W6EwE4qywWBWgYjw9uorK1Eds38RXPOheoKduKTytX5hcOM3EBh36veZewyWPasyUpxlEN0DLvs3xPRQtUKEuFAYzfDz3yoDHuXyjsGGD1QxU0/KSiFzDyyaBSc2qKBH+0vgkhcgzMuDK78Qdb750yBth9rWZhQMvFsFTCzJanlaQAxoa3kAU5gNexeqOiAF6SorZfD90Gc6+EsHGCGEEHVH43K5XHV9Uq3We6vUxljI1GKxEBQURH5+PoGBgQ09nHMy8OVlXJQQypSL6mjt7mk4mGHlqYW7WfSXoXSNDTpv1xVN34U095qUI6vg0wk1t0d0hpu/q32ZyupXVSDBE4MfTHgdvr1Dfd9+PEx8WwUqylnTYflLqs5A+Q1aZGe4fp4KRCx6GBJ/AZdLBUyu+VAVRcw9qm6ogltCQCyU/54pLVKp+8YA9yfE+Sfg40tV0KQqrQ5u/x1ie9X2E7rgydxrguxWlRnh6+H37I4vK+dddRot/GUrhCTU3Je5H94bVnP5izkc7lheex2QggwVBC21QXYi/PyoyhwDVdx00myIHwoGDwVUy+yw9k1Y9lzNfb1ugvEvqowuIYQQog7US6aH0+k89UGizhXay0jLtxEXfH6fYIaaTQCk5dsk6CFEY1eYpdq+epK5V90E1Rb0KMrxvq+0SGWNlEv8RQU5yoMeZSWw/l3Y8on76zL2wtyr4bYlMOl9KMpS6e6mwMqnxVFd3V9TnAuZB2DN/8CaDAnDoe90lc6v1UHShpoBD1AFGH97Bq77RGoPiKbBmq6yMNa/Bw479LgBOlzinpVV27x0OaHUQ7Hx0iIVxPRU76MwC/YtggF3q+KonvhHqq+kDbDwPvd9RTlqqc09ayGio4f3lAar/u35vNvnwbCHJOghhBCiznhPyRBNzqFM9YQlLuT8Bj2CfQ3otBpS823n9bpCiLNQUqg6PXhzeKX3fS4XtB/rfX+rgarAadXU9Iy9lf8uSIMN73l+bX6SakXrEwihbVRtgKAWntPj7QWw6RP46GLY9yMkb4E/X4N3h0LGHnXMvsXex3l8jfcCqUKcjcIsFTBM2aaWlJTWUeHQgnQVUFhwIxxdrQIMi/6qMrXyT1QeV1uNm7C27sHIcjYLHF2l/q3VqeUsoW0q9x/4RQVGamO3wopXPO9zOmDjh+Ao9XDtPO/ndrnc35sQQghxjuqtWtSGDRtYsWIFGRkZNTI/Xn311fq6bLNWHvSIDTq/QQ+tVkOo2UhqvrStFaLR0+rA6F+Zhl5doJcsj1IbHF+nAhexvSFla83zjvk7pGyB0U+pp7Qr/6nS3EttKsW9tLj2YEP2Ifc2tE6HyvjQ+7gHPwrS4XcPS2xKCuHHmXDT97XXJvELr73WgBBnIusgfHMbpG5T3xv8YNjDqpOKOazWl9bK5YLUHXBwac19OYdVodJhfwOdTs2RtqPh0O81jx3+iNpfndagipJ2nQSth6tgqM6oMjN2LACN3nOwpKqSQsg+4H1/+s6TRUqrFTX2NJ6qfGTJlRBCiLpTL0GPl156iSeffJKOHTsSFRXl1tFDunvUnyOZhYSajee1c0u5UD+jZHoI0RSYI6HfrbDmjZr7tDpoNUil0wdEue/LPQqfTVI3dJPeh8MrYPt81fEhYSgMeRBW/wcO/qaO9wmGie+q47+8CfrcrLI3agu4lBdOLLOrpSlb50HqdojpCb1vVPUF9CY4sUndEHqSvFm16+x5A/z5X8/HDLpf/RyEOFf5yfDpFaq2RbnSIvj9ebWsq+8M78tDapN7DDL3wZZPvR+zdZ46f0C0qsHR+2Y1V7bMUUWCY/vAkL+AJRUSl6o6NqGtK5eNmMNgwn9VNsbn11eeV6OFkf8HnSecugOT0ayCJJ6WkoEqHuwpwOEXpsaXsqXmvsBY8I+quV0IIYQ4S/US9Hj99df56KOPmDFjRn2cXnhxKKuQmKBTPD2pJ8F+BtIk6CFE46c3wsB7VZp80rrK7VodXP4qLH9RBR2ufrcy8FFmh7VvqcwLu1V1bWg/Dsb/A0LiIfswfH+P6shSzpantl39juq+krgEBv8FBtylgiPVBbdSN2ROp8oo+eyayrT4w8th3Vsw7RtIGKbaz55KUAu48n/w41/cAyTtLoZukyoLoQpxLjL2uAc8qlrxsqq9ERh7ZufMOgAfXgwdxnsP7oF7pxafYBUQTN0BY58Fk7/qhLT5Uxh0Lyy4SWVcXPyiqn1jOtmSNu8EbPus5nmXvwjtxp16rKYAGPGoCqpUp9WrAKunwIk5HK6ZrZbpWFLc38eUBbXXFRJCCCHOUL0EPbRaLUOGDKmPU4taHM4sOO9LW8qFmo3sT7M2yLWFEGfA5VJLWCbPVTdsh5argp6RXWDTh3DgV3XciQ3qSS+oFPa07VXO4VTHHfgVbvjsZGDBQwHr4lxVN8AnSGWErP2f6pxSnA9bPlZBFICobnD9HHVzmJ+slgpUrwPgKIVvb4c7VkCLi7y/v5ie6sbJ5K+CG/FDVLcaWz60GQlBcWCOOLufnRDVpe7wvq8gXQUazkRxHiz6m5o7R/+A4bNgv5f6ND1uqCwSXB6QHPyAWvqSthMiOqkskG/uqBzH0idVMMUUoIqNesuGAtj4AYS+qIqOZu5TS9VCW5/snlQlozS8E0z6ABY/rOYZqDk2abYqLOxNWDu47TdV9ydth2pZHdNLBSwlK1gIIUQdqpegx0MPPcRbb73Fa6+9Vh+nFx64XC6OZhXRp1VIg1w/1Gwk3SKZHkI0SiWFKpiwYwHkHIK2Y9SSlK3zVB2NkgL1dLmqDe9Dm1EqmFFWCqFt1VKT6hxl6hifIOhxPYS1V6n1O76E3CPqZmzIg6rNrculam5M+xoG3wdFuWD0VXUF/E8GIgoz3TNGqirIUJ1dQlrDsFmwulr3B70JrnhNPUUGtZQmzF8VchSiPoS3977PNwR0pjM7X3EuHFmh/p2fpJaatBwASevdjwuMU1kUlhTI2A07v1HLSHpOVnP7+/tg00fqfFW5XKogang7cJSowIw3lhOw8ytYPMv9PU39Ui1NKW8R7RMAXa9WhYwLs1TAwhxxsvPSKTKqguJUACaqq6r749swf8MIIYS4sNVL0GPWrFlcfvnltG3bli5dumAwuKc2fvvtt/Vx2WYts8BOcamD6MCGWd4S4meksMSB1VZKgM8p1gALIc6fUpuqs/HVjMpsjN3fnUwv/wi+vkUFKaprcZHqRrHyZRUYufSfsNvDZ7ezFLpfD92vUbUB9v6ongQPfVDdcPkGqSyRITPhz9fVE2eNRnWJCK1ynuJcFQTR6uDiF9S5co94uJ5DFTkcdJ+6ufvjv6q4avwQtWwnOOHcf2ZCnK7Y3iqzyJZXc9+gB8A/+szOV33p1tJn4NJ/qKyrXd+q+dP9OvXlcsA3t6oaN+W2zVP1b0Y9DvMmer5GedaHMUDV8Nn1jefj4gerTJYJb6igplanCqXOn6IytkKqZHHoDBDcUn2dLnuBapP9+0sqcBMSr2qJxPaW4IcQQog6VS9BjwceeIDly5czatQowsLCpHjpeXA8W7V+i2qgoEeoWVV4T7fYJOghRGNSkA7f3F5z+UlhFqx8RXWYqJ4xEdNLLRP5cExlTYHd38Glr8Bvz1a2mjT4qpuTbhPhiymV17CmqQKFF90F4R1h/btwwxdwaaxqrZmxRwVazJEqZT7rAPz0Vzj2h3p9ZBcY9xxs/gQOLascl9G/cmmKXyi0HQVxfcFhV0+LT9URQoi6FtQCpv8IX0yurE2h0UDvm6DPTaqzypnwCVYFRxOGqY4qRdlqngXGQbdrVI2Q0DaARhUjrhrwKLd1njo2pLXnwGHrEep/TWYY+hDs/aHmcjKfYOgyUbW0XfkKWJJV0KPj5areT9pO96DHmXI61Nz+8ubKbQXpMHciXPwS9LsFjH5nf34hhBCiinoJesyZM4dvvvmGyy+/vD5OLzw4ejLoERl4hqm0daQ86JFhsdMuMqBBxiCE8CB9l0pj9+TYn6q7Q1UJw1SB0s+udS+iuO0zFcyY9P7JegDZqp1laREsecpzTY+N70OHi9VN4OHlatvPj1bubzUYrnwDPrrEPQ0/Y4/KQJkyX6X1l3d7ufSVml1lpLWlaEgajQrk3b5M3bTbrSpAYY44u/82zeGqPsbvL8Bn11VuD2oJE99TgUKtTnV3qa2zy5Y5KvPp57+5b+81TWVi5R5XGVzZiSrja+Ur6rMCVNbU6KdUnY1FD1e+1ulQAZKsA2oZ2bmwpsJPD3net+wZ6HwFGM8hqCKEEEJUUS9Bj9DQUNq2lTXU59PxbNWu1qQ//+1qQXVvAUi3Sl0PIRoVu5f2sOUCW8Dwv6l6H9E9ofc0VQjRU0eKQ8vU11VvwbLn1E3e9XNVO1tPXC7IPqjW9tvyoKSo2gFOlVpfve4AgLNMdZ4Y/jf1ND2iowq21NbNQoiGoNGoIrxn2qXFk7IS2PY57PnefXt+Enx1M9y5Ui0ZczlVrR5vSgqh0+WQfxx2fq2CMENmqmK+lhPwyeWqSOqG91XdkH63Qcij6r2kbIOs/fDHa57PnbkPSmu59ukoyvG8rA5UkDY/6dwySYQQQogq6iXo8cwzz/D3v/+djz/+GD8/SU88H47nFBEZ0DBZHgAmvQ6zSUe6xd5gYxBCeBDbq/LfoW3UcpTco+qGIzhePVke8Rj0v11lbbw3XLWZ9aTlRWrJSmAsXPOhSoHXniLQqjOp1PnWw2H5S+77Ijqp9rTeJG+C4Q+r5TnZB8HgB32mq+yUurjBFKKxKUhXgQhPCrMgY58KAvqFq7bRW+d5PrbzlXBwGfSYDJ0mqKUi/lEqaPjlTSoI6SxT2VrZB2HpU+p1/lFqfkV1g7HPqOKi+39RWSVVs7lStkK7sWf/Pk/1uaGVZbJCCCHqTr0EPd544w0OHTpEVFQUCQkJNQqZbtmypT4u26wdzykiwr/hgh4AoX7SwUWIRsc/EkY/DVGdIeeIWqJy0Z3qBiYwTrW0BPW/R1ap9Pz8JAjv4N7RpcvV0H6s6uRQnpnhGwKX/kt1Z/nztZrX1hnVcpTyAEV5zYNyxXlqfF7HHgUHflE3ZaCCMuvfUTdck+fW/lohGhuXS9XGSN2h2rRGdYXobiqIUa7MVrmcy5OcQ8BY1TFl8F9U4eDyNrHlwtqpAqXr3oJFD6ngBqhsqWs+qsyWOvALdJ0Iq/6lvg9qoYqWLn0afn1CbdPqVGvcK/8HP9xf+dqgVuf2s/ALg5AEz1lipkAJagohhKhT9RL0uPrqq+vkPM888wzPPvus27aoqCjS0tIA1ab12Wef5f333yc3N5cBAwbw1ltv0bVr1zq5flOSlFvMsPbhDTqGYD8jGZLpIUTjYgyAVgNUjY7S4srtYe3gxmpdG8pvnta/B5e8BN/eodLkDX7Qayp8cYP7097iXPjuDrh1CeyYrwIqVY17HootcO0n8P7wmmNLXALXfKCW1njSdwas/V/N7UnrIO+4BD1E05KxFz69wn1Zh38kTP9JBSSgsjiwpyVfoIr8lgttq+be6n+rAIbOqAIUF90B279QLWurytwP8yapmj1f36KyrAbdr2qSpO2E0U+qltL5SZWvcTpUPR/fYOh4GexbpFpDtxpwbj+LgGhVu+TTKyq7yYBaajPp/TPveiOEEELUol6CHn//+9/r7Fxdu3blt99+q/heV6US+j//+U9effVVPvnkEzp06MALL7zAuHHj2L9/PwEBzaeYpq3UQabV3qDLW0DV9UjNLz71gUKIc2NNVzdOLgf4hqobCG/p4tYUFaworTY3sw/Cr4+r4oimk5+XEZ3U/+YeVTdS18+BwyvUNXZ85blYqcsFa99U3Vn2/gip29UT3L7T1dIZczTY86FFf9XusqrSIrCkwcUvqvT6qufvdxuUFatii54krYcW/U7xgxKikbCmwfypNetYFGSoDibTf1QBEP8YVcfm18drniMkQQUryxVlweK/qeDg0L+qDJFDy9Tc8LZEpiBd1ePwj1TX/v5euOxfUGZXwc2qAY+qtsyBCa+rOTx1gSqGeq5ie8O9a2H7fDixUX3+9J0Bwa1AL8tbhBBC1J16CXqcDpfLdVqtbPV6PdHRNSP+LpeL1157jSeeeIJJkyYB8OmnnxIVFcXnn3/OXXfdVedjbqyS89TNTEMvbwnxM3Ik6xyLmwkhvHOUqY4K39ymio2CShO//FVoN6YyeFFV5n61ZMWT/T+rOgGmAFVA0RyhOj6sexuSt6jskIQR6gnw7u+8jysrEQ78rG62rvkAzFHurTr9QuGqN2HxI7B/kQqUaHVqyYw5TD09nrpAjdXpULUKjP7wZi1BDXkSLJqSwkzP7WNBFQYtzFKBCJ0Oelyvsq6OrIawturfRblw1f8gMEYtE8s5ogKMfW46WXx0Kyy8V53vuk+9Z4qACjQGxKigh90C390FbceoeeeN3QrBCXDfBhWYqYughE6v6gyNeAwcNlX/51S1PoQQQoizUGdBj86dO/PUU09x7bXXYjQavR6XmJjIq6++Snx8PI899tgpz5uYmEhsbCwmk4kBAwbw0ksv0aZNG44cOUJaWhoXX3xxxbEmk4kRI0awZs2aWoMedrsdu71yGYbFYjnNd9k4JeeqoEd4gwc9DGRa7acd0BLNz4U29867/CTVdaG0SheUomz4ajrctlQVGq2uMMv7+fQ+KshQXqhQo1MZGp2vhG9vh/wTKlX+xHr1hDlth+fzhLVTN2J5x1TnheLsmktPAuNURkfvaSrA4hOkCqGueAkOLlWtMAfep64XGKPeY7drYfvnNa+nM0qWxxmSudfAqmdaVVdWZb85Qi0n8w2Bg7+puTPsYRXgzDoIc692z8gIaglTvoDglpCXpJaf1LZEJrqbyrLIOqA6wRRb1Hw11lJ43uivjgluedpv+bRptaCVovdCCCHqT50FPd566y0effRR7rvvPi6++GL69etHbGwsPj4+5ObmsmfPHv744w/27NnD/fffz7333nvKcw4YMIA5c+bQoUMH0tPTeeGFFxg8eDC7d++uqOsRFRXl9pqoqCiOHTtW63lffvnlGrVCmrLkvGK0Ggj19x5sOh+C/YzYy5xYbGUE+UpqqqjpQpt755XLpdq7Vg14VPX786p9rG+w+/bobt7PedVbqjjh8bWV2/b9CO3GwYzF6uZp+3xY+Qpc9zHs+a5my1iNRt2gfXM7jH5KpcsXZsCwWerpsX9E5bHOUvhiSuX3viHQ+0b1eqdDLYcJjFH7DH4w+gl1E3divQqmAOgMailNgGR6nAmZew3MHK6yGJyOmvt0RtWNpVzWAfhovPtSmC2fwmX/VrVsqi9ByU9S2RrXfgIfXwpavVoetvrfNa/lH6mWjxRmw84vVZZWaBsY9biqKxLUQgU7qxtwt8w5IYQQTZbG5ar+F+y5WbNmDQsWLGDVqlUcPXqU4uJiwsPD6d27N5dccgk33ngjwcHBZ3XuwsJC2rZtyyOPPMLAgQMZMmQIKSkpxMTEVBxzxx13kJSUxC+//OL1PJ6eeLVs2ZL8/HwCAwPPamwN6V+/7mPBxiT+N6VPg45jX5qFZ3/cw9KHhtM+qvnUVBGn70Kbe+dVmR0W3KiKf3riFwZ3/1Gz60FhFnx7p1rrX1VUN+h/G/z0kOfz3fA5xPWHdwepc3SdBG1Gwm9/r3yC7BMMY/8OxzeAOVSlzC95svIc/W+HMU+rrA6Agkz1lDp9l+dr3r8Jwturf1vTVT2S3GPqZstRquoRtOinrqNv2My2pkbmXgOzF8Cy52HDuzX3DX1Y1fEw+qqORl/NgMPLax6n0cLUL9WyM09u+k7NE58gtWQmeQtsm1cZaAlvr4qYZh+BbXPV8piqJrwB8YNUjZGMvZXX7H2T6vhky4eQeFXPQ6s925+EEEIIcd7VeU2PwYMHM3jw4Lo+LQBms5nu3buTmJhY0SEmLS3NLeiRkZFRI/ujOpPJhMl04fzBnJJnI6yBl7aAqukBkG6xS9BDeHShzb3zSmtQhf68BT1C4tVylerM4XD127D2bdj0oSp2GBwPV/wXfn7E+/XWvwdX96pcHrP7W7CmwuX/UR0mjP7q6bTLqZ4c71gAa99yP8fGD9QT4vKgh3+EKob4yWUqiFPV0Icrl8TkHoXPJ6sbt3IRnVTdj5AE72MWXsnca2Amfxg+S2VS/PkqFOWoZSzDZkH3a1TAA1RA0VPAA9Rcy9ij6nx4KvBrt6qAxU3fw5KnoevVKnhZagO9UWVN/ThTBVgKM2u+/ue/wf1b4OYf1Ly3W1W7293fwnvD1PV9Q2DaVxDbR+pvCCGEaDIarJDp2bDb7ezdu5dhw4bRunVroqOjWbp0Kb179wagpKSElStX8sorrzTwSM+vE7lFhJsbdmkLqO4tAJkFtlMcKYQ4Y1qtKlq47i3PKfIjHlMFQz0JiFbFSAfcoZ4EG/xUCnz1wENVZTb1lDemZ+UT4eNr3ZfCTJ4HEZ1h3jXez5O+qzJ7AyCmB9y3UXWP0ehUIVMA/cmbvsIs+OoW94AHqO+/vgWmfl35GiGaEv8IGHQvdJsEDrvKVvKPcc+acHmY21WV2VQAtDqNVmV7XfkmRHWFAXfBqn+qZTFV+YaobkyWFA/ntqvAZsv+6riVr8Cqf7kfU5wLc65WXVeCW53W266VwwEFqarmid4H/KNUgAZUsKYwQ43L6FfzZyWEEEKcpkYd9Jg1axYTJkygVatWZGRk8MILL2CxWJg+fToajYYHH3yQl156ifbt29O+fXteeukl/Pz8mDp1akMP/bxKybPRNz6koYeBSa/Dz6gjw1LLjZQQ4uwFtVT1LL65XXVdAFXjYuTjqiVsbfRG9fpyTgd0vko9Ofak2zWqvsbFL8KnV3gYSwv1tPdUBRoN1QoU2gtU4OTQcuh4Kfz6mEql12hUDZCxz0DKFs/nSt6inlBL0EM0VVodBMV53+8TpJaeeVsCFj+4ZiACoMcNKoMrYaj6vv/tamnYzgWVxwTGwuTPYOF93q+vOxlQKUhX2V6elBTAiU3nHvQozITtC1TtkeJcMJrVMpoB96isktX/ga1zVaDHPxJGPQWdr/Ae3BVCCCG8aNRBjxMnTjBlyhSysrKIiIhg4MCBrFu3jvj4eAAeeeQRiouLuffee8nNzWXAgAEsWbKEgIDms7TC4XSRbrER1sBFTMuF+BnJsErQQ4h6YfCFNqPhnjWq2KCjRC1rMUeoG4YzYcuHhCGwNQ4sye77Qtuo5ST2AojtBVMWqNT3vOMqONFuHFz6T3XzVpQDcf0geVPNa+h91HnKOZ2qTe6SJ2DiuzB/amXWisulOlX0nVH7uEsKzux9CtGUmCPU0rOPL1VLS6rqNU1lR/W/HbbMgZJCtcys940qI6s4t7K7SkAUXPZPGPEI5B9X9XcColXQxe6lc49/ZGXhYUeJ9+NAtcw9F2U2WP++ykYpV1IIf/wXYnrDpo/gyIrKfQUZ8OMDalx9b3FviS2EEEKcQqMOesyfP7/W/RqNhmeeeYZnnnnm/AyoEcoqsFPmdBFubhxrtYNPtq0VQtQTvUHd2Jxr60iXQ91wXPkGHPgF9i1WKfJdroTWw2Ht/2DSB6obTMfxENsTbJaTnSbCwOdk8Uu/UNUF5uPx7i0yNVq4ZrZ7x4eCVFj2DPS8ATa8772ThUZTs0sMqO2+DZ/VJkS9iu4Bd62CFf9QWVH+kTD0IRXYeHsgdLgUrvyfWqLmdKiuThveg9uqFSv2DVZf4e3ct1/7CcyZ4J6lpTfBdZ+qJSSglpsFegiIlovrfW7v0ZoOa96ouV1nBJ3ePeBR1fIX1OdRUItzu74QQohmpVEHPcSpJeepP1oaS6ZHkK+BdIvU9BDivCrOg6Islb1hClTFS08VHPAJgZYD4LProP3FMPgBwAWJv8G6t+HilyoLkILqmBIQ4/lckZ3UTdrB39SylbB20HOKujGp2mXFXqCe2EZ0hm2fez7X4RVq2c2e72vu63adehIuxIXAUaJu/gszVQDDHK7mmMFH1eW4+h1VTFRnUPtyj6l6Hvt+Ul9VGXzVMacjtjfcsxb2/KAytGJ7QdeJavlbec2MgGjVeem7u2q+PjjePYPrbNjyVLZHdf6RkFtLFklxrvqZCCGEEGeg3oIeTqeTgwcPkpGRgdPpdNs3fPjw+rpss5OWr/5oaAzdWwCC/YzsTa0lJVYIUbcsKbBoFuxfVLmt7ViVwVFb7QCdXrWi3PyxyvQ4UKXNd2AsdJ6gMitOV3Ar6Hcr9Jnhvdig3kfd3NnyVfDCll/zmI2zYfpPKmizbZ4qvKozQK+bYOSjlRkmQjRlxXmw9wf49fHKm/jAOLj2Y1UrR29QHV9M/pWv8Y+C0U+p5WHVjXpK7T8dOj2EtoahM9WSM0/zVaNRwdDL/g2/v6CCFKCywCa8UbM19pkqL1xcnb1AZZJ5o9F67lIlhBBC1KJegh7r1q1j6tSpHDt2DFe1FGWNRoPDcYrq5OK0peQVY9JrMRsbx/rWED8DWbK8RYjzozi/ZsAD4NBv8P29cN3HtRf9C4mH236D1f+F3d+oG4qek2HQA2e/fKa27gp+4dB1Euz6GvrcDEufrnmMo1Rlq4x/CYY8CKUFqm6Bf2TNoqhCNFVpO+CHB9y3WZJhzpUqCyOsTc3XGHyg11QVsPj9BdUBKayd6szUapDaf6Zqna+hqn5Gx/FqaZvepOawb/CZX6c6czi0uAhObHDfbssDY4AKelZdLleu46Wnn9EihBBCnFQvQY+7776bfv36sWjRImJiYtCcydNCcUbS8lUR08byMw7yNWC1l2ErdeBjaByBGCEaXGE2FGWrdG7fEJU6rvPQdvJMFWXWDHiUO7JCtX89VaeD0NZw2Ssw8jHQoJ6y6uspc8xkhrF/VwVMfYKgy1WwZ2Hlfo1WPUUObqkCHKEJ9TMOIRpSUa4KWnhSZlNBwRGPeN7vFwqdLlcBA0eJqoHhf5pLvgoy1OeQs0y1rQ04jRawOr1a9hJU+2FnzC8UJr0HcydC7tHK7eZwiOgIN36jWuNWLaYa2QXG/wNMzadYvRBCiLpRL0GPxMREvv76a9q1a3fqg8U5SbXYCPVrHPU8QC1vAci02mkZKk9lhSDrIHx3h2q3CiqLYfST0P26c2+9aDvFUjKbhyelnhh8IchLunldC2oB076C3OMQ0gYGz4Tkzern0moAmCNVcESIC1VZMWQd8L4/eTOUlag2096cbqADVLHT9F2q1XX5dc0RcPl/oO3ohgsihLaBW36BnEOQvhvC2qpaIUEt1JjvWaO25x9XxV1DWquuNEIIIcQZqpegx4ABAzh48KAEPc6DlLxiQsyNKOjhq55eZ1htEvQQIv8EfHo5WNMqt9kt8PMjKqOi+7Xndv5T3az4BJ/b+euLf5R7/YEWfRtuLEKcb3oflWGV7CUoGdOz9oDHmco7Dh9f5t7uuTATvrwZbl2igo0NJTBGfSUMdd+u1dVNlyohhBCCOgx67Nixo+LfDzzwAA8//DBpaWl0794dg8E9jbtHjx51ddlmLzXfxsDW5/i0uA4F+50MelikrocQpO5wD3hUtew59Yd+1ZauZ8ocAe3HQeLSmvvih6j190KIxsUvFEY9CfMm1dynM0KPyXV7vT0/uAc8qvr9OZj8Wd3U6RBCCCEaqToLevTq1QuNRuNWuPTWW2+t+Hf5PilkWnecThdZVnujyvTwN+nRazVkSDFTIeDERu/78o5B2TnOE99guOJ1WHgfHF5euT1+CEx8D8y1dEGozlEG1lT1VLikQBVINEdItxQh6kNsHxj/Cix7BkpV63nM4XDtJ6qGRm0KMtRczU9WHZoCYlShX0/KSuD4Gu/nSt8NpUUS9BBCCHFBq7Ogx5EjtfRVF/Uiq9BOmdNFaCMKemg0GoL9DGQVSNBDCCI6et/nF1ZZzLQoG3KPwY4FKhDS/ToIb+/9RqaqoDi49iNVtNSWpwqEmiPOrF5IWQkcXwsLbqwsHKjRQP87VEFF8xnUDxBCnJpfCPSdAZ0uA2u6+iwwR5y6uGjuMZg/RQUrykV2gSnzVTem6nQGVSejakvqqoJbga5xtLwXQggh6kudBT3i4yt/2a5atYrBgwej17ufvqysjDVr1rgdK85eer4KLDSmQqaglrjI8hYhUG0kjWYoKay5b8iDqq5FYZbq5LD548p9mz+GNqNg4runt/zFL/TciqJaTsBn16puEOVcLtjwPkR1h743n/25hRCeGXxU0CG41ekdX5gFX9/qHvAAyNgDX98CU7+s2c5Vo4GeN8Da/6nioNUN/9uZZYQJIYQQTdApepWdnVGjRpGTk1Nje35+PqNGjaqPSzZLaRYbQKPK9AAI8jWSabU19DCEaHiBcXDzQpXVUU6jgT43qxsRrQ4y97kHPModXu65Vkd92LfIPeBR1ep/qyfRQoiGVZQFyZs870verIIinricqhV01cLHWh0Mul9lhjmddT9WIYQQohGpl+4t5bU7qsvOzsZsllaEdSXNYkOn1RDoazj1wedRsK+BdIsEPYRAp4fYvnDXarAkgy1ftWk0h6ubjTI7rHvH++vXvQ0dL6359BagOE/dzPgE154Ofzoy93vfl58EzrJzO78QF4rSIigpVhlcBp/ze227l2Kk5TwVK3U6YO2bkHsUrnpLBTfLbCrLbPd3sPB+uG2ptIIVQghxQavToMekSaoSuUajYcaMGZhMletEHQ4HO3bsYPDgwXV5yWYtPd9GiJ8BrYcAU0MK9jOwKyW/oYchROOg1aq6G0Fx7tsLMtSyl943qZuQQ8vUkpKqSgpqBhysaXBsDax/t7L+R9erIajF2Y8xfghsnet5X1RX0Muaf9HM2Qsg+yD8+QZkH1DLvgbfr4KYBt+6uYajVH0uOMvA6Fezlo5viMoUq/45AWq7b0jN7S6nCpAeW6O+DL6g1YPdqvb7hapjhBBCiAtYnQY9goKCAJXpERAQgK9v5R8CRqORgQMHcscdd9TlJZu1dIuNkEZWzwPU8pbsghKcThdabeMKyAjRoIpz1U1NYZYKdCQugV3fQo/rodc0+O4u92UmHS5zv5GxpsN3d7t3akndpgIgt/wMwafo+uBNwhB181NUc1kiY5/znGkiRHNRVqIKgX5zW+W2tJ2wY76qo9F2jPdsK5dLdVopylZZF+Zw8I8Bnc79OGsqrHsPNn2gAhJRXeGSlyCuH5j81THmCOgyEXZ/W/M6XSd5LjisM0C3a2H/YvV9eaeYcu3Hg2/Q6f0chBBCiCaqToMeH3+s1qUnJCQwa9YsWcpSz9IaadAjxM9AmdNFblEJYf7yhFgIAPJPwMIH4PDv6nuNFjpPgMv/rW6mWo9QRQWXv6j2+wTDwLvdsyzSd7oHPCrOnQSbPoJRj1d2hKmqpEgFW9J2qGNje0No28qU9uBWMONnFXRJ3aa2+YXBJS9DXN+6+gkI0TQVpMGPM2tudzlh4b1wx4qamVygApgnNqn5bUlR23yC4fJXocPFlTU2CjJUgdJjVVrLpu+GOVfBtK/VZ4PeqNpHj39ZLa3Z/oXKCNHqVX2g0U95by/daqBqQZ190H270QzDHgaD35n+RIQQQogmpV5qevz973+vj9OKatLybbQOb3yBpWA/ddOVYbVL0EMIgMJs+Po2SFpXuc3lhD0LQaODPtNh4wdw0Z3qRqjDpTDyUQiu0unKUQqbP/V+jR3z1esDYyq3FWRCzmFVS+TrW9zX/Ed2UU+py7NDIjvBjd+oJ9KOUpVhEhCtCh4K0ZxZ0zzXywAVsCjK8hz0yDuuAhdVs7dsefDNraqORsuL1Lb8JPeAR1W/PAbjnlNzMbQd2CwquHH9pypzRKuDo2tUq2lvnZ6CThZUXveOWsZWWgwdL4NRT0Bo69P+MQghhBBNVb0EPXr37u2xkKlGo8HHx4d27doxY8YM6eRyjtItNvrEe1jD28CCfFX2SabVTueYUxwsRHNQmOEe8Khqz/dww2cq6FGYCfdtAp8A9RT2nK6ZCav/o5bOuMrg6rfVzc6mDyFpg2pz+esTantF+ny456UsDgcUpKraADqjygKRNpdCeOd0wJa53rsirfiH6qii08ORP7yfJ/ugyuiYPRpmLIZFD6nCw2vfcj/u0G8w/UdVoNSToBYw5u8w6D615MYnqHLeCyGEEBe4emlZO378eA4fPozZbGbUqFGMHDkSf39/Dh06RP/+/UlNTWXs2LEsXLiwPi7fLNhKHVhsZYQ2wuUtVTM9hBCoAIQ3LqfKrAAVcAiM9hzw0Bmg73TP59BoYPij7vU/8lOg3Vj44QH4Ygp8eTMseRJ63KBa5gLs+7H2sYEKdOycD+8OhXeHwFv94bNrIfNA7a8T4kIREA1GLwEC/0jwOxkodLkqCw+X2SqXinmSuVdleCy4qfbuS1WXq9nzvXdaytzvvWVtOb0RAmNV5ocEPIQQQjQj9RL0yMrK4uGHH2b16tX85z//4dVXX2XVqlXMmjWLwsJClixZwpNPPsnzzz9fH5dvFjJPBhTKAwyNiUGnxd+krxijEM2eXy1ZERqNurEx+KklJ7WJ6g5tqmXItRoIN30PRZlqCcuKf0D2IXXeb26F9F2VxxZmwqK/qnMExqqAS0mhKtRoSVVfjmrdYpLWw/f3qiKs5VK2wCeXQl7Sab19Ic4bm0XVzzhVAOBM+EfDhNdrbtdo4aq3VZAybScs/psKYmz9THV7qW0+h7ZVxUVPbIDwDp5r8QB0mqAKHkNlcNSb6kVKhRBCCAHU0/KWL7/8ks2bN9fYfsMNN9C3b19mz57NlClTePXVV+vj8s1CusUG0CgLmYIKxmRYbQ09DCEaB3MkxPTy/OS3w3g4thamfKG6OtQmIAomvqvW/294DwJiVD2QzyerJ8ugbqRW/wcmz4OQBEjdXvM8699R3WI2fwx6H1j2LOz8St3E9ZoG/WZAUEtVE2Tp057HUpilAiLmCChIVzdcRrN6Ku7tBq4JS7fYsBSXYtBpCTUbCfS98N5jk2YvhKwD8PvzKigXGKsKAycMO/fuQ3qjmqd3rqzZsjYgBnYsgJ8fqTx+/2JVj2fKF2qeOh01z3nRnbB4lvr3hvdUcdOfHnQ/NrwD9JoKC248OQ6TKlxavY01qNoetQVXhRBCiGasXoIePj4+rFmzhnbt2rltX7NmDT4+PgA4nU5MJilyebbSLSqLIsTcSIMevgbJ9BCinH8EXD8HvpqhbsjKtR0NI59QhQkDYkDv4Ua6OFdlaOQeBd9QdTPXdaJ6rT0fPrq0MuBRzlECC++Di19QHVmqS9sFfWeobi+/PaOWuZRb/W/Y9TXMWKS+z9zn/X0d+1O1uV36JJTZwRQIQ2aqQIy/h/aZTVCBrZT1R3J4euFukvPUk/Rh7cN5/qpuJDTCQtLNVtJ6+GySWmICat58NQMG3gsjHjv3tqwmf4jtBVe/CSUnA3wGH8g66B7wKJd3DPb/AjfMh+/urMyUMviq2hopW1WwEODgMlXQeMp8SNmmjm05AEqL4Ns7Kuf3vsW4+t2KZsP7Na/X/w4VXBVCCCFEDfUS9HjggQe4++672bx5M/3790ej0bBhwwY++OADHn/8cQB+/fVXevfuXR+XbxbSLTaMOi1mY+PsrBDka6jIRhFCACHxMO0rFcCw5YFvmHoC7Rfq/TWWVJXVcXwN7PxavS4gBqb/BL7BUGqDyM5gTam82StXmOl93X5wSwhOUDdXVQMe5XKPqhu2zleoIEt5u81qXEEt0Wz7TAU8QHWQ+P15XKXFaEb8TWWRNHG7Uyzc9ukmt22rE7OY/P5avrt3CLHBvg00MlHBmqayJKrPAYB1b0O/28496FHO4Ofe4tVTC+lyK16EmTvg7j8gPxmsqapwqb0Itn3ufmziEvUV3R36zIADS1QB4iEzVfelXd9SggF9z6lozBFqrvmFq84xpgCI60eJo4wTmQX8vDOVpNxiRnSMoFeLYGLkv1EhhBDNXL0EPZ588klat27Nm2++ydy5cwHo2LEjs2fPZurUqQDcfffd3HPPPfVx+WYhw2onxGzw2CWnMQj2M7I7Jb+hhyFE4+KtO0p1NosqdLjiFcg5BOHt4ao3IXEptB4OexfCrm/UE+B242DA3bDoYfV0uSpHmartUf1mcMiD6hrLX/A+hp0LoMdkGDZL1QGpTm9CE91NLY2pRrP2TVUsNSTefYejDArSVJBE73POLXFLypxkWm2UOlz4GnVEBdZtkCWnsIQXF+/1uC/dYmfL8VwJejQGxXk1/9uvKm0HhLfzvv9clBR63+d0qDocoa3BBfzyqMrwCG0Dgx+Anx6q+ZqMPRDbW7WzXfuWalMdmoBz0mz0viFoi7JVHZ81b0LWj2oJTP/bYff3lPi1ZOz/EnGenO7zNyYRF+zLF3cOpFWoX81rCSGEEM1EvQQ9AKZNm8a0adO87vf1lT8Uz0WGxUZwI63nAaqmhyxvEeIslJXAvkXw/d2V23KPqoDHxPfg6B+qFke57EOw+zu4+h34/LrKmgB6HxUsMZihpEBt02hVLYGIjuopsb6Wz2G9L2iAzldC+m7Y/FFl8MQnCNe1H6P5w0tdpjIbjuI8dFWDHgUZ6gZu7f/Uk2u/MBj+CHS/9qxqLqTl23hnxUEWbErCVuokLtiX/7usE8PaRxBUR/U27KUOdiZ7D96uTsziih6xdXItcQ5OFTgz1OPfG21Hwm9e9sX1Va1hQS1VGf0U/Pq4WjJWnKeClRveVwWFAYxmXJM+QFNWrAoFl9lVgGTQA2hXvaKChp0ud19Ok3tUZYhc9Ta+2Plickte/iOPbSesACTnFfPior385/oe+JukDo0QQojmqd6CHgAlJSVkZGTgdDrdtrdq1ao+L9sspFlsBDfiQnrBfkYKSxwUlZThZ6zX/8yEuLAUpFUWOKzu50fhilfdgx6gagPs+wk6XgZ7Ty5XGft3tYTl3nXqKXhpEYS1A78I8AlQx1x0p3twpaqBd6u0eVOAOtege1X9AoNvZXr/0T+8vo0ynS8Vt6I2Cyx/ETZ/UnlAUbZ68l2UDcMeVvURTlNWgZ2Z87ey/khOxbbkvGLu/3wrr0/uxZW9YuskC06r1RAZYKqooVRdQpjU9GgU/EIhrg8kb6m5T2c8dVekcxHYEjpPqJx3Fdc1qNodh5ZDi75qKdl3d6r6IgHRavlZwhDoeQOk7aTMP5a9pVG4NL70WHJ75ZKxS15SdXkKM+G6T1ULak9+eQzdlf9jwJK/MvuSd/i/zcH8diAPgKV70sgp7CxBDyGEEM1WvdyNJiYmcuutt7JmzRq37S6XC41Gg8PhoZK5OCMZFjvtIr2s128EygMymVY78WES9BCihoIMtTxFqwdzFOhOhgisaZWZGdXZ8tRNnCf7F6sbqvxkGPWEutEymdVTcI0WcIHRvzLgAdB2lOpucXS1+7k6XAqxfSu/9wlSafrH18CG2VBahObKN1WHl/yabWsdCSOwaoOoKFVdmAVbPvU87jWvQ+8bay6FqUVqXrFbwKOql37ey4A2oUQHnfvT/cgAE3ePaMuzP+6psU+n1TC+W/Q5X0PUAb8w1Tr2o/FqjpTTaODqd8E/qv6ubQ6Dy/8DbcfAmjdUYd9WA1Rh0T/+q2p+6Ixw+zI1DxbPUgERo7/a3nMKtOhPWWArvliVx81dnWoJDKhATtoOFfAANZftFs/jsFvUPC/KJuKHaTw9+XcOZpdwd/8gOocbiHBmgSPmguysJIQQQpxKvdyNzpgxA71ez08//URMTEyjrTvRlKVbbfRPCGnoYXhV3ko3w2onXp6GClHJlg9JG2DJkyrN3TcEBt2vbvwDzuEmWquHNiOhy1WVxVHzjsOq/8COL9ST44ShcMnLENlJ3XAFRMM1H0D6LrX0RKOFfreq4qj+1TpBpO2EP1+v/H71v2HCa6pLjDWtcntkZ44NfQVfY2DltoI0z0UmQY2rOOeMgh67kr3c+KFqbRTY6yawrtFouKJHDNuS8li4rbKYq0mv5a1pfYgJavqFWi8YEZ3grlWwfxEcXgmhbaHPTapo7xlkEZ0V/yjodwu0vxjStsOh3+HrWysDFI4S2PuDWpqyb5EKIIZ3UBlOa9+EP1/DR2fk6c7Xogl/SM1LaxpEdlWfFaer/G8tRylRh7/m+2uuInjJTBU4MQWqzK7+t0PgKVpjCyGEEBeYegl6bNu2jc2bN9OpU6c6Pe/LL7/M448/zsyZM3nttdcAFWD59FP3J4gDBgxg3bp1dXrtxsRW6sBqK2v0NT1AZaQIIU5yuVS6+1fTK7cV58Lvz6tWlVe+oZaT+IZUtrisyj/SexZI75shqFVlxkh+Msy5CnIOVx5z9A/4YAzcuRKiTqb8B0Srr9YjVQ0PrYdfC/YC94AHqFoCi/8G454DnYlCax55/m3YXRhEUUEgE1pXaUluPEVWWm21RTyICPTe7lyn1WDUa8/ofLVeK8CHZ6/syn2j2rErOZ8AHz2dogOJCjRh1DfO7lnNkkajAmcD74V+t6uMhvP9wMVZCvOnVdboqGrNG3DHSjUX9v8MIx5Rx5a3o3WU4LPrczjxB66r3kYzb5IKmlStd1NarLJairJrnt8vTC1hO8mUtQuTPUcFPECda/W/VWvfaz++YFpKCyGEEKej7v4yrKJLly5kZWXV6Tk3btzI+++/T48ePWrsGz9+PKmpqRVfixcvrtNrNzblgYTywEJj5GfUYdBpyLBK21ohKlhT4ZfHPO/b96N6uqv3gUtePLkkpQqtHq54DVJ31nxtWFvoPa0y4AFwYqN7wKOco0R1bbFb3bfr9J4DHuWv8RSEyTkM394J697h57K+3LtCiyEohlEdI9HpqozfHAHBXmo5Rfc440KmnaID8PPSrvuybtGE+9dtQDjYz0iHqAAm9WnBuC7RtAz1k4BHY6Y3nv+AB6gixJ4CHqAympI3wuWvwj1rcG2dVxnwqCrvuMqMatFfFS/udEXlvvXvqM+G6oVbtTq1ff27ldvCO0H+iZrnP7ra45I0IYQQ4kJWL0GPV155hUceeYQVK1aQnZ2NxWJx+zpTBQUFTJs2jdmzZxMSUnNJh8lkIjo6uuIrNDS0Lt5Go1UeSAhpxJkeGo2GED8jGdLBRYhKdosKfHiTul1leeQchakLoNc0aDVItX+dMl8tQxl0H0z7CtqNUctVrnoLbv4RglpUnsfpVB1dvDm8wnttAE9MgSp13wtnq4GM6tGGD2f0Z1SnyJrdUwJj1Pj9qn02B8aqp85nGPSIDvTh41v642Nw/xXWMdqf/7ussxRPFg3DFOA+D6uL6QUmf9Dq0BxZ6fUwzd4f4fq5MPQh9Vkw+kkVxEneAnsWwpQF0GeG+mzofZP6fs/CykKuehO0Gw3ernHsz7N9h0IIIUSTVC9/GY4dOxaAMWPGuG0/20Km9913H5dffjljx47lhRdeqLF/xYoVREZGEhwczIgRI3jxxReJjIz0cCbFbrdjt1fejJ9NIKYhlQcSGnPQA1QmiixvEVU19bl3znQnn0B7q2/hGwK+wdB3Bix+WB0f01M9sd0yBy59RQUKguIgfoh6qmwKqHkerRb8a6kP4hMMmjPIVNDpVSbJhvfdC0UC+ASh7XsLYYGnqN0T2QXuXKXa32bth6huqg5DUNzpj+MkvU5Ln5YhLH1oBNuS8kjNL6ZXy2ASwsxEBkqdDU+a/dw7HwJjVLeVL2+uua/taDV3Qc09n2BV9NQTc4SqEzLsYSgpVG2ou0xUWRqlRRAQC2OfUctp0vfAN7dB0cns2sBYmDgbVv3Le9bJWbSIFkIIIZqyegl6LF++vM7ONX/+fLZs2cLGjRs97r/00ku57rrriI+P58iRIzz11FOMHj2azZs3YzJ5Xvf98ssv8+yzz9bZGM+3dIsNg06D2dS406uDfY2kW2R5i6jU1OfeOfMLh3ZjVdp6dQbfytaaQXFw9Tuq24PdCj6BqtWsb1Dl8cZTBBn63Agb3vW8b+A9YPYeGPYoOB5uXwpLnobEX9S2duPg4he8L12pSqNRRSWDW0LH8Wd2bQ8Mei0tQ/1oGep3zudqDpr93Dtf2oxUWVq/PgHZB1VQsv8dMOCuymCDf6SqPeKtNXW/21TgEtTcB/ALgfB2NY/1DYE7V6igh0anruETDAFeOtZo9SpDRAghhGhGNC6Xt0eODS8pKYl+/fqxZMkSevbsCcDIkSPp1atXRSHT6lJTU4mPj2f+/PlMmjTJ4zGenni1bNmS/Px8AgMDPb6mMXnll318s/kEr9/Qu6GHUquP/zzCkaxClv51REMPRTQSTX3u1Ynco/DpBLV2v5zOAFO/hPihqh5BXSjOV21ilz7lvj1hOFwz++w7xdgsKtvDhcpK8Wkm/781cTL3zrOCdFV4VGtQQY7qrWKtafD9vXBomfv2UU+oLiu+wed2/bwkmHu1CryU0+rguk9VsLK+O9oIIYQQjUi9LXxevXo17733HocPH+arr74iLi6OuXPn0rp1a4YOHXpa59i8eTMZGRn07du3YpvD4WDVqlW8+eab2O12dDr3bIeYmBji4+NJTEz0el6TyeQ1C6QpyLDYG3UR03IhfkY2WL2k74pmqanPvToRkgC3/gqpO1Q3ldDWJ1Pf4+ou4AEqK6TvDOhwCez9SdXw6HgphLap2Y72TPgESqCjCZK5d575e8m0KBcQDRPfU0HQfT+pjJBOV0BAjHtG19kKbgnTf4KMPapjVFALaD9OLX+RgIcQQohmpl6CHt988w033XQT06ZNY8uWLRVPl6xWKy+99NJpd1cZM2YMO3e6dyq45ZZb6NSpE48++miNgAdAdnY2SUlJxMRcuH3oM6w2gn0bdz0PgBCzgbziUuxlDkzS6UCISoGx6qsOlnnUqjxAEdERUHWVkvOK2bDlBFuO5dIxOoCRHSOJCfJBr6uXutanJa+ohJS8Yn7ckYqt1MFl3WNoHWYmPEBu0sUFzD9CfbXsf9anyLLaOZJdyOKdqfgadFzRI5bYYB/V0j4wRn21G3PqEzWAis+jIzkVn0cjOkYS28CfR0IIIS489RL0eOGFF3j33Xe5+eabmT9/fsX2wYMH89xzz532eQICAujWrZvbNrPZTFhYGN26daOgoIBnnnmGa665hpiYGI4ePcrjjz9OeHg4EydOrLP309ik5dtoHX6K9fyNQHlgJtNqp0WIrLsXoqEdSC9g8vtrySsqrdjmY9jLZ7cNoHerELTa89/mM7ewhLeWH+SDP45UbPv4z6OM7BjBP6/pIYVJhfAiw2Lj4S+3s/pgVsW2t1cc4u7hbbhrRFtCzI374Uhj/DwSQghxYaqXUPr+/fsZPnx4je2BgYHk5eXV2XV0Oh07d+7kqquuokOHDkyfPp0OHTqwdu1aAgI8dDS4QGRa7Y2+cwtQ8QeXtK0VouFlFdh54IstbjcYALZSJ3fO3UxaAxUdPpxV4BbwKLdifya/789ogBEJ0TQs2ZPuFvAo9+6qwxzNLmyAEZ2+LGvj/DwSQghxYaqXTI+YmBgOHjxIQkKC2/Y//viDNm3anNO5V6xYUfFvX19ffv3113M6X1NjL3OQV1zaRGp6qDFmyB8vQjS4nMISDqQXeNyXXVhChtVObLDveR1TSZmTT9cc9br/w9VHGNs5inB/WeYiRFWZVjsf/1kzWFhuztpjdG8RhF7bOJeJ5BQ1vs8jIYQQF656CXrcddddzJw5k48++giNRkNKSgpr165l1qxZPP300/VxyWYj82TWRFPI9PA36dHrNKRbJNNDiIZWUuZ0+z4q0MRdF4XSJ9pIUakTX63jvI/J4XKSW+1Jb1VWWxkOZ6NtMCZEg3E6XViKy7zuzy0qweF0oT+bmIfTobrPOB2qlXZ5q906VP3zqLriEu/vTQghhDhT9RL0eOSRR8jPz2fUqFHYbDaGDx+OyWRi1qxZ3H///fVxyWajfKlIU8j00Gg0hPoZSZdMDyEaXIifEX+TngJ7GRO6hPBEPweRfz6K9o8NYArA2fc2CLhLFVg9T3wNei7vEcPqxJop+gCjO0US5Nv4P+uEON8CffWM7BjBV5tPeNw/oUfs2RUQt6bB1nmw7i0oyoGobnDJixDbF3zqbtlwiJ+h4vOoOq0GyfIQQghRp+ot7/HFF18kKyuLDRs2sG7dOjIzM3n66ac5fvx4fV2yWciwNJ1MD1DjlEwPITxwuaCkCBzn54lmZKCRR8Z3JDbIhyf7OYj+6kq0yRvUTrsV7ZrXYMFNYE13e53T6aK4xIHDUfuTWVDdGE732HLD20cQ5+EGx2zUcefwNvgYpPOTaCIcZVBadE6nKHM4KSopw+WqPcPJ16jnnpFt8TO6zw9fg46bBrZiSLuwU56jhsJs+Okh+P15FfAASN8Fc66CY3+c2blOITLQh0fHd/S477ahbWRJmxBCiDpVL5ke5fz8/OjXr1/F99u3b6dPnz44HOc/jfpCkWG1oddq8Pep1//r6kywn0EyPYSoyuWCvGOw+3s4shKC46H/bRCSAKb6K8Bs0Om4smcsI1vqiFp8M7g8BCaSN0HOIQiIoszhJDmvmIXbktl4NJc24WamDmhFy1A//Izunz9Op4sTecUs2p7CmsPZtAr146aB8cQF+1JYUoYLFQD1FMCIDfZlwV0DeWf5Ib7ecoIyp4uxnaP42yUdaRkqXZ9EE1CcBzmHYcNsKEiDjpdCh/EQ3Oq0T2EpLuVYdhFz1h4lzWJjdKdIxnWJqrXzWatQPxbeN4R//bqf3/amM31QPFf1iiOrwM7uFAsH0gtoG2Em7nS7p1lTYf9iz/t+fhRie0NA9Gm/p9oYdFom9IwlKtCHV37Zz6HMAmKDfJg5tgNjO0diNjWNv3GEEEI0DfJbpYnJsNgJMRvRappGK7cQPyOHMj0XKxOiWcrcCx+NB1t+5bbNH8OVb0K3a8BYfzf6wX5Ggu1OSNni/aDEpRA/mL2pFq5/bx3FpSpIvToxiznrjvHmlD6M6xKJsUrq/P50K9e9u7YiVT0ywMS4LlF8seE4C7en4HC4uLxHDPeMbEt8WM122y1C/HhqQhfuH90OFyp1398ky1pEE2CzwKaPYdkzldsO/Q4r/wm3/gphbU95igJbKQs2JfHior0V21YnZvG/3w/y9d2DaBPh7/F1ep2W9lEBvHp9L6y2UtIsNlYfzGLBxiSyCuz0bhnCbcNaU+JwnV6b++TN3vflHQO7pc6CHqA+jy7uGk2fViGUOJzotRppUS2EEKJeNM6y3sKrdIuN4Ca0xj3EzyCt54QoV5QDCx9wD3iU++lBKDy3Fq15RSUcSLeyeEcqaw9lk5JXXLMQqFanihN64x9JbmEJDy7YXhHwKOdywcNfbXNrQ51TaGfWV9vd1ua/cHU3nvx+F5+uPUZeUSlWexnzNyYx8e01JOV4Tv/3MeiICfYlNthXAh6i6ShIcw94lCvMhKVPgd16ylNkWu28tHhvje05hSU8/9MerDbPxX5dLhcFtlKyCuwUlTh4e8Uh/rPkACdyi7GVOll7OJs752xif5rF6znc+IZ436fRgrZ+ltWGB5iIDfaVgIcQQoh6I5keTUyG1d4kipiWCzEbsdrKKC5x4GuUtfmimSvKUUtIPHGWQcp2tczlLGRYbDz5/S6W7KmsyRHka+CTW/rTPS4Ive5kjNscAb1uhI2za55Eo8HRdizZhXa6xQWSlFNESbX6HLZSJ8eyiyrS7nMLS9mdYqnY36NFEPvTrZzILa5x+pzCEuZvOM6D4zpg0EnMXVwADi33vm//z2rOn2LZ2trD2Xgrv7HiQCa5RaUE+Kjf++XLzpbsTic6yIedyXmUOJxc3SuOpXvSa7ze6YL/LDlA55jAinN4FdMT9CYo81CHq8N4MIfV/nohhBCikarTvzp37NhR69f+/fvr8nLNUrrFRnATKWIKlQVXJdtDCMDlpZ6R3gRaPTjOruhvqcPBp2uOugU8APKLS5n2wXr3+ac3wdCHILKr+0k0GuyX/Y//rrNy4wcb0Gk1fDC9H11jA2tcr2q7SWe1u7W+8SFeu7EA/LwrjbyikjN4d0I0YmVV5pZWD76hld+7nJ5r51Rjr6V9q8ulauaof7vYfiKfS15bxbz1x8AFQ9tFEOJrZPUB73MuMaMAW+lpFBcOiIbr56r3UVVwK7jk5XqtOSSEEELUpzrN9OjVqxcajcZjxfDy7ZomUouiscqw2OjRIrihh3HaQs0q6JFusZ3emmIhLmS+wRDWDrIPqu97TFZ1PGx5oDNBVFfVAUJXy0ezo0wVHLSmgKMUglpQoAnh07XHPB5eVOJgW1Kee0HEoDi48RvI3AeJS3CaI8mMG8t/1xcwf3sqAN9uSWbJ7nTentaHe+ZtprBEBWx0Wg2twirPFehroGWoL0k5KrPDXuqs0VHC7Udg1KHTnp/fAw6niwyLjTSLDXupk9hgH8L9TfhJkURRV9qOhkMrYMyTKgBis6r5lXUANn4IPkGnPMWgNt4zKHq0CCLw5JLWdIuN+z7bgq3UyTV94ogMNHHvZ1vw99Fz48B4r+fQaMCgO405pzdB6xFw30ZI/BVyj0Gb4RDdU70nIYQQoomq07/8jhw5UpenE9WUlDnJKSolpAktb6ka9BCi2fOPgglvwJwJMPJxFeyYP1UtbQG1pv66T6HVINB7yOgqtcGRVfDt7ZV1QXQGgkY9wcPDRvDsb6keL3ssy0MdjcAY9dV2FHuT85nw5h9UL/9RYC/js/XHmNg7jnnrVbvx24a2pthembESFejDPyb14KYP1+N0wbJ96cy6uCMr9md6HMutQ1oTaq7/dpQlDgdbjuVx97zN5BWpegZ6rYZ7R7ZlxpAEr2PILyqhzOki0NcgS3AuVAXpkHUQ9i1SQYkuV0Jg7GkFKGoIjIfRj8OCG8GSorZpNNB9Mlz9NviF1vpyh9NFkK+Ba/u04OstJ9z2GXVanr+qW8Xv0ezCkoqsrS4xQSzclkx2YQnZhSW0jfBHp9XUrOEDjOoYSUTAac45gw+EtYGwe07veCGEEKIJqNOgR3y89ycN4txlFqjU95AmtLzFx6DDz6gjLV+CHkIAENcX7lkHJzbA78+77yvOhc+ugXvXe+76kHsU5t8AzirLZBylaH97hmmT5/NhiK/HWhq9WgXXOqTPNxyvEfAot3xfJq/f0IvdKRamDGhFal4xGQU2oPIGsU98CD8+MJTXf0tkW1IeWq2GCT1i+HGHexBmaLtwhrUPB1TWWnGpA4NOS7i/0a0bTF1IzbVx84cb3GqSlDldvPH7QTpEB3BFj1i34zOtdjYdzWH26iNYbaWM6xLFDRe1omWIr2QoXgAcDicZVjshzhxM39+K5vi6yp0rXoKxz0Hf6Sob60wUZ8K8a1UAs5zLBTvmQ0g8DJvlOYAJnMgt4vutySzakco9I9syoE0oH/95lKwCOwPbhHL/6PYkVMmqKnNUTlKdVsOyfZWFjxdsTOKJyzrz/KI9bvVBIgNM/HVcB0x6CeAJIYRoviTHtwkpz5ZoSoVMAcLMRqnpIUQ5gw/4hcDatzzvd5TCrm9gxCPu251O2DbPPeBRhfGPf/HI8Df4y0L3ZS6tQv1o66XlZTmTwfsNkUGnITrIhyHtw3ljWSIpecX8/vBIt2N8DTq6xgbx6uSeWIrLKHM46dUimGkD4/l28wnKnC6u69eCdpH+GPValuxO48XFezmWXYTZqOPGgfHcOrQ1UXXYveHnXWk1irCW++/SRAa0Dqt4+p1dYOfvP+xi8c60imMSMwr4bP1xvr9viCzNa+IyrXa+2ZzEtuM5vBKzHJ+qAY9yvz0N7UafedAjeYt7wKOqDe9DjxtU5kQ1STlFXPvuGtIt6mHGX+Zvo12kP3cMa82gNmEel2GFmo2E+BmYOiCeiAAjxiqBjF93p+Fj0PLh9P6sPJBJltXOoDZh9GoVzG970gnxMxIXUkvXJiGEEOICJqH/JiTjZOAgxNx0Mj1AjbdJZnqUlUBhFl7L6gtxthylkOe5BgcAaTtVkKOqMjtk1lIMOvcoY1r7uAVFh7UPZ97tA4gOqj2YMLGX9/X6l/eI5YPVR3jz94OcyC3mxoHxmL3U7ChzuFi8M5WLX1vFmFdXcufcTWg18MDodvRPCCXc38TqA1ncOXczx7LVkpvCEgfvrTrMw19uI7vg7Aq5erIn1UNb4JOScoooqxIQOZ5T5BbwKJdfXMprvx2gqEo7XtG0WG2lvLpkP//4ZT/XdDQStP1D7wdv++LML5B1wPu+4lyPxYlLyhx8suZoRcCj3MGMAh79ZidHs4s81p2JDDAx97YB7DiRx4uL9nF1tXm7cFsKd87ZxKGMAm4eFM/C7clc8b8/iAz0wYX777GUvGJ+35vOW78fZOmeNJI9ZIgJIYQQFwrJ9GhC0i129FoNAU2sCF+In5GU/Cb2B9W2z+HXx9UfrbG94dqPILTm0zohzoreByI6qqfEnrQaBNpqMWmtDqK6QeISz6+J7Iyvjw+L/zIMq60Uo15HmNlYUQSxNi1C/ZgxOJ5P1rgHYlqE+DKpdxx3zN1I2wh/bhzYCqfLRb6tjIiaTV3YeDSHFxbtrfjeUlzG/E0nWLg9lV8fGoZRp3PbX9UfB7NJzbcR5l839T76JYTyw3bPNU7aR/m7Zbf8tCPF63kW70zlsfGdpPhpE5VdUML8TUkAmA1asFu8H1yQ4X2fNzE9ve/zj1JzvZrcwlJ+2O79v7kvNhxncLsw9NU+A2ylDv716/6K7ki3Dk2gY1QA+9OtFceUOV34GXXsTM5n49FcAN5ecZBBbcNwOJzodFoOZliZ8v76iiWzACF+BubfOZCO0R4mthBCCNHEyV9xTUi6xUao2djk1peHmY3sS6vlD83GZus8WHgftBml6i9s/wI+vhzu/gPM3qvsC3Ha/EJhzDMw58qa+4z+0Omymtv1Ruh6Nax/B0o9BBEH/wWtXyixRl/gzNLYQ/yM/GVMBy7vEcvctUfJLy5jcNsw2kb4s/1ELs9c2Y0sq50vNhznQHoBCdPNtIt0XzKTVWDn3796fupdXOrgtz0Z9IkPqXWp244T+XSLO4tikh6M6hhJgGk/Vg9ZGo+O7+RWyFSr8Z70qKFpfd4Kd8l5xRXJemuTS+gfPxzDwV89H9z16jO/QFRXCIhRHZWqG/pXCPZQ60yD1/+qIvxNRAeaPO7PLixh5YHKAsEPf7Wd567sRoG9lN/2ZmDSa7m4azS5hSW8/PO+iuNO5BZzIN3KT9tTmNSnBX/5YqtbwAMgt6iUO+Zs5uu7BxFZh8vMhBBCiMZAlrc0IekWe5Or5wFqeUum1e6WTt5oZR2Enx6C9peoP1jbjISLX4SSAvj5bw09OnEhiekFV78LPsGV28LawYxFENTS82vMkSrrKKR15Ta/ULjsXxDWHox+nl9XRbrFxqajOczfcJw1B7NIzVMBlFCzkf4Jofzrup7885rubDmWy+1zNvH7vkzWH85h/ZEcjmQVAnjMHiktc3I4q8Drdbcl5VFoL6O2brWBvnUXh48L9mX+XQNpG1FZjyPQR88r13SnZ0v3wMqEnjFez3Nlr1iCzU3vc1coAT6V/019ujmbzIseBZ2HJaJh7SCuz5lfILQ13PSdms/lDH4w/FHoPKFmxhZqrl3Xt4Xbtt4tg3nvpr48NK4DbSL82ZtqJataYMJSrLoQaTVq6dr4rjF8s+UEi3emccuQBMwmPS8v3ssLi/a6dXHRazU4XS7+s/QAj327g4l93K8NqvZP/4QQiks91wwSQgghmrJ6yfRIT09n1qxZLFu2jIyMDFzVaiI4HPJL9WykW2xNqnNLuVCzEacLMqx2YoMbeSG1xQ+DXxhcdKdqOwhgDoe+M2DNGzDoPpX9IcS58g2C7tdD62FQlANavcok8o/y/prAGHA54dJ/gqNE/dvgq5ZeBbc65SWP5xQx/aMNFcELUHUCPrt9AGH+Ro5mFTF33VHyisq4uk8sD43rwM+70tielEfrcDMP3jOYLKud+LCawRWDXkvrcDMH0j0HPhLC/dielMuYzlEs3ZNe88dh0BEfeuqgzenSajV0jQ1i/p2DyCksodThJNRsJDLAhL5aK9oWIX5c168FX21ybxka4W/iL6Pb4WuQpMimKjLARFSgiXSLHYutjEdWFPOv6xcTufZ5dEdXqvnT+0YY8qBqW3tWF+kMN3ymlkOWFqtApH8MmDwXwDXotEwdGM/Pu9Po2yqEsV2i8DXouGfeZgpLKv8+GtgmlNcm966oyRPgY2BIuzDuHNaGFQcyOZxZSLtIf0Z2iMBHr2PRjlTsZTUfLlzSNbqihfTqxCymDYjHz6ijqMSBv0nPs1d1xVbi4Pf9Gbzy8z5uHBhPfJgfWq0Gs1F/WkvkhBBCiMZM46oekagDl156KcePH+f+++8nJiamxnKMq666qq4veU4sFgtBQUHk5+cTGNh417OO/c9K2kX6M31wQkMP5Ywcyy7ksW938s09g+kbH9LQw/Hu0HKYezWMekLVVKjK6YAf7oPYPjB5boMM70LUVOZeo1Nqg6JswAWmQPA59c8ur6iEO+ZsqljnX9Wj4zuSZbXz4Z9HAVXL45kruzLzi61uN2FaDfz3+l6Ai4Ftw2t0W/l1dxp3zd1c4/y+Bh1vT+vDfZ9v4Z1pfXjupz0cyqwMvJj0Wv51bQ86xQTQIer0/zsosJWRXWjHXurE7KMnykNA43RkWu3kFNqxlzk5ll3E4h2p9GoVzOU9YmgRUneBmMakucw9l8vF7hQLU2evw2JTS51CzUYeGBzBlJ4h+Bj0KrCtr5taMqfL6XRxOKuQt34/yBU9Y7j/860esyymD47nicu6UGAvxVJcyvGcYm7/dJNbZyKjTsvb0/oQ5Kdn2mz3Ns3tIv158vLO3D1vM7ZStf2+Ue1YsT+D3SkW3rmxD28tP8iuZPclqBd3iWJou3CW7EnnkfEdaR/pj69Rgn9CCCGapnr5DfbHH3+wevVqevXqVR+nb7bSrTb6tw5t6GGcsfLChKn5xUAjDnqs/CeEd4CWA2vu0+qg85Ww/l3IO35aT9WFqDcGHwjy3nHFk+yCkhoBj5ahvvRuGUKXmECm/7KxYvstQxJ45ed9bgEPAKcLHvt2J3NuvYj7P9/C29P6EBFQGfi4KCGUJy7rzL+X7K944hwd6MMbU3rxj5/3UVTi4KEvt/PEZZ1BA3tTLUT4mxjWIZyvN51gSLvw034/ybnFPPfTHpbuScPpUktjHhzTgat7x7rV66hNSZmDncn5PPL1Tg5lqgyV9pH+vHJND7rGBmIyeO5SI5oOjUZDl5hAFs8cxo6kPBIzCujeIohO0YH4NGDmYVaBnae+30lSbjE9WwV7XVay7XguielWHv12B7cMac0/Fu+r0Yq5xOHk/77bycL7hvDbX4ez9nAOiRlWusQEUuZ0MXP+toqAB0BUgIniEgd940PYl2qtEfAAWLInnUu6RbM/3crVb/3JvNsGMPgM5qcQQgjRmNRL0KNly5Y1lrSIc1Nc4sBqKyOkCdb0MBt1+Bi0jbttbdJGOL4GRj5euaylujajYNPHsH0BjJD6HqJxcDhdJOcV8/u+DLYdz6VjdAD9EkLZl2phQJswYoN98Tfp3W6qwv2NPH1FF/KLy7DaSvl9X6bbOVuE+JGY4XmZSnGpg6KSMnq1DOFwZmFF0CM5t4iNR3O4qHUI3947mAyLHZNeS2SgCbNRR9LJlpg5hSU8/NV2wsxG4sPMBPjo6BgdwF0j2nrt3FJUUobVVoZRpz1ZI8jGHXM2sie1smuFpbiM537ag0GnYeqAeHS1FQ85KSmnmCnvr3e7iUzMKGDK7HX8PHMYbSL8a3m1aCq0Wg0tQvw8Zu1kWW2kW+x8tzWZTKudkZ0i6dMqmPgwz0tTTkdyXjEbj+Swcn8GV/aKI8BHz/fbkokK8GFc1yhWH8hkZ7Kan/cnhHIww+r1XA+N68gN76/Dai/D36SvUYC0XKbVTlaBnR4tgokIMPH2iiKe/2kPuUWlbsfptBqGtg/HiQun08W8dce9XvuXXWmM7BDBV5tP8OT3u5hz20UVP8Pyz50V+zPYciyXrnFBXNwlitggXwx6KRcnhBCicamXoMdrr73GY489xnvvvUdCQkJ9XKLZKe94EGpuejU9NBoNYWYTKXmNOOix/h0IiIVWHrI8yhl8IX4wbPsMhs/yHhwR4lRcLtXtoaRQpdWbI8kv05JXVIrTBUG++tPOVtiVnM+U2esoqpKVYdJree2GXjzw+VZuHZrAhB6xBPkaMJ28Gfn3dT155ofdHM0uYnSnSML83T9XnKcIWhfaHYT7G0jJK6bU4eBwZhG3fryBZ67qxmPf7mRvqvtN3DMTOvPejX2Z9sH6iuBLdmEJpU4nX9wxkPaR/hj1NbMqShwOjmUV8faKQ6w7nE2Yv5F7RrSjQ7Q/e9M83yj+97dExnSOOmX9IHuZgw//OFLjqbna52TuumP836WdPI5LXBhyCu0s3J7C8z9VtlFeuD2FVqF+zLn1IhLCzeQVlZBffBrzsqQQCrMoKbFx6EQJs//M4/LuMSzclsL325LpGBXAg+Pac/Vbf7plXfgZdcy97SLm3nYRqw5k8eWmJPJPFiztnxDCpqO5FR2ITjUvSx1qv69Rz+T+Lflhe4pb0EOrgddv6EVMsC83D0wgKbeIj04uafPEXupAf/Kz4XBWIQfSC9hyPI/BbcNIy7cx+b21Fdlg329L4d+/7mfubRfRLz4U7WkEHYUQQojzpV6CHpMnT6aoqIi2bdvi5+eHweCenZCTk1Mfl72glWdJhDbBQqag2taq5S2NkDUN9iyEvrdALa0rAWg9HA4tg7SdENPj/IxPXFiKcmD/z7DsWShIB70JZ89pZHW5hys+OUxxqYOusYG8OLEbXWICa73pTrfYuO/zLW4BD1A37U8v3M2j4zvyt693MKB1GNFBPtw6tDVp+Ta+35rC0ewiQHVVeeqKLm5FPAtsZUQGmMiw1nyqrNNqiA/3429fb+fTWy/iSGYh6RY7T07oQoifgYu7RHMkq9Dtxu6ZH/fy21+Hs+Sh4axOzGRPioXerUIY0CaUuGBfr22496VaufadtRWBidR89X6v7RvHPSPa8vaKQzVek1NYUuPn4UmhvYzNx2rWNym38WgOhXaHBD0uYJnWEreAR7njOUW8tuwAD47pwKPf7GD9EfU3S7e4QF68ujudYwMw6qr8d5F/An57DnZ/g9FZxvDgVnQb+nfSQhP456/7AbhjeGueXrgbW6mTthFmooN8SMmzcSSrkJnzt3HvyLbsTbXwxg29ePqH3RzLLqJNhD+7kvPdxlZegLQ6P6OOwCqdalqE+LHgzkHsTsln5YFMYoN8uaRbFFGBPvieXLYVG+zDZd1jeH/VYY8/n1GdIt0yQcocTv7yxVYm9oqlW4ugGsvf7GVO7pm3hZ8eGEpMlaBjga2U7MIS0vJt+Jn0RPgbiQr08TrvhRBCiLpWb5keom6ln8z0CGmCmR4AYf5GkvMaadBj61xVs6PdmFMfG9MTTAEqSCJBD3GmnA7Y+wP8OLNyW5kd7eaPiM/YxwsXv8zDi5IBdUP2Q3oK+9Ks9GgRRJ/4EGICfci3laLVaAj2M5JTWMKJXM/zKtNqJ9DHgMsFaw5lMXVAPLcNbc2x7CKmvL+u4ricwhIKbGX0TwipqPkxd90xHr64I499u4PqD5dvGZxAkd1BYYmDlQcySbfY+HLTCUx6LZd0jWZIuzDm3zmQu+duqchQA/h8/XGentCVqQPiT+tHlVNo54nvdnnMxPh6czIfTO+HQaepeLpdzqjTVmS01Mao1xIb7MP+dM8ZIy1CfDEZJE3/Qvb7vgyv+xbtSOWSrtEVAQ+AXckWrnt3LYtnDqVdZIDaaE2Dz66DjD2VL847TuhPt6G/ei4do4LZn24lwMdAZICJlyZ252h2IceyChnTKYq4EF/+/et+wv1N/HEwi31pFl6e1IPHv9vJ2M6RLN9fOcYFG5L4y+j2/OOXfTXG+5fR7TFUK+IbHeRDdJAPYzpXdoTKKyrBVmonyNeIQafjpoHxfLvlBFkFJW6v7RDlT7i/qaLWzeC2YRVBwu+2pXBZj1hMem2NbjHZhSVkWO0VQY8sq503fk9k3rpjlHfRjQ704YPp/egSEygZIUIIIc6Legl6TJ8+vT5O26yl5ttO1sZomk8dw/xN7Kz2xKpRcDph8yeQMAyMp7F+X6uHFv1h/2IY81S9D09cYKxpsOw5j7v0SWsYMMhGh6gAHhrXgYfmb6tIawcI8jXw4fR+vL3iILmFpUwfnED3uCCPNx7lShxONBqwFJdS6nBSUuYk1GyoEUh4cfEe/nVtT4a2j+D7rcmk5tmwlzn48s5BvL4skb2pFuJCfJk6oBXtIvwrnl5nWu2kW+wMbhvG3SPasnBbMh/9cZToIB9emNiNjUdyeO/kU+S8arUFTsVSXFbrZ8au5HzaRvizr9oyl2v6xBERcOqlQf4mA/eMbMvy/Zke9985rC1+0q3iguFwusiw2ih1uDDptEQF+WC1ef9vstThqhHwAzWn3l15mOev6qq6mWQfcg94VBG46u/MHPgx9/1gxaTXep3Xr03uVfF9VkEJJ3KLeGliNz5YfYQ7h7fhiw1JuFyw4kAmCeFm3rihF3PXHeNwZiFtIszcNDCeE3nFGGsJ9mVYbKw7nM0na45iK3UyoWcsV/aMoWWoH9/dO4RP1hzlpx0pGHRarugRQ9/4UGZ9tR1QNYDuGt6GmQu2VZxv6/FcOkQFeJyj9jKVAeJ0uvh+WzJz1h5z259msTF19joWzxx2wXZHEkII0bjU2190hw4d4uOPP+bQoUO8/vrrREZG8ssvv9CyZUu6du1aX5e9YKVbbE2ynke5MLORrIIS7GUOTI0pXfzw7yo1eciDp/+aFv1h5SuQlwTBLettaOICZLeebDXrmTF7Lw+MvoS/L9ztdmMEkF9cyt++3sGMwQn8/YfdbF2wjUFtQnnuqq48+s3OGucy6bWY9DoMWi2XdIvmjWWJfPLnUW4d2po+rULYcrxyaYet1MkDX2yld8tg/nt9T3anWogN8sHfpOPWIQmUnnxEG+ij552Vh9hw8un3Ra1Dmbv2GHeOaOPWRnN/upWVBzK5b1Q7JvaO47utyVzZK/aMflSnegIc7m/C38f9V9iQtmHMHNvhtIPDHaICePyyzrzyyz4cJ9+jTqvhycs70zby7AtZisYly6oKlb6z8hA5hSXEBPnw2PhOjOwY4XGJFECvlsEc9FLMd/2RbArsZfga9TiPr8VrqCHnMG2CNLhcEOFv4o45mzzO6+d+2sPrN/Sq2PbnwSwCfQ2sP5JD71bBvDypO09/v5sSh5NP1hzll90mnrisCxH+JlYlZvLFhiRmXdIRp9NzzY9Mq42Hv9zO6oNZFdv2pFr4dM1Rvr5nEC1D/Xh0fEfuHN4Gh9NFmUPVtBnUNoyeLYJoF+nP0z/sdgtcGvRajzVGDDoN0UEqyyPdauMdLz9fi62MrcfzJOghhBDivKiXoMfKlSu59NJLGTJkCKtWreLFF18kMjKSHTt28MEHH/D111/Xx2UvaGn5tia7tAXUDQqo93EuVfHr3JY5EJIA4R1P/zWxfUCjg8Ql0P+2ehuauADpfdRSKqfnmhNOv0gCfHRel4IdySokNriyRezawzncPCiB6EAft6UkALcNbc3Cbcm8PKkbz/+4l99Ppsl/seE4L07sxj3ztlBW7SZJp9Ww7UQ+7644xD+v68GhzELaRfnz3opDpFvtrDucXZGi3qtlELZSJ1f2iuXfv+73uAzl3ZWHmH1zP/anWegcE3jaPyaAYF8Dw9qHszqx8katU3QAA9qEUeZwMqJDOKM7RXI8u4jMAjttI/2JCjB57QDj8Rp+RqYNaMUlXaPYl2oFjbpGRIBJsjwaMZfLRYG9DL1Og6+h9v+frLZS/vvbAT5bX1mbIjXfxswF25h/x0AGtQll7WH3OmM6rYaHxqp6Hp5E+JsqlpI4AuK8Bz0MvgT5q5v6nKISUrx0MDuSVVhRvBTU78vsAjtD24UxtnMUB9KtvHdTX1LzbfgZtbSJ8MdW6mRXch5jO0cxqlMkK/dnEOZvZHiHCCL8Tfj7VNZS25tqdQt4gApO9GwZzPakPML8jfga9EQFVgYL/zquI3PWHuH7bSk1ChMDjO4YyZu/H6yx/YHR7Yg4Wfy01OEiu7CkxjHlEr0sLRNCCCHqWr0sWH7sscd44YUXWLp0KUZj5Y36qFGjWLt2bX1c8oKXkl/cZIuYAhXdIZK91B9oEIVZsG8xtBt3Zp1YjGaI6AiHl9ff2MSFyRwOna70vM8niDRjKwrtnpeqlKtew+L7bcnMvrkvnaIDMOq0tI/058WruxHiZ6RPq2A6RQdWBDwAMqx25q49xvs392NY+3BMei0RASbuGdGW24e1Zn+ahVev74lOo+GVX/fxzA+7uap3HEa9FoNOW5Hq/s9re/LCoj3EBPlyKLPQ41gdThe5hSV8fEt/ogJ9PB7jTaCvgWeu7EqIn4EQPwNvTe3DxN5x7E+zkJRTxL40KwadhoFtw5jQM5YuMYFnFPAoZzbpiQ8zc0m3aC7pGk18mFkCHo1Ycm4xc9Ye47ZPN3H/Z1v542AW2V7auAJkF5Tw+QbPbVnv/mwzr1zTg7+Oa09kgAmTXsuQdmHMvrkfRr2mRiCx3D0j2xF88vexNaIv6Dz/bi7qfiNOvwhuGZyA1Vbm8ZhyVfdf2TOWO4a3oU98CK/8so9ViVkUlThIyy8mt7CUvy7Yzs0frSfI18js1Ye57t21vPH7Qf7+wx5G/2clCzYlYTkZRClzOpm/0f39j+wYweyb+xHsZ+DjP4/y/I97OZBudWtr7WvUcXHXaDIsNX+2sy7uQGxwZV0Oo05L2wh/3pram5sGJqhlP6i6OdG1zPtucUG1/kyEEEKIulIvf9nt3LmTzz//vMb2iIgIsrO9p3afyssvv8zjjz/OzJkzK4qlulwunn32Wd5//31yc3MZMGAAb7311gW3hCYt30a7yNOoOdFIhZ1s83eiMRUz3fa5Cna0GXXmr43pBfsWqSf22ka0XEc0biZ/uOQFyD4I6VWWpJgCSL/yc55elsujl8ag12pqZGGAWrKir7bsQ6PR0CkmkM9uH0DJydoeOq0GF+qJ9PfbkmucZ1ViFjuS87m2bwseHd+JnEI77SIDKHM62XI8l7WHs9lwJIeknGKScorZk2Lhmr4tuKpXHA6nk8gAH5bvy6BDVMAp37K/j56owNrbx3rTNsKfH+8fSm5RKQ8u2FZRVLH8PYzpFMk/rulORMCZBVRE03Qit4jr313rljGxbF8G1/Vrwf9d2sljO9mUvGKPtTlA1Zmx2Mq4b2Q7Lu8ey47kfPakWPjrl9u4pk8L7h/djreWH3R7/YzB8fSND674flWGgSFXziPix5uhrHJcZS0GcbD9bVAM47pEEhXke8p5rdXAP6/pQaCvgamz12GpEgj5ZVca945sS1yIDy5cxIeayS0q4eddaW7ncrng+Z/2MqB1GN3igtAAVT8xusUFcnWvOG7/dFPFWDYdy2X+xuN8cHM/hneIQH8yi6V1uD8L7x/Ckt3pLNuXToS/iRlDEkgIMxPsZ2RURx96tAiipNSJXqetUUsnKsDEQ+M8Z8yE+xsl6CGEEOK8qZegR3BwMKmpqbRu3dpt+9atW4mLizurc27cuJH333+fHj3cO2b885//5NVXX+WTTz6hQ4cOvPDCC4wbN479+/cTEHDqP8ibglKHk0yrvUnX9DDqtYT4GRpPpofLpQqYxg8GnzNLuwcgthds/xxStkGLvnU8OHFBC2oBN34DeUchdTsEtcAR2RWHK4ybBuUSEWDipoHxfLzmaI2XzhicwMJtKW7bpl7UCoNO6zXLwduyuLyiUj784wj9E0K5a+5mfrx/CN1bBHPzoASOZBXyvyqp6xZbGR//WTmeN6f0Zvaqw/zrup5YbKW0j/Qn0UP9A51WQ6foc/scjgn25YftKW4Bj3LL9mWwN9UqQY9mwF7m4L2VhzwuEflq0wluHBDvMehhNtX+Z45Jr0Wn0xLsZ+CD1YfZnWIB4MM/jnBd3xbMu+0iCu0OTAYt8aFmQsxGgnwrl47Ehwfz4M9+PHnNMsIse9AVZWKL7MWmPH8WrrfSLhLeX3WYp6/ozB3D23iscXH3iLb4GnUsnzUSo07LzAXb3AIe5d5ZeYhfZg6jfaQ/3eKC3dpMVzd33VFeuro7Op2WKRfFs2inCo7cMqQ1Ly7aWyP44nTBQ19u5+eZw4it0m62RYgftwxJ4IaLWqLXamsUSw3z8DMvp9FoGNclkqyCjvzv98SKNtadYwL435Q+btcRQggh6lO9LG+ZOnUqjz76KGlpaWg0GpxOJ3/++SezZs3i5ptvPuPzFRQUMG3aNGbPnk1ISEjFdpfLxWuvvcYTTzzBpEmT6NatG59++ilFRUUeM02aqkyrHReqGGhTFu5vajxta4+shJxD0H782b0+rL2qz3B0Vd2OSzQPAVHQcgBcdCd0vAxdSDyxof5c168ltlIHLUJ9ef6qrhWp4XHBvjx9RReCfA0s2placZrh7cPpFOM5qGAvU+nwrcPMXNIlis7RATw2vhP/uKY7dwxrQ4ifgRHtI1h/WGXflafXtwjxO+VSuhKHE4fLxQNfbKXIXsZLk7p7bBP713EdKC5xkFvLuv5TyS60M39jktf989Ydq8hwEReu3MISvt1SM2up3LdbPAcAIgNNRHrp5tM1NrDiYUKYv4l/TOqOz8k2xXqths4xAbhcKk/CXuokt8iOrdS9Hk+Y2Uh8RDCXfnqMicvDmbKtG+MWWHlySRo3D0rgy03qv93nftrL9f1a8I9J3SvmdYsQX/5zXU+mD45nWPsI4sPMlDicFYWCq3O5VJHTHi2CGdEhnOxC78t60vPtFYGNjtH+jOwYAYC/SU+ml+VA+cWlZFpr7tNoNPgZ9bV2h/Em1Gzi9qGtWfrQCH58YAhLHxrO3NsGNOnMVSGEEE1PvWR6vPjii8yYMYO4uDhcLhddunTB4XAwdepUnnzyyTM+33333cfll1/O2LFjeeGFFyq2HzlyhLS0NC6++OKKbSaTiREjRrBmzRruuuuuOnk/DS01XwUKanui0hSE+Rs50VgyPTZ+AMHxEHWWy6B0BojsAkdWw9CH6nZsollzOl08/9NePrv9Iv4yRtUOKLCVER/my7rDOfRqGYxJr+WG/i0Z0i6cSA9ZDsdzCnl/5WEW70rDoNXw6uRepOQV896qw2RYbHRvEcQbN/QmJsiHxIwCBrYJo0WoKrhYfiPlb9JTYPdch6BzTCCvXNMDi60UhxO+3JjEB9P78fPONHYm5xMT5MOMwQn8tjed8a+v5tYhCTwwuv1ZFWN2uajorOKJai3qfX9eUQm5RSWUOVwE+hrOuLaIaBxc4HFpSLlSD4V0AaIDffhwej+mzF7v9t9zhL+JN27o7ZYh1TkmkJ9nDuOz9cfp3SqYCH8Ti3emsXhnKmVOF2M6R3Jd3xboNBrCA0wczy7EaivjtmGtGdclik/WHiWnoITJ/VsyqmMkLy3e69bxJC3fRvtIM09c3hmjXotOo5ZwZVrt5BWVEmI2UstbBKDQ7mDh9hRO5BbTNz6EX3enezxuTOdITCe7GEUE+PDPa3uw5VgeRl3tgQtPHVnOlcmgo2WoHy2RTi1CCCEaRr0EPQwGA5999hnPPfccW7duxel00rt3b9q3b3/G55o/fz5btmxh48aNNfalpal0zaioKLftUVFRHDt2rMbx5ex2O3Z75dMMi8VyxuM6n1LyVDpvU17eAuqPzG1JeQ09DMg9qupxDLj7zAqYVhfdHXZ+BY4y0Enhw9PR1OZefSuwlZJVUMKRrEJ8T94YhPmbMBt1bEvK5+ddqexKrvwZ9WkVzMA2oZQ6XLQM8yOroIRMq51Ui42EMD/CzCastlImvr2momvC7cNa8+2WE3xT5Sn5nwezWXsom/9O7sVbyw+SmFHAiPYRPHNlV37YlsLWpDxuG9qa15cl1hjzZd2jMRt1RAWY2HI8j3dXqnT977YmM6ZzFMPahzOwTRhGvYaPTi6J+ejPo0zsHXdWQY9Qs5GresV6bS065aKWFTd31R3KLODxb3ew/ohqzxsX7MtzV3VlYJuwUy57uNA09bkX5Gvgsu4xfLfVc7bHxN4tyC20k2kt4XhuEWFmIzHBvkQH+tA1NohfHhzGlmO5JGYU0KNFEF1jg2osr9DrtLQO9+fR8R05lFnI3XM3czS7qGL/V5tO8PveDBbcNZCtSbk8++OeikD+6I6RPHNFF5bsSWftoWw+XbPBLYBxSdcodibn89LifYCaR+O6RDP9440cz1HX6N0qmNcn96J7XBA7k/M9vs8xnSN5bVkiqXnFfDSjP7/vy6hR3Djc38ioTpFu2yIDfBjfLZqknCICffVYimsGNM1GndesGCGEEKIpq5flLeXatm3Ltddey/XXX39WAY+kpCRmzpzJvHnz8PHx/nROU+3G1eVy1dhW1csvv0xQUFDFV8uWLc94bOdTan4xvgYdfsamXTAzIsBEar6t1qe258W6d1UHljajz+08Ud2gtAjSPLc1FDU1tblXn7IL7by5/CCj/7OCWz7ZyA2z13Hxf1eSYbHx3FXd+GzdMWaOae+2bEQFGQ5TYC/j551p3DB7LRlWO7d/uomxr65iwcbjzF13rCLgodNqGNw23C3gUc7pgv8sOcBNAxNwuWDFgUxu/HA98eF+LN+fgV6n4bFLO1UUJ/Q36bl3ZFsev6wz0z5Yj16nZfbqwxXnK3O6+HV3Gm+vOMRDC7bV+Az+erP3+gO1Mei0TB3QiqjAmjdjvVoG07NlsMfXncgtYvJ7aysCHgDJecXc9ukm9qY2rRv+utDU556fUc/MMe3d6mmUG9kxguggHx74YhsXv7aK2z/dxMS313DN22tITLei1WpoEeLHlb3iePjijozrEl1rPQmH08WfB7PcAh7lsgtL+GrTCeZvTHLLXPx9fwaT3llDz5bBrDiQWSNj49o+LfjPkgMAhPgZmNi7BX/9cltFwANg6/E81h/J4a/jOnhcLja5f0sMei1PXNqRUH8jGg28NbUPPVqogqAaDYzsEMGr1/dC5+VvoJggH16e2N3jvmev6lajGKkQQghxIaiXR10ul4uvv/6a5cuXk5GRgdPpnnb67bffntZ5Nm/eTEZGBn37VhaKdDgcrFq1ijfffJP9+/cDKuMjJiam4piMjIwa2R9V/d///R9//etfK763WCyN+g/AlDwbYf7GWgM5TUFEgA9lThdpFhtxDVXArDALtnwCXa4GwzmmuYe1U60Kj62BuD51MboLXlObe/VpzcFs3l152G1bYYmDye+v47e/juDNqX34ZXcqH83oz9ebT7DleC7h/iau7duC/OJS/vGzemK85XgufVqFsOV4Lq3D/flsw56K88UE+XDQQ4HRcsdziiraSQOcyC2muNRBVKCJ/yw5QN/4EGZd3JHoIBNBvgZ8DTryi0sZ2zmSQ1mFXgOY2YUlZBW41/HwtlQmu8COxVaGTqshxM9AgE/Nm9oWIX58c89gvt58gh+2pWDUa7lxYDxjO0d5XK6Snm/j930ZNcZQ7uWf9/Hh9H4VbUfzikrIP9niM9jXQFATbg/uzYUw9+LD/PjxgaHMW3eMJbvTMJv03Da0NYPahPHaskT+OJjldnxyXjE3frie7+8bQkzQ6f/OKSpx8NveDK/7f9mdxpT+rVhW7ZjcolIOpFkZ0SGClQcyK7brtKqgaonDyaiOkdw1vA32Mif/ua4n321NZnVi5bjNJj3vrDzE7Jv78c2WE2xLyquY95biUlbsz2TGwATGdY3hmR92syfVwtQBrbhnZFsANhzJ4f7Pt3D/6HbcMaxNjb8b9DotIztGsvC+Iby5PJH9aQW0jTDzwOj2tI/yx6hv2g9XhBBCCE/qJegxc+ZM3n//fUaNGkVUVNRZ36yPGTOGnTt3um275ZZb6NSpE48++iht2rQhOjqapUuX0rt3bwBKSkpYuXIlr7zyitfzmkwmTKam8zQjJa+4yS9tASqeIJ3IKWq4oMcf/wU00GnCuZ9LZ4CITnB8DQy+/9zP1ww0tblXX7Ksdt7wsHQEVMbEd1uTeWhcBzpE+5NbUMK4zlGM6BBOVkEJby0/6PaEeVeyhbaRZrIK7EQEmvCpctNSUuasKMzoja5aC9y9qVZah5tJt9jZfCyXuGBfRneK5Lkf93AosxCdVsPYzpFMiQ4gLtiH5Lya3TTAvU0mwNW93Tt32Uod7ErO56mFu9ibakWjgdGdInnq8i4khJtrnK9FiB8PjGrHjQPi0Wrx2KkD1I3u28sPYiku9bgfYFdyPsUlDvxNThIzCnjy+11sPqYyQga0DuHZq7rRPjKgxs+mKbsQ5p5Go6FVqB9/u6QDdwxrjU6rIdRs4nhOEd94ySRKt9hJyilyC3oUl5SRVVBChtWGUacl3N9EVKAP2pP/f/sZdfh6uflvHW6mf0KI13m15nA2j1/WidGdItmZnE+Ev4mLWodi1Gt55ZoeHMoo4M65m7DYygjyNXDTwHhu6N+KD1cfZkdyPlqNho1Hc7hjziYu6x7DlP6tyC8urZj3L03sxuLdaRTay9h+Io+sgpKKDBK3cRzM5uZBCfh4WPplNunp2TKY/07uRXGJE1+jDv9mttxLCCFE81Ivv+XmzZvHt99+y2WXXXZO5wkICKBbt25u28xmM2FhYRXbH3zwQV566SXat29P+/bteemll/Dz8/t/9u47PKoye+D4d3qfSZ/0QkICofcqYEOxF2wIir2vbZtlV13b6q71Z1vXvopdFLuAFOm9BkKA9N4zk+nl98dAYMiEYnp4P8/Do7n3zr1vkrm5c8897znMmjWrXcfuScoa7H2iJWP0gYJxxfV2xnXHAGr3wbo3Ychlv69NbSjRA2HfokC1xV6eiSN0HbfPd9RORrmVFnw+P3qVgvpmN06vl81FgSe+j184mPdWFbQ8SY41qbh0ZCKD4k1sKqzjyrFJPP7dLgCqLE7iwzQoZJJW8/4BRqWEs+OI2gFmg4rV+wIZEvEmNWcNiuXOjzdxsL6h1+fn552V7Cq38Na1Y/hxRzn/9+teDq9/mB6tDwrMjEoJJ8sc3GVmf7WVK95c05It4vfD4l1VbCtp5OvbJ5EQ3jowKpNJiTpK+r3X6+PLjSXsq7YyOMHU5nYxRhUymYSSejuXvLYK+2EdOdbm1zPz9dV8/4fJpES2Dr4I3U8hkxFtOHQz73B7j1rktPywNrf1zS7mrS1izf5ahiaF4fR4WV9Qx0PnDGREUjgKuRS1Qs41E1P4NfdQJsegeCP3nNGfwlobFY0OFDIpr88eyb9+ymV/TXPLdglhGr7eUsYP28vJiNGTX9PMG8v3sfjeqXySXxw0zavR7uaVJXu5bmIq10xMwaBWYHV6OKV/FMv31LSqXyKVwKiUCM56cTl/PisLs1HdZjZTYoQGxTGKlupVCtrodC0IgiAIfUqn1PQwmUz069evM3bdyp///Gfuuecebr/9dkaPHk1paSm//PILBkPoNo69UWmDnSh978/0UMqlRGiVFNe1nifd6XxeWPAH0ETAoIs7br8xA8FWC3X7j72tIBygUcgYGNd24G1ieiRSqYTSehvXvLOOez/dygerC3l+4R5u+mAD5w2NY1JGJBIJXDEmmds+2sijC3by/upCRiWH8951Y3jxiuGcNcjMJ+uKePjc7FYxuXCtgrtOy+DjdUUty1TyQOr7wSkxV4xN5s3l+wjV0KGozsbmonoabW4ePndgy3K9Ss7jFw7iq80lJEVoePjcgbw2ayQxh01DsTjcPLdwT8jpMdUWJyv2Vgctq7EGntaXN9hxe72tXnNQvd3NN1tKWZdfx6T0KNpK1LhjWgbhGgUfrC4ICngcZHV6+GRdcZsdQYSeRauUoTtKzat+UTp8Pj8VjXbKG+1My4rmtIExLM0NBNlmjkxiX3UzZY2HAnWD402cNzQwbTY9Wsc9Z2Ry76dbeeL7Xby1Ip+Hvt7BQ/N38Pfzs4kzBd7bEglcMjKBt1fkU1hrY/GuKjYW1uP3g83tbbOt7odrC9GpFNzw/gb2V1t5+NzslocEh3vsgkEs2xMIxDg8Xq6dmNrm93zR8IQ+lakkCIIgCO3RKZkejz76KI899hjvvPMOGk3HTmNYunRp0NcSiYRHH32URx99tEOP01PYXV7qbW6i+sjjmGijqmOCHl43bHwPdnwJtppAfY0B58HgS0ER4j23+B9QuBKmPwnyDsyaiRkASKBoDUSmd9x+hT4tTKvkL2dncfl/1rRaZ1TLOTUrhmanh6d/3E3+YU+RITD95aH5O3ht9kguHpHA91vLqLa4uHB4PKcPiOGB+dvZVW5Bp5Qxc3Qi5w6Jw+f3s+DOyXy5sZiyBgcTMyLJjjPywFc7Wp4Ua5Uy3rp2NMkRGhbeO5XPNxYzKjmcFxa2Tp0/aEtxI1VNDkalhHPFmCT6Rek4fWAMGoWM168ehUohJUqnapk2cJDV6WF9fl2b+128q4pLRyXicHnZWtLIP77NIbfSgl4lZ874FK6dmEqsqfV5LAHkUik+P3yyvoinLhnCI9/sxOk5FLy4YkwSZ2SbaXJ4WLm3ts0xrNhbwy1T+hHWB6YW9nVmo4qbp6aHfK8OijcSoVPy3uoCXluylxqri8RwDddPSmN6tpkXFuWxLr+OaVnRpEVqW7J7ogwqHr1gEHPGp2B3e/nrl9tb1aWpa3bx1A+7uGFyGv/6OZe/nZeNhMD70KCSkxShpdHupqzRTrXF2WY7WrfX3xJge2nxXs4ZHMfnt05g2Z4qVu6txWxUccWYZLRKKTNeWsFNp/TjzIFmNhXVM2tsMvMOC1wqZBIemDGQ9QV1DEkwtdnZSBAEQRBOJp0S9Ljsssv4+OOPiYmJITU1FYUiuDDdpk2bOuOwfdLBJ09HS+nuTaL1KgrbG/RwNMHHVwYKiCaNC9TVqMuHb+6EX/4GY26E4VdBeFqgcOmvj8Om92H0jRA3tGO+kYOUeghLhpJ1MOLqjt230KcNjDPy6qyRPLLgUOBhYJyB5y8fTmK4hv01zfy4oyLka11eH80OD8MSTfzx8230i9JxalYMf/hkS8s2zS4v768qZEdJE49fPIi3fstnWKKJK0YnYXN7SYnQ8frVI9lZ3kSEVkF6jJ5Yoxq5TIpereCvMwZS3mgnXKug3ha6PkaUXsneKgvz1hVx6chE9tdYeXdlAR+tLWJUSjgvXjG8VcADQCEN1FFocoQubhofpkEulbKhsIa57x5qV251enh92T42F9fzyqyRrYLBETols8Yl88iCnfy8sxKby8srs0ZSbXHS7PIwvl8kSeEawrRKLA430QYluZWhfz/RBiVun8j06A0UMhlXj0vG6/Xx5m/7cbh9gRoxWTE8esEg3luVz39/K2jZvqTezj++y+Hu0/tzxsAYFu2qYmluNVeNTQ7ab5ReRZRexY7SRiqaQteu2VNpZURyGK/MGslHawopqmvmi1snUFhnY3eFhWi9iv4x+pAFeoO/h0OJt++uKuCpi4dw7cQ0rhyT3BK42FPRxCuzRvDxumKW7anhu21ljOsXyXvXjWFftRWVXEacSc0n64uJ1CmRH3buNTs9VFucrM2vxe72Mi4tkliTmvA+WLRXEARBEI7UKUGPuXPnsnHjRmbPnt2uQqYClB6YFx/VR542xhhVLN9TfewN2+LzwmdzoHwrnP1PMA86tK6pHHK+htX/B8ufDWR8uB2BzI4Jd0Lm2e0ef0jRWVC8tnP2LfRZBrWCGYNjGZkSRoPNjUImIVyrJFKvIq/SQv5RuqMA5Nc24/X5uHpcMv2idbyyZG/I7TYW1ePz+blqbDLz1hbx444KhiSYuGpsMhkxejJj254KGKNXccPkNP4dolCiVAJjUiN4fdk+sswGiupsvLk8n5euHM5Ha4vYWFjP3HfX8/HN44g5oiZRlEHFLVP78Zcvt7faL8CVY5Oosjh4dMHOkOvX7K+jrMHeKughkUg4a5CZzzYUs7Osid/yavgtr4YovZJT+kdx4bD4lo4tBrWCW6ams6KNbI8LhiXwj+9y+Nu52UFTc4SeKUqv4s7T+nP56CSaHB40ShlReiV1zS7eXlEQ8jVv/baff18+rKVTy8KcSs4aFNtqO5srdHDuoGqLk1s/3IRaIeWO0zL4y5fbyCm3tKzXKGR8dftEUiO1IdvgDog1UFB7KKOrxurE6/cjRRKUqWHQKPhwTRHL9lRTUm/jyjHJ/OO7HP63uoBYoxq3z0+1xQnA93+YjOxAIMXicLNgSxkPf7MjaKraRcPjeejcbNGmVhAEQejzOiXo8f333/Pzzz8zefLkztj9SaW0wY5UAhF9oKYHgNkQKLxmc3nQKn/H22/1K7B/GZz5eHDAA8AYB+Nvg1FzoXwbWMpAbYKEUYH/dpboAZC3MJCB0lEFUoWTglQqIc6kCeosUWVxcPP/NnLD5DSSIjQU14UueDowzsjdn2zm/64agcfrb7M17blD4libX9dS3BQCQYOP1xXz+a0TjlpbRCaTcvnoJNYX1Ae14JRLJTx2wSA+3VCM3w/j+kWys6wp6LVpUTqumZBCRaODumYX4Rol5sOmpJw2wMwFw2pZsLXs0M9DAk9cNISEMA01VlfLDaJSJiXGqMLq9NBwIOtkfUE9QxPDWo051qTh7WtHs6Ggnk/WFyOXSpgzIYUhCaZWwYtB8SZumdKP/ywPrskzd2Iq+TXNfLu1nIGxRm6Zmi7qI/QCSrmUxAht0LLtJY1tTitpdnk5PJmnrd+w2ahGJpWEDEJqFLKWBzsXj0jgg9UFQQEPALvby2tL8vi/q0Zw/XsbqLY6W9bFGtX8+7JhPPPT7pZlk/tHhxyH3eVtOQ/3VFpRyqWcNzSO77aVU3agWKtEAo+cP4jk8EM/h5J6Ow99vaPV/r7eUsbkjChmju5drYsFQRAE4UR1StAjKSkJo1Hc/HWEknobkToVcmmn1JztcuYDNx1FdTYGxJ7ge6SxFJY+DQMvgPjhbW+n0EByF/aHiR4I+KF0I6Sf2nXHFXotj9eHvI3OCrVWF/k1zXy0tpA7T+3PX77c1mqbcWkRFNbacLh9LMqp4vxhcShlUlxeHwNiDQyINWJ1uvktr4ZLRyVy0wcbWu3D6vTwly+38egFg2iyu8mI0WM2qFHIg8cVY1Tz/OXDKK6z8WtuFXqVgowYPR+tLWTxriqMGjmnD4hh7rvrODPbzJp9tYxNi+D6SWk889NuHlkQeIIdb1LzzMyhjE6NQKOQEW1Q8dgFg7htWjpr82vRKuWMTY0gxqBCq5LTYHOjU8q487SMA10wbEToFGgUcl5ftveoxZ1jTRrOG6bhtIExSCSgUYS+1EXolNx+ajpnDYrlt701SIAhCSZ+3V3FSwdaCr+1Ip9LRiYQa+qmNttCu+jaaMWaEqlleFIYCWFq9Co5VqeHy0Ynhtw2Sq9i7sSUkBkj101KZf7mUlRyKVeMTuKD1YUMiDWwuyI48HFmdiwPHih8anN5Ka63kRKhRSmX8tD87cyZkMJveTUkhGkYGGsIWWW+yRE8zezv3+zg9lMzeOva0WwvaSTGoGJSRhTRBlXL9+33+9le0sjpA2NYkVcTVN8G4PVl+5k2IKbP1A0TBEEQhFA6Jejx3HPP8ec//5k33niD1NTUzjjESaOk3k6UoW9keUCg4BxAYe3vCHosewZkKhjew9oRmxICtT1K1ough9Aml8dLSb2drzeXsqvCwqjkcGYMiSUhTBMUALG7At1EdpVb2FRUz/OXD+P1pfvIq7JiUMmZNT6Z4Ulh3HOgfkeVxUm4TsHsccmMS48kv6aZTYX1ROpVvD57FBJoc5rMtpJGqi1ObvnfxpZCpqNTw1HKgosfujw+GmwuzsyO5eO1hfzzx0DWyKlZMVw3KZUnv9+FUaPg6nEp3DlvE6/MGslNH2wIusEqa3Rw7Tvr+P4Pp7Rkl4TrlITrlCGzTSJ1St69bgwvLMzjmZ9yW5Yb1XKenTmMIYnH/vtxPNlkJo2Srzfnsia/Dr8fXly0JygzoK7Z1WamgNDzxRhVRBtULdM+wrQKHrtgEI12Nyv31vK/NYU8c+lQ7C4P/aL1IfehU8m5ZkIq4Vol768qpNrqJN6k5uYp6dhcHmL0Kt6ZO4YvN5VidXq4aEQCaVE6/vFtTktraqVcxvbSRu76eDPR+sCYKpsc1DYH6vncrlZw/tA4LhudhM/vxwccWYLUoAquC+Lzwyu/7kUpk9IvWsd/5owKarNcWm9jxd4aFu+qIlqv4pVZI1m8q5JP1he3bFPb7MQTop21IAiCIPQlnRL0mD17NjabjfT0dLRabatCpnV1bVftF4IV19n61BMYk0aBWiGlsLb52BsfrqEItnwEI68Bpe7Y23clifRAXY913T0SoYfyeH2sza/junfX4zlwB70wp5KXf83jk5vHB03TiNApkUoCNzSfri9mXX4dV41NJjFcg8vjIy1Ky8WvrWq5EZ+WFc2g+DDU4+Vc/dbaoIKLn64v5qFzB3LJyAS+2lQacmxSSaBVrc3l5bp317Pw3ikkH3Hj9Ni3Ozl1gBmX189lo5K4eGQiMglYHB5e+TWPU7OimZAeyWPf5jAmLYJfcipbPVGGwPf06pK9PHvpULRtPIE/SC6T8sP2ClbvD6650eTwcP9nW/jxnilHff2JOKV/NB+sKQq5bkRyGBrRAaNTNNnd1NlceLw+DCoFMUZVh9cAizWqefua0Vz13zXY3V6eu2wY//xxN3mHTQf7clMp109K5fSB5pD7qLI4uP69DUToFNx/ViZGtYJmpweHx8vE9Eg2FTVw9VuH6jr9klNJrFHNszOHctuHGwPTaA4rplFtdQZNcYFAMEatkLEuv5YJ6VFIJRLKG+zYXB4UcimROhWReiXj+0WwZn/wZyiX10e4VoFJc+izVmFtM5e9sZoqy6HjfLK+mD+flcWFw+P5ZktgWtnY1Aj0avH+FgRBEPq2Tgl6vPjii52x25NSSb2dyRlR3T2MDiORSIg1qkMWczuqtf8JTFvJPKdzBtZeUVmQ+z34/YFJ1YJwmEqLkzvmbWoJeBxkc3n5w8eb+ezWCS3FPlVyKZeMTOSLjSUA5Nc089QPgcyK2eOTidIrWwIekTolpw+Mwepw8+9fckN2mHjqh128c+0Y5m8ubSliKJXAdZPSODUrmiaHh+cuG4bX7+eNZftYnldNVpODaIOacK2ClftqOW2gmZcX5VHW6EAigUnpUdw8pR9xYRremDMav8/Ho9/msK/ayrSsaLYWN7T5s9he2kizy3PMoIfF4eK3vNBFj5tdXnaVN5F8RP2G32twoilk/RSJBB46ZyDhfaSQdE9SWNvMI9/sZFleNX4/xJnUPHJ+NhMzolDKpFRbnBTX2fADyRFaogyq3xV8kkgkDEow8fM9UyiotbI0tyYo4HHQOysLuGhEQsjfda3Vxb5qK/uqA7VkDu0b3rl2DP/4LqfVayqaHHy4ppCLRyTw4doi9Co5WqUM24FMrsMZNXLkUikzhsSxZl8tyRFaNhTU4fB4KWtwEKZRoFXKyIo18MIVw7nnk82szT80jrGp4Tx3+fCWIr1Wh4cnvs+hyuIkMVxDptmAxeFmY2E9//oll3fnjmHB1jJkEgn3npmJXnX0zjKCIAiC0Nt1StDj2muv7YzdnnQcbi9VFicxxr6T6QGBuh77q08g08NpDbSc7X8WKHpoF4XoAbB1HtTug6iM7h6N0EVcHi/ljQ4W76piT6WFsWkRjEuLJCE8uP5DZZODJnvoDhAFtTbqm10tQY/aZhdj0yKI1Cn5eF0RTQ4PYdrA1JE4kxqby4tEAtMyo/nbedkkhmsprrPxS07o3qt+P+RVWcgyH6oz8M9Lh7Jybw1z3lnXEgiJ0Cl56uIhFNY189zCPazLr+PrOyahlkuDWuH6/bBibw37qq28ctUIzEYVJp2aJy8ewr1nZuL2BqbCbCisDzEaSArXoG7j5tXm8lDe6ODH7eUU19u5blIaYRoFj32X0zI94aCS9ra+PkycScO8G8fzzx9389POCrw+P+nReh67IJvsoxR6FX6f8gY7s/67tmXqB0B5o4NbP9zEB9ePodnl5Z5PtrRkCyllUh65IJvzh8VjPEbr11B8vsB0EYNayVebQ2c8AXy+oYSEMA2r99eyIq+G1EgdZw2OxXug2mmETsmFw+MxG9SU1Ada0m4raWhzf4t3V/HqrBE4PT5SI7Xce0YmT/6wq9V2fzl7ACa1nDeW7uXRCwbRYHfz1op8Fh52TpuNKv41cxj9zTremDOaWquTepubMI2CSL2KiMOCNfU2F9tKGnnlqhE02N1sLqonI0bP7dMy+HJTCTvLmpieHcvNU/rRL6qHZU4KgiAIQifosKBHU1PTsTc6QBQ5PT4l9YEP9Ue2e+zt4kzqVum5R7Xjy0DgI6uHZnkARGcG/luyTgQ9ThIer4+NhfVc+856XN7ATdEn64uJ0Cn57JbxZMQcagXrCjHV43Duw+bUu7w+/vLlNqZmRvOPCwejlEtxuL18vbmUV5fU8NVtE1n2p1MJ1yowHLgB9Pn9R21v6/XBExcN5ukfdxNvUlPRaG9Jbz+ortnFvZ9u4bNbJ/DOinx8/sB0NJlUwquzRuLx+fhuWzmLd1Xi8wduUvfXNFNUa2NyZhTRBjUGtZzSejszRyXyZRvTaa6ZkEp9sxuNQnZELRMPi3IqufvTLS2BmE/XF5MUoeG5y4Zx64cbg56SZ8d3bEempAgtz84cyl9nDMDj86NXyUUrz06yo7QxKOBxuKd+2M1FIxKCpke5vD4emr+DAbFGRqWEn9CxvD4/W0oamP3WWp66eEhLzZxQLE4P//o5N6jmxb9+yeX1q0fy8LkDiTNp+GR9EYt3VZERo+dPZ2Wycl/olscHj50QpsHp8WFzeymsbebVWSP5eF0RBbXNpEfruWpsMkt2VzI2NYLnrxiOzeXh03XFQQEPgMomJ3d/spmPbhzHr7tLuWhEAmNSQ7eb9vn8/GvmMB76entL9pJBJcfl8XHjKWlIJRIuGZkQ1DVKEARBEPqyDmsJEhYWRnh4+FH/HdxGOD5FdQeDHn3rg3esSU1FkwObK/ST71Y2vgeJo0Ef06njahelHsJSRF2Pk0hlk4Nb/rexJeBxUF2zi3s+3UJd86HMhPgwDQpZ6GlPYVpF0FPaeJMao1rO0txq7vl0C7d/tIn7PtvK8rwadEoZMUYVyRHaloAHgEGtYGhi20GAKf2jGJ0awdvXjuZPZ2Xx3qrCkNvZ3V62lzSiVym4f3omq/bW8Ni3OdwxbxMPzd9BYriGF68Y3vK9bC5qoLzJweaiBgAK62yc98oKvtxUyt/Py0Z1WCcYuVTCXadlsLGwnrNeXM6+I7K9qixO7v1sK/4jYjfFdXY+WlvExSMSWpZlxOgxaTo+UVGnkpMUoSUtSicCHp1o1f62AwW7KywkhYeetvTakr00O4/zunFA1YHz1OnxsaW4ntMHtn0dmZYZzZLcqqBlXp+fuz7ezNi0CO6Yt4nf8mooqrPx6+4q5r63nskZodvLAgxPMuH1+fklpwKZREKdzcXfvtlBplnPNRNSSYvS8devtuH1g0ImwaBWYHN6+fzA1LYj1dvc7K9uZkVeDZe8toriNrKdtEoZ89YWtQQ8EsM1vHr1SAprbVzy2ioueGUlN32wgS1F9TjdbQeBBEEQBKGv6LBPjUuWLOmoXQkHFNXakEslhGv71nzyg0+XCmpsZMcfI+unajeUbYJpD3TByNopKjPQwUU4KZQ02GlyhL4B21HaRF2zmwhd4MY5Sq/knjMy+dfPua22feT8QWgVUvIqLfy6uwqH28vrs0exsbCeFxflBWVw3HVaBooQ7asjdEoev3Awl76+qlXdkLMHmWlyuHlp0R6mZEYTplVQd6BjRCi5lU2cmW3G4/XzwPwdLcutTg/vriygJNvMzVPSeXXJXpIjtOyrsbIkt4oxqeG88uteGmxuvthYwrSsaP7vqhFYHB48Pj+D4418sr4Ym8vDdZNS+WVnBbEmdUvxxfUFdW1mqyzaVckrs0bw8boipmXFcM2EFDYW1rfK9qhqCmSfrNxbg9moZnJGFGaTWhQi7WFSI9ueUhGuVdDcRkC8qM6Gw+1tsw0tBKac2ZxelAopWqWciiZHy/vd7fMzd2IqS3ZX0XxExsewRBNOj4/KJmerfTo9PvZVWZnSP4rleTUty+0uH9tLGzhrkJmfdwZnZsilEh67YDAD4wwsuncqpQ12zh4Ui0mj4KO1RTg9PtQKKZeNSmJ4chgWh4e9VRYa7W7sRwlElDbYkUoDQcKP1hRy/1mZKI7otuTw+Fi469B4Hjk/m/s/3xo0RWxHaRMz31jNd3dNZoCYwiUIgiD0cR0W9Jg6dWpH7Uo4oLDOhtmoRirtW4Ux4w8EPfbXWI8d9Nj6MaiMkDi2C0bWTtFZsG8xuJp7XocZocMd64mz+7AMEK1SzqxxyWSZDbywaA/FdTYyYw38aXoWaVE63llZwMu/7g16/blD4njpiuE8OH876dF6rpmYyvaSBqqsDsym1lPeBsYZ+O6uyTy/cA9r8muJ0Cq5ZmIqRrWcOW+vw+Pz88KiPF67eiSJ4RpK6kNPL+gfYyDOqObuT7eEXL8wp5KrxiajlEmYmhXNs7/sJstswOr08sP28pbtluZWszS3Gr1KjlQK95+ZxfRBZj5ZV8zyvGoyog2U1ttQyfWoFTIa26h5AoGn7fEmDW/MHsXKfbXc/tEm3r52dNA25Q12bvxgAzvLDk21PDg1Z2pWtAh89CDTsqJRyCRB07oOumZCKl+3UXdjSIIJrTL079Ht8VFcb+N/awpZX1BHgknDLVPTUckPXT9Hp4Tzzx938/rsUXy+obgle+rC4QlcODyeS19f1eaY65pdnD4whlnjUnhw/vaWQMrTP+zml3unMDUzhjeX76PO5mJ8WiT3nZlJv2gdSrmMxAgtLq+PWW+t5dwhcbxwxXAgUB/nh+3lLNhSylVjk7lz3maemTmUCJ2yzcDkgDgjry3dB8DXW8q4fnIaMcbgn4nHd2i626B4I3mV1lY1cQ5u9/zCPTx/xTBRzFQQBEHo0zqlkOny5cuPun7KlI5rNdiXFdY097kipgB6tRyjWs6+qmMUM/X5YPtnkDIJZL3gA1n0APD7oHQTpJ3S3aMROllalB6JhFbTMYBW7SMDy5SMSYvgpSuHY3V6MKoVmI0qciusrQIeAN9vL2dKZhT/vHQIuyss/Oun3ZQ1Orh0VGLI8SjlMgbEGXn+imFYHV6anW7u+XQr20sbg7Z7dcle7j69P3/6YlurfcQYVAyINVBlcWI9SlCnotHO67NH8v22Mnw+OGtQLHKpBHmIAK3V6WFieiRqhZRrDiucuqO0iW+3lfHu3DFMzohiXFpEm8frH6NnQ2Edj38XKAKZFKEhLVrfst7p8fLakr1BAQ8IBEvumLeJX++fSspRsguErhVnUvPWNaO55cONONyHgoNnDzJzycgEXlva+nyQSSXcMjUdjTL0x5ac8iYu/8/qllogO0qb+DmnkofPHchFw+P5eksZPj/sqbJw8/82cP7QeP46YwAOt5efd1aQHKklTKuk2RU6GNgvWse6/DrSovS8MXsUTQ43v+XVMG9tIQ63l1njkjkz24zX50evlrUKIkTqVZw+IIYFW8tYsDW4ns6nt4xn9ltr8fj8fLmxhGsnpvLCwj2txpAaqcXu8tBodwMgb2PKnEEtp3+MnrwqK1mxBjYVhS4qDLCuoA6rwyuCHoIgCEKf1ilBj2nTprVaJjmsjafXK+aQHo+C2sDT4L4oPkzD/prWbQODFK2GpjKYeHfXDKq9TEmg0AaKmYqgR58XpVNyzfgU3l/duj7Gw+dmYzYGZ2OUNdh5aP52luQG2rBKJfDYBYNYm992Ud8PVhdyZraZlxcHbgKjj+jSEIpepUCvUvCPb/e1CngA7Cxroq7ZxUPnDuT/Fue1TNEZlmji/ulZ2N1eFLKjl3vqF61nf7WVV5bsI0Kn5NKRiUTqVVw8IiHkz2POhBT+GKJeh9fn5/7Pt7LgzknEhamZlhXN0tzgNrUSCfzh9AyeX5gHwNi0cP49cxixh/18aywuPmujDoLX52fV3loR9OhBlHIZEzIiWXTfVHZXWGiwuRmSYCLGoEKnkvHRjeO477OtLdlI8SY1z84cRmpk6FofNVYnf/pia1Dx04Oe/nE3n986ga+3lLFkdxXnDI7jk/XFfL6xJKh2hsXh4eHzBnLbh5ta7eOCYfFE6FTsqrDw+rL9QOD8PXtwLK9dParlnAxVB6bZ6aHa4mRTYR13n94frVLGt9vK8fr8yKUS5k5MpaCmuSXrZdW+Wk4dEMNtU9N5f3VBS/He8f0iuOeMTO6at7ll35ePTiJS3/qYUXoVj180mKv+uwaLw9MyzS6USJ0Sj/fohZYFQRAEobfrlKBHfX3wUwW3283mzZv529/+xpNPPtkZh+xzvD4/xfU2pmS2XSStN4szacirPEbQY8eXgeKlMQO7ZlDtJZUF6nqIYqYnBYNGwV2n92dAnJFXft1LWaOdLLOBB2YMYHhyGLLDsh7qmp3c+9kW1h7Wtcjnh9xKK7VHqa/RYHOjPfBkWy6V8PwVwzAfRzcnn89PZZOjzfUvLc7jtatH8NTFQ5BIJCjlEnZXWLjvsy3MHp/CxSMSGN8vImSXpSi9kopGBwqZlC9um0CkTkWYVoFSLuWmKf1YtKsqqDOHTCpBJZO2qqFwULXFSV2zi0HxJp69dChfbirh7RX51Da7GJkczoPnDMRsVPLKrBGoFTIidUrCjqhz5PX5Q97wHlRlaftnIXQPpUxGYriWxBBFS8emRfLlbRNpsLnw+wNZUqGmdB3UYHOzp43ridfnp7LRwSuzhvPvX/bw2PmDWLG3ptX0riyzgZFJYXxy83ie+mEX20sbidarmDUumckZUfzzx92sPqwAq88PP2yvQCaVMCo5DJ/PT7XFic/vR6eSY9QoaLK7+XJTCf/4Lge/P9B69/IxSXx80zh8frA43OyusFBUFzyWJ7/fxVmDzPz7smH4/ZAapcXj9fPjjgqqrYFpKunROmaOSgz6O3O4IQkmvrx1Iv/6OZfLRyfy2YbikNtdPjoJtaLDatoLgiAIQo/UKUEPk6l1F4EzzzwTlUrFvffey8aNGzvjsH1KWYMdt9ff6mlxX5EQpmH1vhp8Pn/omiVeD+ycD/2mgaQXfSCLHgB5vwTmPEj6Vi0WobUovYqrxiZz+oAYPD4/Krk05JPXWqsrKOBxUE5ZE6f0j2J1G20vJ/SLwOb0cOWYJK6blEZqpPa4avxIpRLOHhzH99srQq4fmxbBpsIG3lqRH9QGFkAhlRCpV/GPCwdTUNuMx+vnu21l/LC9AoNKzquzRvL15hI+3VDC+9eNZcZby5nQL5I/nz2A9Ggdn986gV93V7Fgaxl6lZxrJ6SgPs56GjFGNTdPSeeSEYl4/X60SllLgCPxKI2/tCoZWWYDuZWWkOsnpkcd1/GFnsNsVHfY9U8igfOGJDA2LRKfz8+8m8axYk8Ny/fWoJBJuHx0EtlxRiL1KswmDW9dM/pABooLn9+PQiblmokpnD8sng9WF7C74tD77Ptt5fzhtP58ubmUt3/Lp67Zxdi0CB6YMQCv389j3+a0bOvy+vhwTSEfrinkmUuH8May/VQ1OXh25rBWY/55ZyU/76wkLUrH3ImpPLJgJ3+cnsl9Z/YnzqRhckYUcWFtt5zVqeSMTAnn9dkjqW5y8PC5A3nyh11B2VbTs81kmg1tThkSBEEQhL6iS6900dHR5Oa27l4gtJZfE6h3EX+Up1u9WUK4BofHR2mDnaSIECnLBcvBXgdpvaz+S8xA2PYJ1O6DqIzuHo3QRWKOcXMWqoggwKaieu48LYNovarlCe5BGoWMK8Ym8/POCrRKOWEaBaoTKMY5KiU8ZMFShUzCPWdksr2kgWcuHYrD7eXFRXmUNtjRq+ScMySOP36+hV92VuLzB8YxZ0IKX90+kaomJ0//uIvbpmXw8foSVu+vJTvOyJLcalbsreHr2ycxKMHE7PEpXDQ8AYfby6MLdnLhiHi0SlmrAAsEMkcOn7Ijk0qO+lQ/lCi9ikcuyGbWf9e2Wjc0wUhqVOhpEULfEKZRkB6tZ19162wPqQQGHuhOEnMgS6rJ7mZceiQGjQKdSkZCmIaiWhur9tWSFWvAbFQhAdQKGc/+nMveqsB+E8M13D89ix+2l7MwJ9AdxecPdJV58vtdLcdcsbeG/yzfd9RpYh+uKeKi4fG8sCiPKouDkclhbDrQ+vlwd5yazn8OTKl5buEeFt47lYyYQD2bNh8aHP6z0SppdnkpqbPx7twx7Chtwu72MjwpjD0VTSSEa47aDUcQBEEQ+oJOudJt2xZcIM/v91NeXs4///lPhg1r/URDaC2/phm5TEJUiKfGfUFieOAJ1Z5KS+igx875YIiDiPQuHlk7RWcBkkBdDxH0EA6I0Lddh+Oxb3fywQ1jeXlxHr/kVOL1+ZmQHsnNp/Tjrd/2cenIJPJrmnnk250khWu4bFTScd+oPH3JED5bX8xPOytwe/2MSg7nztMz+PfPuazYG2i9mRiu4amLB/PXr7bz2AWDuP/zrUE3X3a3lzeX78fv91NSb2dLcSNWhwe9Sk6N1YVRHSiA6Pb6eeL7HN6YMwqTRoleLSevysJ328tpcrj501lZLU+9jWo5F41IoH+MnuFJYcc1ZedYhiWE8fFN43js2xx2V1jQKGRcOSaJW6b2I7oD9i/0XFEGFf+aOZQr31yD64j6FH85e0DQdbTG6uSFhXv4aG1RyzKVXMpjFw5i8a4qFn5cyYzBZu6fnsV5/7ciqNBqSb2d+z/bwn+vGc2qvTUtU7ZCNVo2qBUU1LZdrLvG6uS8YfHklDfy7E+5vHTlcFbureHzjSXYXF76x+i5bVo6q/fXkncg6OL3w4aCOjQKKQu2lrGluJHhSSZmDI4jMVyDvI0gS0KYhutP6cfzv+SSW2FBKpWwtbieB2YMFAFBQRAE4aTQKUGP4cOHI5FI8B9RtW78+PG88847nXHIPmd/tZXYPtiu9qBInRKNQsaeSiunDzQHr/S6IWcBZJzR+6aIKPUQngLFa2H4rO4ejdCFvF4flRYnNpcHlVxGtEHVMq0jSq9iULyxVXcRgPRoPd9uLSMxXMM3d0yiqM7GjtJGHvt2J/+4cDD3f76VaouTflE6xqVFkF/bjMfvR6eU4fcH0thDFVBcvKuSJ3/YxflD43nhiuGYjWr2VVn565fbqGw6lFVSUm/nkQU7efua0Xj9hHzaDIEn0/++bBg/7qjA7fMhlQaKn76+rKZlm9X7A50gDnSl5re8wLrleTXEmjS8OWcUOeVNjEwO54PVBfywvZwYg5rbT01nQr/IkFODjpdOLWdCehQf3jgOm8uLXCohSq9EKRetak8GgxOM/Hj3Kby9Ip+NhfXEh6m5dVo6WWZDUIBwye6qoIAHgNPj46H5O3jrmtEs3lWJzeXji40lQQGPg3x++HR9MRcMj+fjdcVMy4omp6yR5y8fRqbZgNvrQwKs3FdNmFbJyr2hp64NTwrj5x0VXDepHw+fOwipBCb2i+SaCansrrBQ1mjnlV/3sr8mOHDicHu56YON5JQH/pb8vLOCFxfl8dGN4xiVEh5UNP5wyRFanrp4CPU2Nz6/H4Na3qo2jiAIgiD0VZ0S9MjPzw/6WiqVEh0djVotnrYdr73VVuJNbc/X7e0kEglJERr2hJqDn78MHA29b2rLQdEDoGhNd49C6EJ1zS6+3lLKy4vzaLC5UcqkzByVyB9O70+sSU2UXsUbs0dx64cbgwIfE9IjuXpcMnfM24TD7WNSRhQvL85jd4WFG09J47+/7afa4uTy0YlcOjKRD9cUthRFHJMazh2nZvDeynxum5bBsKSwliCLz+dn1b5aHG4f32wpo7jexp/PGsAP28sZGGdEJW+mqM7WMo6CWhsymYTC6rafTNvdXiSSQEFVs0GNUa3AqFEETZ9RK6RBcUqj+tAl5rMNxeSUNXLvmZlc9956vL5AULzG6uLOeZuZPT6FP52V1arV74nqq9lxwtEp5TLSY/Q8ckE2VocHlULaqg1rtcXBK0tat8OFQMHTlftqGJcWyeB4IxsK227zmlPexJzxKYxPi+AvZw9ALpPw3C+5fLyuCJ1STm2zkzkTUhmeaOLDNYUtLWYPUsgkXDEmids+3ITd7eXflw3jwuHxKGRSam0uXlvauv3yQWnROvZUWsgyG4gLU1Pe4CC30sId8zbxzR2TiD3K5watSo5WTGURBEEQTkKdcvVLSUnpjN2eVPZWWZnQr28X30sM17KrPMQHux3zwZgA4WldP6iOEJMNe34CWx1oI7p7NEInc3t9fLGhmKd+3N2yzOX1MW9dEUV1zfzfVSMI16lIitDyylUj2F/TjNXpIdaoxunxsXp/DSaNAofbSY3VyRuzR5JXFcj0unHrBl67egQ+H3y5qZSkCC3vXDuGj9cV8UtOJfd8uoXnLx/OrLfWsuDOSQyKDxSRlkol9I8xYJ6oZnL/KNbur+XrLaXMmZCCXCqlrMFOmFbJk9/nUNYY6GxS1+w6am0S6YGAx42npLGn0sJTFw/hj59vDdrmstFJQUGHMwbG0GB3I5FIWL2vhunZsTz1w+6WgMfhPlxTyNyJqe0OeggnN5VchkofOrvH4/VT0dh2J5+KRgfhOgVxJg3xYW1vlxim4YyBZqZlRaNWyHhvZQFXjE5mbX4tdc1upg+KRSmTsre6mR/vPoU/fr6VVQeKFQ+KN3LXaf3572/7sbsD02MeW7CTCemRJIRpiNQr+dNZWdz0wYaWNrYHXTkmiX1VVt6YM4q8Sgv7a5oZlxZBf7OBFxbuobbZddSghyAIgiCcrDo06PHrr79y5513smbNGoxGY9C6xsZGJk6cyBtvvMEpp5zSkYftcywON5VNTuLD+nZmTHKElt/yqnF7fYcKvnlcsOtbyDy7901tOSgmO/Df4nWQdXb3jkXodFVNDl7+NfTT4xV7a6myOAnXBQIBKoWMT9YVMWdCCh+tLWJHaSMxRhV/PmsAJfU2+scYqLe7ufl/G/nP7FH8a+YwHpy/PSibQirZxxMXDcbl9bE0t5ptJQ0MTTDx4qI9vHD5CPQHsisuGRnPvHXF3Pj+hpbXfrC6kJHJYdw0pR+PLcjh2ZlDuevjzTTa3cQY1OhUcuJMaspD3BhOz44l2qDiijFJFNQ0c9uHG4Pa0PaL0nHb1HSU8sC5XNFo59fd1fy8swKvz88ZA82MSgnnH9/ltNr3QZuL6luKNApCR9MoZQxKMLKpsCHk+sHxJr7eUkqsUc3FIxKYv7k05HZXjUsmp7yJt37bzyMXZJMcqeX699e3dEb5bEMxaVE6nrhoMG6Pl7tOy+ChcwZSVG+jsMbGE9/nBJ3TFqeHOquThDANepWC9Bgd78wdw7y1RWwpbiBKr+Ky0YlMzohiX3Uz93yyBavT0/J6vUrOi1cOp4/OhhUEQRCEduvQXqAvvvgiN910U6uABwTa2N5yyy08//zzHXnIPmnfgRTzhKO0o+sLUiK0uL3+4Ir7+34FZyOk9eLAmN4M2kgoWtXdIxG6gMXpCboBOdL+w6aMmA0qrp2Yyo3vb2TB1jL21zSzZn8d93++9UD2h4oIrZK3rx1NptnAR2sKW3Vf8fnhkQU7mT0+kFG3tbiRDLOe0no7Ts+hIITVGShAeqRNRQ3sq7Ly/OXDkEklPH/5MC4YFkekXkmsSc0H149t1TVqdGo4t01L595Pt3DhKyvpbzbw1e0TuemUflw8IoH/zhnFvJvGEX/gb1ZFo53r39/A3xfsZFe5hT2VVl5buo8aa+guNgcdb2tbQfg9wrRKHpwxMOS6CJ2S9Bg9uyssrM2vw+Pz8ffzslHJD31Mkkkl3DY1HbVcSkK4hr+ePQAJEh4/MOXscPk1zXyyrggkEiqanNjdXvx+SInSkh59KLCnUcgYftjUNICkcB0Z0Xqum5jKI+dnc+8Z/ZmaGY1UEih8fOTfG6vTw6MLdnZY69kai5PciiY2F9VTUNuM1ek+9osEQRAEoQfr0EyPrVu38swzz7S5fvr06fz73//uyEP2SXsqLEgITP/oy5IjA99fTlkTA2IPBMp2fAFhKYF/vZVEAjGDoGBld49E6AJquQyJhFY3PQdFHta5pd7u5rFvc1p1mAB4a0U+l41O4r5Pt3D/9CyK6mws3FUZcp9ur5+iWhtJERoGxhmYlBHFqORwPlhdyIBYA2PTIvh8Q3GbY/5wTRFGjYK/f7OT0SnhPHfZsJaihv3NBj68cRy7yi3UNjuJM2nYX21l1n/XtGR2lDbYGZcWyUPntg5wA6zZX0dOiJoEm4sbGJsazrqC1vUSZFIJQxNNbY5ZEDrCgDgj/71mNH//ZkdLRtPI5HD+cHoG//g2h0idkrkTU+kfo+enHRW8MmsEDTY3bq+fOJOaxbsr0SjlzHx9Fc9eOgSL00uI2VoA/LSzgrvP6M8T3+VQ2+wCQKeUcfcZmaRF6jCbVPSL1rOpsJ5le6qRSSVEG1QY1AriwjTEHfHgI6essVUQ9KCSejvNRwm+Hq/91VZu/XAjeyoDDyOkErhyTDL3ntlfdEESBEEQeq0ODXpUVlaiULQ9H1sul1NdXd2Rh+yTcistxJrULWnifZVWKSfWqGZHaROXjARczbD7exh0Se+d2nKQeRCs/2/ge1Lquns0QieK1Cs5Y2AMC3OqWq2L0isxqBTUWJ1E6VU02twt7SeP5PfDirwaLh6ZwBvL9nHtxNQ2b6YgMA0uOVzL2YNjuf699dRYXS3rLhuVgO2wqSehXqs58GR5Q2E9f/h0M+9cO6ale4rT4+O+z7ZgUMtpsLnxHDEQq6Ptmyur082nbQRcPl5bxEtXjuDOjzfRYAt+evzkxYNDdqERhI6kV8k5M9vM0EQTjXY3cqkEnUqO0+3ljdmj0KvlLZ3T7j4jk/+tLuDnnRX4fIGMpwuHJ/DIgp34/PC/NUWckW1u81juAzVEDgY8AJpdXp76YRff3zWZf/+yh2d+ym1Z98T3u3j43IFcPjoJY4jaNqGCpUHH8xx9/bFUNDqY/dbaljo/EMgsm7euiAidkj+cniG6IQmCIAi9UofeVSckJLB9+/Y212/bto24uLiOPGSftLuiicTwvj215aCUSC07ShsDX+z+Adw26DetW8fUIcyDwecJ1PUQ+jSDWsGjFwxmYJwhaHm4VsEzlw7lzo838+T3u6htdh4zlieRBFrYrs2vo9riJDWy7WyvgXFG/nBGBvd9tjUo4AGwNr+eieltF0KemBHF1pLGlq+3FjcGTT0xaRRolDJqrK5WAQ+JJFC/46Amu5v8aivbSxsprG3G7fHRVry2ttnFeyvz+fqOSfz9/GymZUUze1wyP959CucNiUPbQen5gnAsZqOaTLOBftF6zEY1yZE6MmMNxIdpWlrFJ4RpuH96Jm9dM5rbTk3H7fVz4/sbyD/QRraozsbI5PA2jzEg1tDSWvZwQxJMrNxXy5Lc1oHSJ77fRXG9rdVygEidKmi6zeFUcikRuva1oC2oaQ4KeBzu3ZX5VFmOPj1NEARBEHqqDv2Eec455/D3v/+dGTNmtGpPa7fbeeSRRzjvvPM68pB9jt/vZ1e5hWlZ0d09lC7RL0rHN1vL8Pr8yLZ9EigCaojt7mG1X1gSqE1QsALST+3u0QgdzOn2UmVxsr20kQabixHJ4fz3mtHsrbSyo6wJs1GFRiHjnz/uZl+1lX3VVq6dmEJyhJbBCUZ2lLa+EZJJJaRE6lrm63+4ppC7Tu/P/Z9tbbXtqVnRpEfrsDg97A2ROVJUZ8OgltMvSsf+muA2tCq5lNnjkrlj3uag5TVWF1kH/t9sVPPwudmturMAXD02mcgDGRllDXYenr+dX3MDGXxyqYTLRyfy0DnZnPt/K1p1aZFLJVwxNokYg4q5E1K5emwycqkEmaxvZ7UJvZdCFgj+/fXLQw90pBK4/dQMRiSFIZHAuH4RrN1fF/Q6iQTuOzOTJ77f1Wqf5w2N4+N1RW0ec97aIh6/cDASCTg8XhRSKXKZFIVUwnWTUnljWetaPddNSj1UEPx3CqqvdYRmlxeHu32ZJIIgCILQXTr0k+bDDz9MXV0dmZmZPPvss3zzzTcsWLCAZ555hqysLOrq6njooYeOe3+vv/46Q4cOxWg0YjQamTBhAj/++GPL+rlz5yKRSIL+jR8/viO/pS5XbXFS1+wiJeLkmBLRL1qPzeVlX35+oIhp+mndPaSOIZEGsj3yl3f3SIQO5nB7WZ5Xw+nPLeP2jzbx4PwdzHjpNx5dsBOb28tXm0p46odd3Pnx5qCpLPM3lxGhU/HMpUPRKluniP/l7Cy+3lzaMu0kr8rKb3tqeOWqES21LiJ0Sv44PZNnLh1KeowBj7ft+S+PfpvDm9eM4rpJqWiVgbojUzKj+OTm8RTV2dAcMYbow1rNyqQSzsyO4Z25o+l/oJtKrFHN4xcO5p4zMjGqFdRanfzh480tAQ8Aj8/PvHXF/G91IX88s3/Q/k8fGMP7149l7f46bvnfRp5fmEtZowNxGyX0dOFHZFA8efEQ8qubueH9DVz37nqunZDKjaektbRbHpkcxtvXjKbW6qSornXWhkEtp97marX8oMomBzVWJx+tLeLmDzby4Pzt7ChtxOL0oFcpeOT8bJIjAllgyRFaHjk/G4NaccxCwcdyeIHVI+lV8pa/TYIgCILQ23RopofZbGbVqlXcdtttPPDAA/gPVPaTSCScddZZvPbaa5jNbc9/PVJiYiL//Oc/ycjIAOD999/nwgsvZPPmzQwaNAiAs88+m3fffbflNUpl+9I7u9vBVNiUo6S19yX9onVIgC3rlpAplUPq5O4eUseJHRqo6+G0gkq04ewryhsd3PrhxlZZDIt2VZFpNhBrUrfKrgDw+gK391lmAz/efQqfrS9m9f46EsI03HBKGtF6JW8u30+/aB0T+kWyen8tX28pZV1+LVeOTeaWKemoFVLGpkZgOHBzZdIoUMqkIef6N9rdeH1+HpgxgLkTU2m0u1m2p5rbPtxIUoSWxy8czA/by1mwtYwRyWFEGoL/dpo0Sk4bYGZoYhgujw+ZVEKMQYXkwBydGquTDYWtC5ICfLaxmJ/vmUJqlJ4P1hTSL0rHlMxornlnXcvP7be8Gv77Wz4f3TiO0akRJ/hbEISuE6VXthTgHZxgpNHu5vvt5UCg/s0d8zZxSkYUfztvIIPiTfyys4KHv9nBM5cORSGT4D4iOJlTbmF8WgQ/7QxdqHhMagTPLdzDGQPNqORSPttQwmcbSvj7ednsrbKSW9nEdZNSMRvVVDY5+GxDMbvKLZzSv+0pbccjNUpLQpiG0obWxVJvmJxKjLF3f74SBEEQTl4dPoE6JSWFH374gfr6evbu3Yvf76d///6Eh7c977Ut559/ftDXTz75JK+//jpr1qxpCXqoVCpiY/vAdIgDdpQ2olPKTpqCflqlnKQIDZv3FHN52iRQ9qHgQNywQF2PwlWQOb27RyN0kIU5Fa0CHgd9sr6Yv5w9gFX7alutu2h4AgBymZSUSB33npnJLS4vKrkU1YEnqB/fNJ57P93KfdMzkUph5d5ayhodPL9wD6cNiOHJiwe3BDwg8MR49vhk3llZ0Op4l49ObHmi/NKiPL7aXNqyrqLJyYbCjfx75jBUcin3nNGfSF3ovzlR+tDLy9uY+w+BAo4Ot5cZQ+KYkhlFXbObc//vt1Y/N6fHx92fbOGr2ydiNorOEELPFKFT8eKVI/jj51s5dUAM89YGT03x+2F5Xg3L82qYMz6FojobZQ0O/rt8P+/MHcN9n26l+kAWhl4lJ9Os48Jh8fy6u7pVwDLWqCY1SsfTP+7mi40lvHzlCHIrLRTX2Xni+xzenTuGa98t5bFvc4JeZ1TL2zxXj1esScOHN47j9o82sqvcAgSyvmaNTWL2+FQUMpHpIQiCIPROnVY1Ljw8nDFjxnTY/rxeL59//jnNzc1MmDChZfnSpUuJiYkhLCyMqVOn8uSTTxITE9Nhx+1q20sbSY3SIe3t3UtOQH+9i/X18dD/+LOAegVjAuiiYf9SEfToQ9pqGQlQ1+zCpGn9Z3V6tpnUqOApa3KZFKMmeIZhf7OBd68bQ32zk0fPH4TP76fZ5SVMoyBSr0SjkFHZ5EAqkRCpUxKlV3HxyEQi9SreW1VAtcVJpE7JnAkpnD7ATJRBxb6q5qCAx0F+P7ywaA/zbhxHwu9oj320GyyphJaipDqVgryqZprsoTu+lDbYqWt2dWvQw+P1UdfsAglEapWixojQSnyYhldnjaS+ORBEbEu11dkyzWVrSSOZZj0L7pxEpcWBy+PH7fXxn2X7+HFHJa9ePZLXl+5jU1E9MqmEMwaauXpcMn/9chsAXp+fV5bkMXtcCk//uBufP/D3Jy1K11JM9aCnLxlKTIiHJTanB4vTg1wqaenOdDRpUTr+d8M4aq0u7G4v4VoFUXoVOpUoMiwIgiD0Xj3+KrZ9+3YmTJiAw+FAr9czf/58srOzAZgxYwaXXXYZKSkp5Ofn87e//Y3TTjuNjRs3olK1fXF3Op04nYfmvjY1tS4q2F22lTQyKuXEs2J6syzHFhb7h1Bv0tOnvnOJJJDtsW9xd4+kx+jJ597xOiUjig9WF4ZcNzjBSHa8kTtPzeDnnRUY1HJumJzG2LSI434KG21Qhcz0Kqqz8fLivfy4vRylXMrs8SmcNzSO9CgdWqWUgbFGfH4/UgkkRWiJM6lRymRsLW5o81gl9XYsDg9FdTYidAr0qrZbjh8pxqAiI0YfspDq2YNig75fzzFabbaVOdMVShvsfLquqCUwdNmoRC4bnUR8WN/qoNUXzr3uFq5TIpNKGJEcxm95NSG3mZweSaReRUFtMxcMjcfngwVby5i3rgi3x8eMIXHcdXp/XlqUx6KcCp66eDBWp4fKJicr99Zw64cbg9pN7yq38IfTDwUlfcCbc0bxwsI95FZayTLruW1aOv2i9cgPC9a5PF4Kam288ute1uyvJUqv4tap6UzMiDzm36IovardWSOCIAiC0JP0+KBHVlYWW7ZsoaGhgS+//JJrr72WZcuWkZ2dzRVXXNGy3eDBgxk9ejQpKSl8//33XHLJJW3u8+mnn+axxx7riuGfkCqLg/JGBxlHKSbW51jKGFi/FBjC2govZ6f1sSes8SNh7yJoLAVTQnePptv11HPvRAxONJEYrgmZ8fHwudkkR+i454z+XDcpFZlUQpi2/fPgi+psXPTqykA2wgFPfL+LrzeX8ta1Y0iPNhBn1GBze9EqZUGtX4/1hLakwc7tr65keraZB88ZSFKEliqLgxqLkwabG7NRTaRe2er7iDGqefva0dz0wQb2VB4KfEzOiOLv52ejVx86rtmoRiWX4vS0Dn6EaRXtbrX5e5U12LniP6uDfpcvLMrjy02lfHrzeOL6UOCjL5x7Xc3l9VLd5KSi0YHX7yfOpCHaoOLPZ2Wxcm8NR8bqovRKzCY1f/9mJ/eekcn4fhFc9956dldYWrZ5e0U+C7aU8fbc0cxbW8Sfv9jGjCGxPPNTbsgxSCUg4VDm5/i0CPqbDfz7smGB810hQxviHN9dYWHm66tbps9UWZz84ZPNzByZyEPnDmxVnFUQBEEQ+rIef4epVCrJyMhg9OjRPP300wwbNoyXXnop5LZxcXGkpKSQl9d26inAAw88QGNjY8u/4uLizhj6CdtS1ABARsxJFPTY+Q1RSjdmLawuDZ3+3qvFjwh0ctm7qLtH0iP01HPvRMSZNMy7aTzTs81ID9yLJEdoeXfuGAYnBLqsyGVSIvWqDgl4uDxe3lmRHxTwOGhHWRPbShoA0KoCc/oPD3gADIo3omxjusbYtAi2lTTi9fn5cUcFc95eS36NlaveXMM5L69g1ltrOf35Zdz36VYqm1rX8EiJ1PHRjeP58e5T+OjGcSy8dwr/d9UIYk3BwYJog4oHzhkYcgz/uGBQyLT8zubz+fluW1nI4FVRnY2FuypbinH3BX3h3OtKNpeHX3dVMf2F5Vz6xmou/88aznh+GZ9vKCYhPPA34GC3E4kEJmVE8txlw3j8u12UNzr4z/J9rC+oDwp4HFRtdfL5hhJqrE62lTYyOMHU8rfkSKcNiGHVvkBWyVVjk1qywFrO9xABj7pmFw9/vSNkgeMvNpVQZWm7Ho8gCIIg9EU9PtPjSH6/PyhF93C1tbUUFxcTFxd31H2oVKqjTn/pLpuKGojQKbvtqWeXs9dB3s+QNpXBzXKWlfTBoIfKANEDAt/nqGu7ezTdrqeeeycqOULL85cPp67Zidvrx6CWE9NJNSnqbW5+ONApIpTPNhRzalYMCnnowEa0QcWzlw3lvk+3BD2ZjtQpuWNaOvd+trVlWUGtjU2FDTQ5gs/FX3OrePan3Tx+4eBWN1ltTcc5nFoh4+IRCWTG6Hlh0R4Ka21kmvXce2YmmWZDUFp+V2mwu/hmS1mb67/aVMoFw+I7JHDVE/SVc6+rFNfZue2jTRwe93J6fPztm51kmg2M7xfJJzePp7LJQWmDnY2F9fzhky002t1AIKC4YGvb768luVXMnZjKyr21RBtU/PPSofz5i21B28Qa1Vw3KY0XFu3hlVkjmNAv8rjej00ON9tKGttcv3JvLVmxxmPuRxAEQRD6ih4d9HjwwQeZMWMGSUlJWCwWPvnkE5YuXcpPP/2E1Wrl0Ucf5dJLLyUuLo6CggIefPBBoqKiuPjii7t76L/L+oI6Ms36lpaQfd62z0Aqh5SJDKuRsrjIQ1GTj2Rjj09AOjGJY2D7Z+B2gEJ0qOgr9Gp50BSOzvSvmUNpdnlRyKQU1TXzn2X7qbIEgr8quZSj/ckwqBVMy4zihz+cwvzNpZTW2xmcYCI9RscjC3a2yiDZVtpIerSOaktwcPmbLWXcfXp/kn9nQUOTRsHEjCgGxRtxeHxoFDKMmuOvIdLRpBIJiqMEW5RyKbK2Hr8LfZrb6+WDVQVBAQ+9KlCfZ2iiCavTw74qK9EGFSvyavjnT7tb7cPj9aOQtf3+UcqkZMcZ+fmeU4gP05AYrmV4UhhfbiqhtN7OGQPNjEoJR62Q8uac0Sf0MEQqCWSftJWopFaILiyCIAjCyaVHBz0qKyuZM2cO5eXlmEwmhg4dyk8//cSZZ56J3W5n+/btfPDBBzQ0NBAXF8epp57Kp59+isFg6O6hnzCH28u2kgZmjU3p7qF0DUsZ5P4IGaeDQsPgaD8KKSwscHPD0D72NDJxLGx6H/KXiy4uwgmpsTh5fek+/remsKXYZ6ZZz79mDuWhr3dQUm9nzvjUY2ZKhGlVhGlVPBBnpM7q5KYPNvDPnxpCbhutV7HW5m613OMLdJJpL5NWiande2m/MK2SayaksKWNQq/XTkjBoO6+oIzQfZxuP3urD9WpMajkvHzVCN5cvp+XFh+aPjs928y9Z2bywqI9rerVLNtTzZ/PzmLRrqqQx5g1Lpnx/SKRHgis6WVSMs0GHpgxEL/f366HH+FaJdMyo1mSW91qnUQCE9Ijfve+BUEQBKE36tFBj7fffrvNdRqNhp9//rkLR9O5NhXW4/b6GRjX+wI2v8v6twNTP1ImAqCRSxgcJeXnfE/fC3qEJQfa1+7+VgQ9hOPm9vr4eF0R760qCFq+p9LKX77czl9nDODX3ZVkmE+sBpBBo2BCeiQbD9QQOpxCJiE73si/f2ldVFEll6LvY20rJ2VEMTY1gnUFdcHL06MYkypuDE9WaoWUoYkm1uYH3he3Tkvn1SV72VBYH7TdLzmVSKVw3cRU3li+P2hdlcVJtEHNjEGx/LizImjdoHgj5w2Nbwl4HKm92Z4GtYK/nZfN1pLVrTK5HjpnINGiM4sgCIJwkulbn2B7sVX7ajGq5SRFaI+9cW9XtBqK1sDwWSA7lLI7Ll7Of7a4KLP6iNf3oSkuEgkkT4Bd38K5L4BMnHbCsVVbnLx5xI3UQRVNDkwaBY+cP4jIE7yBUcikXDMhlW0ljSw/rO2mSi7ljdmjWLSrMuTr5kxIIcbYt26WzEY1r8wawfbSRj5aW4REArPHpTAowUiMQUxFO1nJZVKuGpvM+6sKcXl9DIwz8q+fQ3dX+XlnJd/fNZn3VhfgcB/K9hjfL4IBZgOPXzSYqyek8MHqAlweH5ePTmJkSjixnVQD6KB+0XoW3DmJhTmV/Lq7ilijmjkTUkiN1KEXGUyCIAjCSUbcffUQy/OqyY43Iu3r9TycTbD61UBxT/PgoFVjYmW8K4P5e9zcMbJv3VyROhl2fAH5ywJTegThGOxuLxZn28V9C+uaOXVAzO/ad4xRzYtXjKCiyc7mogYi9UoGxZswG1VkmvUU1jazYm8tEMj+mDUumVum9EMlb7sWQLXFSUm9jdwKC3FhavrHGIg1qtt8mt1TxBjVnG5UMzkjCiQc9XsUTh6JERo+vHEsf/p8G0321tO9DvL7we31s+i+qeSUNVFtcTI00URcmIaoAwHJyQYVY1Mj8Pn9WBxu9lU1s2R3FenROpIjdMSaOicAkhiuZe7EVK4cm4RCKu2WgsGCIAiC0BOIoEcPUNfsYntJIzdP6dfdQ+lkfljxAnhdMOgSjqy+qFVIGB8v46NdLm4ZrkTew2+WTkhEemCKy44vRdBDOC5quQyDSt5m4CMjun2trSP0SiL0SrLjgytsJIRreWXWSGqbXdhdXoxqOVGG1q1wD1fWYOemDzaws6ypZVmYVsGHN4xjULyxVxRnVonijsJhlDIZY9Mi+ezWCdRYQneMg8BlzKiWkxiuJTG87UxNpVzK/mors99aS1njoZaxSREa/nf9OFKjdB06/kPjk6BRiI96giAIwslNhP17gGV7qvADw5LCunsonWv751C8DgbPBHXo2iVnpSkos/r5fl8fa18rkUDaFMj5BlzN3T0aoReIMaq48ZS00OsMKtLbGfQ4mjCtkvRoPYMTTCRH6o4a8LA63Tz+XU5QwAOgwebmmnfWUXHYDZ4g9DZmo5q4MDWjU8JDrp+ebT6uKWbVFic3fbAxKOABgda4d8zbRK217cCKIAiCIAjtI4IePcDCnEoyYvSEa4+/JV2vU7QKNn4A6adBzIA2N0szSRkZI+W59Q6c3jb67fVW6aeDyxqo7SH0OXVWF7vKm/h5ZwWbCuvbfbOvkEm5elwKc8YnB7VOTY/W8/FN44kL07R3yB2i1uri5yMKNR5U1+yisNbWxSMSTlZ1zU5yKw4/B+0dst8InYqXrhrBuLTgwMdpA6J59IJBx9V6udbqZN9hHWEOt7OsidojCo4KgiAIgtBxRM5jN7O7vPy6u4oLhyd091A6T+VOWPYsxA45rqkdV2Ur+esyB69scnL/mD5UTNAQC7HDYMO7MOzK7h6N0IHKG+3c/ckW1uUf6gISb1Lz/vVj6W/+/R2Zogwq/jpjIDee0o/6ZhcapYwInYpoQ8+peeN0+/AdJT5ZZRGZHkLnK2+0c/9nW1m1r7ZlWZxJzXvXjSUrtv1d0RLCNLwxezS1zU4sDg8mjYJIvRKT5vgeVtiO0e7Z3gHtoAVBEARBCE1kenSzX3dX4XD7GJ8W2d1D6Rx1+2DxY2BKhiEzQXLst1yiQcolmQpe2eRiUUHbBeR6pawZULwmEAgS+oRmp4cnv98VFPAAKGt0dMj0Dp1KTkqkjuHJ4WTFGntUwANAp5ZjVLcdP89oR9BHEI6HzenhXz/nBgU8AMobHcx5ey3lHZTxEa5TkhFjYERyOP2i9ccd8ACI0CmPLGPVQi6VEKYVHVUEQRAEobOIoEc3+3pLKenRnVe9vVs1FsEvD4MmHEbMBtnxf6i7qL+cMbEyblto57t9fSjwkTwetFGw5rXuHonQQWqsTn7cEXp6R3mjg9KGjrnh6qnMBhV3n5EZct34fhGY+1ibW6HnqWl2smBLWch1VRYnxXXdP8UqUq/kspGJIdddPT6Z6BNsPS0IgiAIwvETQY9uVGN1smR3FZMzort7KB2vqRR+ehAUOhh1HShOLKgjlUi4a5SSsXEy7lxk59GVDhyePlDjQyqHgefDts+gKfSHdKF3sbu9eI8yv6OyqW9P75DLpFw8Ip5/XDCI8ANPq5UyKVeMSeLFK4YTqRM3c0Lncrh9eI5yDpb3gGK6BrWCP52dxW1T+6FVBjoF6VVy7j49gztP7Y9WJWYbC4IgCEJnEVfZbvT5hhIkEpicEdXdQ+lY1gr4+cFAZsfo60HZdhu/o5FLJdwxQklGuIePclysKvXwxnQN/cJ6eWvJzBmw4wv47Xk499/dPRqhnfQqORqFDLs79Jz8lMjf9/7vTSJ0Kq4en8IZ2WZsLg8quYwog1K0yhS6hE4pQ6+SY22jvXO/TmoHe6KiDWruPTOTq8en4HB7UStkmI0qFLJefk0TBEEQhB5OZHp0E4/Xx/9WFzC+XyT6o8yH73VstYEMDz8w+gZQta+tpkQi4ew0BU+cosbq9nPR/GY2VPTydrZKLWRfDBvfg7r93T0aoZ1iDCpunhK6teyI5DDMxj44dS0EmVRCfJiGjBgDSRFaEfAQukyMUcWtU/uFXDc0wUicqWd0OgJQymUkhmvJiDGQGK4VAQ9BEARB6AIi6NFNfthRQVmjgxmD47p7KB3H0Qg/PwReJ4y5HtTGDtt1slHK45PVJBqkzPnexubKXh74yL4g8PP55W/dPRKhDS6PD4vDjdfrO+p2SrmMORNS+cNpGWgUgRsYqQRmDI7l1VkjiRJz9QWhUylkMq4am8y9Z2S2TB2RSGB6tpk35owmqhuK/3p9fiwON06P6MoiCIIgCN1N4vf7+0ChhPZpamrCZDLR2NiI0dhxN+pt8fn8nPPybyjlUh6YMbDTj9clXM2BKS2WChh7M+g7p06J0+Pn6bVOqm1+vr1ER4KhF8ft8pfD8mfhyo9hwDndPZpu0dXn3vGwONwU1tp4Z2U+xXU2xqZFcNmoJBLDNchlbb/fXB4vVU1OLE4PWqWMSJ2qb2VxCX1KTzz32svl8VJlCbSU1ShkROqVGNRd2xXF5/NTXG9j/uZSVu6tIc6k4YbJaaRGaU+o24sgCIIgCB1HBD3o+g9/320r4855m3nk/GwGxPaBD5teJ/zy90B72jE3gbFzs1eanH4e/s1BnF7CFxfqUMra6APY0/n98OvjUJ8Pt60CfUx3j6jL9bQbL7vLw9dbynjgq+1ByzUKGZ/fOoHBCaZuGpkgdKyedu71FbkVFi59fVWr+iJ/O28gV45JRicKlgqCIAhCl+vFj8l7J6fHy7M/5TI8KayPBDzc8OtTULMHRl7b6QEPAKNKwt2jleys8fHCBmenH6/TSCQw8S7weeDz68Dj6u4RnfSqrS7+9vWOVsvtbi9/+mIrtdZe/H4TBKFTNdhcPDB/W8iCqk9+v4sa8fdDEARBELqFCHp0sbd+y6ek3sbV45K7eyjt5/PCb/+G8i0wfDaEp3TZodPDZMzMUvCfrS429ubCpppwmPoXKF4D39wR+JkK3Sa3oqnN1pe7yi3U29xdPCJBEHqLBpubTYUNIdf5/LCpqL5rByQIgiAIAiCCHl0qv6aZlxfnMWNwHInhvbyNpd8LK16AwlUw7EqI7t/lQzg/XU56mJT7l9pxeHrxLC3zYDjl/kAb269uAreju0d00jpGzVLEbEBBENriO8bfB49X/P0QBEEQhO4ggh5dxOP1cf9nWwjTKpg5KrG7h9M+Pi/89jzkLwsEPMyDumUYMqmEW4YpKbX4e/c0F4DUUwIZH7u+hXemQ83e7h7RSWlArAFpGyViUiO1mDRdWxRREITew6RRMDDO0Ob6kSnhXTgaQRAEQRAOEkGPLvLCoj1sKW7g9mkZqA+0teyVvE5Y+hQULA8EPGKHdOtwEgxSZmYp+O+2Xj7NBSBlEsx4Fpqr4fUJsOhRsNV196hOKlEGFfecntlquVwq4ZlLhxJjVHfDqARB6A0i9SqevngIyhBdnm4+pR/Ron21IAiCIHQL0b2Fzq9i/9OOcm79cBNXjkniwuEJHb7/LuNohMWPQ91eGD4Logd094gA8Pr8/GOVE7vHz48z9eiVvbSby0EeB2z/HHK+BokUBl0KQy6FlMkg71stD3tiB4n6Zhc7yxp5ZcleyhocjEgO4/ZpGaRGalH15oClIBymJ557fYHL46WozsYby/azvqCOaL2KO07NYFiSiQidCHoIgiAIQncQQQ8698PfhoI6rn5rLSOSw/jDaf2RSHrpDXnNHljyJHicMGIOhCV194iCVDb7eGC5gzNS5Lx8uqb3/pwP52iE3B9h32KwlINSB8kTIGk8JIyEuOGgi+zuUbZLT77xarK7cXp86FQytErRZlLoW3ryudcX2F0erE4vSpkEk7ZvBasFQRAEobcRn+Q70YaCOq59Zx0ZMXpum5rRO2/EfV7Y+RVs/hAMcTD6BtCEdfeoWjHrpNw8TMlLG11kR7m4bXgfeKKmNgWmEA29Aur2Q9lGqNgJK18ElzWwjTE+EPyIHxkIhCSODrxOaDejqN8hCMLvpFHK0YhgqSAIgiD0COKK3EkWbC3jT59vJSNGz/1nZqGU98LyKRXbYd2bUF8AqZMh40yQ9dy3zPh4OUVNPp5Z68SolHB1dh95uiaRQGR64N8QwO8LZH7U7oO6fYH/5i8/EAiRQOxgSJ0C/aYG6oSo9N39HQiCIAiCIAiCIHSLnnsH20vVNbt46occvthYyuSMKG46pV/vCnj4vVCyEXZ+DRVbwZQE427tcdNZ2nJZlgKbBx76zUGpxcd9Y1TI22rH0VtJpGBMCPxLmxJY5vdBUxlU5UDljkBNkDWvglQBSWMh/TTodyrEDevRgStBEARBEARBEISOJGp60P65zX6/n71VVr7YWMJHa4vw+/1cPT6FaZnRvWNKi70eqndB6WYoWh342pQUuKE2ZwdusnsRv9/Pt/s8fLrbTf8wKfeOVnF6ihyFrBf8LjqK3x8IgpRtgvLNgawdtx1UBkgcC8njA1NizIPAEBvIJukGoq6AIHQPce4JgiAIgnCyEEEPoLGxkbCwMIqLi4/64a/W6uKTTeU02t00OTxUW1wU1NmptrpatsmI0nJWdhQGVTc/Tfc6kJVuQOKxg8+LxOcBrwuJxwFuGxKXBYmjMVCz4zA+XTS+8HT82qhuGnjHKWhW8EmxCa8/cEMfpvDS3+AiVu0m2+jiquRGVLKT5O3v8yKr34e0aieymt2B98WRm2ij8Guj8Wsj8atM+JU6kGvwy1UgU+CXqXEPuRK/PvaYhzMYDMcV8Dvec08QhOMjzj1B6B7He+4JgiAIXU8EPYCSkhKSko49fSNi+h0YRszoghEJXeE9xT+ZJtvW3cPoVT7Z4eaqL1sHTI50vE+Pj/fcEwTh+IhzTxC6h8iaEgRB6LlE0APw+XyUlZWdcJS+qamJpKSkPv+kTHyffUtXfJ/Hey61de71tt9Fbxsv9L4xi/Een/aee31Zb3sPdQXxMwnWnp/HyXQuCYIg9DaioiEglUpJTEz83a83Go0nxYcF8X32LT3h+zzWudcTxngiett4ofeNWYy3Y7T3uteb9dTfSXcSP5Ng4uchCILQt/SuCpWCIAiCIAiCIAiCIAjHSQQ9BEEQBEEQBEEQBEHok0TQox1UKhWPPPIIKpWqu4fSqcT32bf0hu+zN4zxcL1tvND7xizGK7SX+J20Jn4mwcTPQxAEoW8ShUwFQRAEQRAEQRAEQeiTRKaHIAiCIAiCIAiCIAh9kgh6CIIgCIIgCIIgCILQJ4mghyAIgiAIgiAIgiAIfZIIegiCIAiCIAiCIAiC0CeJoAfg9/tpampC1HQVhK4lzj1B6B7i3BMEQRAE4WQhgh6AxWLBZDJhsVi6eyiCcFIR554gdA9x7gmCIAiCcLIQQQ9BEARBEARBEARBEPokEfQQBEEQBEEQBEEQBKFPEkEPQRAEQRAEQRAEQRD6JBH0EARBEARBEARBEAShT5J39wAE4Xg12l002Nx4fX6MGgVRelV3D0kQBEEQOp24/gmCIAjC7yeCHkKP5/f72VfdzMNfb2fN/joAMs16nrp4CEMSTKgUsm4eoSAIgiB0jr1VVnH9EwRBEIR2ENNbhB6vpN7OzDdWtXzgA9hTaeXKN9eQX9PcjSMTBEEQhM5TUm8T1z9BEARBaCcR9BB6vIW7KmmwuVst9/j8vLgoj2anpxtGJQiCIAid65cccf0TBEEQhPYSQQ+hR3O4vSzPrW5z/cbCeqziQ58gdCq/34/b6+vuYQjCSeW4rn8Ocf0TBEEQhGMRNT2EHk0hlZAQrmlzfbRBhUIm6cIRCcLJw+318X+/7uX9VQU02t2M7xfJExcNJiNG391DE4Q+T1z/BEEQBKFjiEwPoUeTyaRcPS6lzfW3TUsnQieq2AtCR3O4vcx9dx2vLdnLxPRIrpuYSnGdjYtfW0luhaW7hycIfZ5MJuXq8ce4/okuLoIgCIJwTCLoIfR4SREanrx4MNIjHmjNGpvEhPTI7hmUIPRhfr+fv3y5jfX59fx1xgCumZDK9EGx/OPCQYRrldz20UbsLm93D1MQ+ryk8NDXv6vE9U8QBEEQjpuY3iL0eAa1gouHJzA5I4rNRQ3Y3V5Gp4QTbVARplV29/AEoc+Zv7mUb7aUceepGQyKN7Us1yrl/OG0/vz1q228uyqf26dldOMoBaHvE9c/QRAEQWg/EfQQegWtSk6KSk5KpK67hyIIfVpds4tHv93J5IwoJmVEtVqfEK7hjIFmXluyj6vHpmDSKrphlIJw8hDXP0EQBEFoHzG95SRXb3Oxr9rKrvImyhrseDqgQ4PL46Wk3sau8ibya5ppsrdutycIQs/0/MJcvF4/c45SS+DC4fE43F4+31jchSMThL7F4/FRWNvMtpIGdpY1Utpg7+4hCYIgCEKfJDI9TmL5NVbu/3wrmwobADBq5Pz17AGcMyTud6fN1lqdzFtbxOvL9mE7MOd/SmY0T108mMRwbUcNXRCETpBf08zHa4u5cmwSRk3bGRxhWiXj+0Xy/qoCrp+UhvTIggOCIBxVfbOTlXtreey7HKotTgAyzXr+NXMYg+KMyOXimZQgCIIgdBRxVT1JlTXYueI/a1oCHgBNdg8Pzt/Bqn21v2ufHq+PLzeW8NzCPS0BD4Dle6q5/r31VFkc7R22IAid6MVFewjTKpieHXvMbc8YaKa43s7a/LouGJkg9C17Kq3c+fHmloDHwWVXv7WWwnpbN45MEARBEPoeEfToBbxeH1VNDiqbHLg8HdMxYWdpI1WHfdgC0KvkzB6fgs3loeQ4P3Q5PV4qGh1UWRzU2Vy8snRvyO32VFopqRepu4LQUxXUNPPt1jIuGBaP8jieMmea9UQbVHy7rawLRicIfUe1xckLi/aEXGd1evhhe/kx99Fgc1HRaKfB5sLl8VHZ5KCqyYHX5w/artnpoaLRQY3V2caeBEEQBKHvE9NberjyBjufbyzh0/XFuL0+zh8Wx9yJaSRFtG+qyMai+qCvT+kfxQ2T0/hwTSFPby8nUq/k1qnpnNI/mmiDqtXr/X4/hXU2/rtsP7/kVKJWSpk1NplnLx3GHz/fitXpafWaPZUWRiaHt2vcgiB0jjd/249BrWBaVsxxbS+RSJjQL5Lvt5Xz2AWDUMhEDF0QjofN5WFXuaXN9VuKG7C7PGiUrT+iNdpd5JRZeG5hLvurm+kXpeOmKf3YXtLI11tKmTU2mYtHJhCpU5Jf08yLi/JYl19HhC5wTZ+SGfqaLgiCIAh9mfiU2oOVN9iZ/fZanl+4h9IGO1UWJ2+vKOCS11ZRXNe+9Nf0aH3L/yeEaZg9PoWbPtjAol1V1Da72FNp5b7PtvLUD7uob3a1en1hrY0LXlnBR+uKqLY6Ka6z88xPubyyJI9HL8gOeUxR00MQeqZaq5MvNpRw9qDY48ryOGhsWgSNdjcbCuqPvbEgCAAoZFISwzVtrk+J0KIKcR66PF5+2F7BVf9dw4aCeuqaXWworOeW/21Er5KTEaPn2Z9zuf699RTV2Tjv/1bw444Kaptd5FUFani1dU0XBEEQhL5MBD16sFX7a9lX3dxqebXVySfrinC3o9PK+H6RqBWBX//V45J5eXEebq+/1XbzN5dSeUQtDofby+tL99Jkb53NsaO0CbfXT0JY8Ae6SJ2S9CjRbk8QeqKP1hYhkcDpA48vy+OgtCgd4VoFS3KrOmlkgtD3xIdpuGVKesh1UglcNjoJqbT1x7Mqi5PHv8sJ+bqXf83jyjHJAOwqt7Auv67VdRhCX9MFQRAEoa/r0UGPp59+mjFjxmAwGIiJieGiiy4iNzc3aJuvvvqKs846i6ioKCQSCVu2bOmewXYwm8vDV5tK2lz/7bZyGmy/vxVsnEnNRzeMI0yrIC1ax86ypja3XX1EYdMGm5ufcyrb3H75nmpGpRyaxmI2qvjoxnHEhfgAJghC93J5fHywuoDJGVEY1G13bAlFKpEwLDGMRbva/nsgCEJrY1LDuXlKPw5vfKRVyvi/q0aGDFYA1FldQUXCD2dzefFz6MHFol1VjEmNCLntqr2/r1i5IAiCIPRWPbqmx7Jly7jjjjsYM2YMHo+Hhx56iOnTp5OTk4NOF8gaaG5uZtKkSVx22WXcdNNN3TzijiOVgEoua3O9Si5F0o4ukXKZlOHJ4fzwh1NocriRSMDfOtEDAI0ieBwSCSiPMn9fq5Rx35mZXDAsngi9kniTmliTCHgIQk/0884Kaqwuzhp07I4toQxPDmPpnmpKG+xt3qwJghAsLkzDrVP6cfnoRPZWWVErZKREaIk1qUPW8gCO2RpaetiHApVciquNbNAjr+mCIAiC0Nf16KDHTz/9FPT1u+++S0xMDBs3bmTKlCkAzJkzB4CCgoKuHl6nUivkXDsxhV93h04bv2ZCCpE6ZbuOIZNKiA/TYHDImZYZzZLc6lbbSCQwrl9k0LIonZKrxibz0uK8kPudNS6FhHAtCaKGhyD0eP9bXcjAOMPvLo6cHWdEQiAjbOaoxI4dnCD0YRF6FRF6FRkxhuPaPlKvJFqvojpEJ5Zogwqr49CU0wuGx/PQ/B2ttpNIYHx6ZKvlgiAIgtCX9ejpLUdqbGwEICIidMrm8XI6nTQ1NQX964kGxZmYEeLp6/CkMM7MjkXSnlSPwxjUCv52XjZR+tZBlEfOH0TMEZXeZTIpV45NItOsb7X9lWOSSIsUwQ4htN5y7p0s8iotrCuo48yB5t+9D4NaQWqUllX7ajpwZEJHE+de72c2qPm/WSNaZVoqZVL+fl42b6/IB+Di4fEMjDMSKjHkkfMHie4tgiAIwklH4ve3NamhZ/H7/Vx44YXU19fz22+/tVpfUFBAWloamzdvZvjw4Ufd16OPPspjjz3WanljYyNGo7GjhtwhaixO8qosfLimEJfXzxWjkxiSaMJsVHf4sUrr7SzJrWLxririTGquHp9McrgWgyb0PP+KRjubihr4clMJOqWcORNS6BetI1InPlAJofWmc+9k8OiCnczfXMorV41A3o6Ws/9bU8imwnpWP3BahwVjhY4lzr2+weX1Ulrv4KtNJWwvbWRwvIkZQ2L5YVs5hXU25oxPIT1GT5RedcLXdEEQBEHoq3pN0OOOO+7g+++/Z8WKFSQmtk6hPpGgh9PpxOk8lB7a1NREUlJSj/7w5/H68Pk5oXaSv5fL40MmlSA7xvzhg9weHxIpyENUmxeEw/XGc6+vcri9jHlyEadmxXDV2OR27WtDYR3P/bKHFX85VbSm7qHEude3+P1+XF4fSpkUiURy1OvwiV7TBUEQBKGv6dE1PQ666667WLBgAcuXLw8Z8DhRKpUKlap3ZSO05ynsiTrRwIqiCwIxfYK1GrxOkMpBb6ZdlWh7qd547vVVP+4ox+LwcGrWibWpDSXTHKhJsLGwXgQ9eihx7vUtEokkqNj50a7Dx3VNd1rB0QBIQBsJio7PJhUEQRCE7tKjgx5+v5+77rqL+fPns3TpUtLS0rp7SIJw4uwNULIeFv4dqnLAlAin/BEGnAf66O4enXCS+mRdMYPijcSa2n9zY1QrSAjTsKGgnguHJ3TA6ARB6BI+L9Ttg8VPQO73IFPAsKtg8r0Q1r4MMEEQBEHoKXp00OOOO+5g3rx5fPPNNxgMBioqKgAwmUxoNIHWiHV1dRQVFVFWVgZAbm4uALGxscTG/r4WjILQYXxeyP0Rvr710LLGEvjuHqjcAac/AmqRWi50rf3VVtbm13HnqRkdts+MGD0bCus6bH+CIHSB+kL472ngtAS+9nlgwzuQtxCu/ykQpBcEQRCEXq5Hz0t4/fXXaWxsZNq0acTFxbX8+/TTT1u2WbBgASNGjODcc88F4Morr2TEiBG88cYb3TXsPsPr8+Hx+dq1D4/Xh8/XK8rGnDivB47187GUwy8Phl634W1obt0mWBA622cbStCr5IxJbV8nrMNlmg3kVlhodnqOvbEgCC06+jrp9/txeY7j2u1xwOpXDgU8DtdYDHt/7bAxHXss7q47liAIgnDS6dGZHsdTY3Xu3LnMnTu38wdzEqm2ONlTaWHe2kJ8frhybBID44zEGI4/Db680c6GgnoWbC3DpFZw9fhk0qJ0hGlbt8XtdRrLoGQdbPsUVCYYcz1EZoA2xA2kvQFsbTz99vuhdi9EpnfqcAXhcG6vj883FjMxPbJDCyOnR+vw+WFnWRNj0zoumCIIfVVFo4OtJQ18ubEErVLG1ePb1wHN4nBTUm/n47VFlDbaOTUrhmlZ0W3X2bE3QN7Pbe9w51cwZCYoO6lOj9cTCK7s/BqK10BMNgyfBaYkUVNEEARB6FA9OughdL1qi4O/frmdxburWpb9uKOCiekRvHDFiONqlVvWYOfqt9aSX9PcsuyLTSXcMqUft01L792Bj8ZS+N9FULPn0LJtH8PYm2HaA60DH7JjfK9KfYcPURCOZmluNbVWF6cOaH8B08MlhmtRyaVsLW4QQQ9BOIbyRjs3vLeenPJDWRZfbynjijFJ/PnsrBMOfNhcHr7bWs4D87e3LFu8q4pInZLPb51Av+gQ1xqJDFSGtneqCQsU3u4sFdvgvXPBbQt8vecnWPUyXPUp9JsGMvERVRAEQegYPXp6i9D1Nhc1BAU8Dlq1r45V+2qO+XqXx8vbK/YHBTwO+s/y/ZTW2ztknN3C64b1bwUHPA5a9ybU57dero2E+BGh96c2iUJxQpf7dH0R/aJ0pEbqOnS/MqmE1CgdW0saOnS/gtDXeH1+vtxYEhTwOOjT9cXkV7e+fh5LlcXJQ19vb7W8ttnFY9/m0GQPMX1EHw3j72h7p2NvAXknPaSwVMKXNxwKeBzk88CX14O1vHOOKwiCIJyURNBDaGF1uHl3VUGb699dWUCDzXXUfdQ1u/h0fUmb6+dvKf29w+t+zdWw6f2212/+sPUyXSRc/J9A8ONwMiVc8REY4jp2jIJwFFVNDpbsrmZqVud0DeoXpWNLcUOn7FsQ+oJqi5P9NVbmrS1qc5sP1xSecI2PjQX1tPWS5XnV1Ld17e5/JqSf0Xr5uFshKvOExnBCbLVQtz/0OkcjNJV13rEFQRCEk47IHRQA8Pn8VDQ5cLq9bW7jcHvxHuODmN8PTk/b++jVRQ79fvA4214fqhgcQHQW3LwMilZD4UqIHgCZZ4MpQaTvCl3qi00lyKQSJqVHdcr+06J0/LijgkabG5NW0SnHEITeqqjOxs0fbGD2+BScRyk0anN58fn9SJEc977tR7l2+/1HqbltiIWLXw8EIHZ8BXJVoI6HKSl0naqO4jvGZwHP0R+wCIIgCMKJEHdcAgDljQ4eWbCDqVkxbCpqCLnNBcPij1mPw6CRc/oAMz/trGhzH72WJgyyzoHtn4VeP+zKtl8blhT4N/TyThmaIByL3+/n03XFjE2LQKfqnD/9aVGBKTM7yxuZ2EmBFUHojWqtTm7/aCO7KyysL6hjalY0X20Knfk4c1QictmJJeIerRPTgFgDRs1Rznl9TOBf8vgTOma7aCNAEw72+tbrZIrA9VIQBEEQOoiY3nISqGpyUFDTTGm9rc0sjJ1ljazcW0d2nIGkCE2r9WajiotHJCCTHv3Jk16l4I9nZaFVylqtG5MaTnpMLy7cqdTB1L+ELvyWMBrMg7t+TIJwnNbm11FYZ+vwAqaHizdpUMml5JQ1ddoxBKE3qrG62FEaOC9+3F7BRcMTQgYisuMMDE0MO+H9m40qZo5MaLVcJpXwxEWDidSrsDrdFNfZKKxtpq75KFmLXcEQB+c8F3rdtAdB13l/pwRBEISTj8j06MOa7G7W7K/lie93UVRnQyWXcvmYJG6flk6cKTiwUVAbKJz24PwdPHvpUJbtqebHHeX4/DA928wNk9NIaKvt3RFSI7V8d9dkXl2ylyW51ehUMuZOTOO8oXEn1Pa2R4roBzcvhRUvQu4PgUDImJsC6cCG2O4enSC06ZN1RcSZ1AyMPUq3hnaSSiWkRGrZKYIeghDk8KmdLq+Px7/L4ZWrRvLNllKW5lajVsi4elwyF49MINZ04tfJMK2Sv84YyIT0KF5fto8aq5PRKeHcd2YW6dE6CmubefL7XSzaVYnPD0MTTfzjwkFkxxlRyls/pOh0Ulmgnsh1P8Gvj0NVDoSnBrqgJY7pvDa5giAIwklJ4vf7T6xa1jHk5uby8ccf89tvv1FQUIDNZiM6OpoRI0Zw1llncemll6JS/b4e9J2lqakJk8lEY2MjRqOxu4fTYX7eWcEt/9vYavmI5DDenDOaaMOh38PKvTVc/dZaAKQSmJYVw9TMaCQS2FbcwP1nZbUKlByL3e2h0eZGKpUQrVchkRz//OQez2UHR32g5Z8uGqQiaer36KvnXk/TaHMz9qlFXDIigQuGt34a3JHeXpFPYW0zC++b2qnHEdpHnHtda3+1ldOeWxa0TCmTMmNILGNSI5iUEUlyhO6Y2ZTHo8bixOPzoVPJMagVlNbbuPi1VVRZgrM75FIJ3901mQFx3fz7tzcEurjIVa2LfguCIAhCB+iwO7XNmzdz5plnMmzYMJYvX86YMWO45557ePzxx5k9ezZ+v5+HHnqI+Ph4nnnmGZzObk6t7OMqmxw8/l1OyHWbixooqQ9uE5cerScxPBDU8Pnh191VPLJgJ3//ZicjUyIw/44MDY1CTqxJQ4xB3bcCHgBKDRjjwWAWAQ+hx5u/uQSPz8+UzM7p2nK41Egt+6ubcRylsKIgnGwi9SqmZ5uDlrm8Pr7ZUsaGwjqi9eoOCXgARBlUxJo0GNSBYsK/7a1pFfAA8Pj8vLAoD6szRDvbrqQJC1xPRcBDEARB6CQdNr3loosu4k9/+hOffvopERFtF9RavXo1L7zwAs899xwPPvhgRx1eOILN5aGk3t7m+s1FDYxIDm/5OtakZt6N4/jjF1tZlx8oLKZTyrjr9P6cPdiMtIM+jAmC0LX8fj8frS1iVEr4MQsRd4TkCC1ev5+9VVYGJ5g6/XiC0BuYNAoev2gwOpWMBVvL8fr8yKUSLh2ZwP3Ts9CrO2e2scvjY1FOZZvr1+bXYnV40atEtyVBEASh7+qwq2xeXh5K5bE/UE+YMIEJEybgcol2ZJ1JIZWikElwe0PPXooxtp5ilByp4805o6ltduF0ezFqFJiNKhSybpjvKwhCh9hYWE9elZUHZgzokuMlRQTm4u8qbxJBD0E4jNmo5omLhnDPGZk0Oz3oVXKi9Cq0ndRNCQJTWGKNbWdqRuqUHZZhIgiCIAg9VYddaY8n4NGe7YUTE2lQctHwBD7fWNJqnUouZVgb1eHDtMoueRosCELX+HBNIbFGdZcFINQKGbFGNbsrLF1yPEHoTXQqeae1jA5FKpUwa1wyH64tCrn+linpQfW9BEEQBKEv6rQr77p161i6dClVVVX4fL6gdc8//3xnHVY4QKOQc++ZmewsaySn/NDNh0ou5a1rR2M2tf9DToPNRbXFyeaiBrQqGUMTwog2KNEoO+5tVWd1UmlxsrWkgXCtkuw4I2ajqnuqzbtsYK2E0k3gsQcqzOtjQBN+7NcKQjeoa3bx/fZyLhuVhLQL6+okRWjYVS46uAjC0bRcQ4sb0ChkDE00EWNQtbqGOm0WpM1VeEs2gtuBLHkMXl0MasPx1cBICtfy6AWDeOzbnRxeuv7cIbGd2sJaEARBEHqKTgl6PPXUUzz88MNkZWVhNpuDilj2uYKWPVh8mIb3rh9LYa2NjYX1xBrVjEoJx2xSoWznlJUai5Nnf9rNZ4dlksilEv512VCmD4xF1wHzk6uaHDzw1XYW765qWaaSS3lzzijG94tEpejCwIfTCrsWwII7wXdYgcZRc+HUhwLBD0HoYT7fUAzA1C4oYHq45AgtS3Kru/SYgtCb1FicPLcwl4/XFbcsk0sl/PPSoZw9OBb9gWwQh7UBSc58FD/dj+Kwa49vxHU4pj6AOszcat9HMmgUzByVyNTMaFbtraHZ5WVyRiSxJg0ROpHZKQiCIPR9nRL0eOmll3jnnXeYO3duZ+xeOAExBjUxBjVjUtsuLvt7LN1TFRTwgEAl+Ps+28rPd5vIjDW0a/9er49P1xcHBTwAnB4fN7y/gcX3TyUlUteuY5yQhiL4+rbWyze+BymTYOjlXTcWQTgOXp+f/60pZFxaJEZN1xYpTArXUtfsosbqJEovUucF4Ui/7a0JCnhA4Br6x8+3MiTBSFZsoI2stLEI5Q/3tHq9avO7OFMmw/CZx3U8vUqOXiUnLaoLr5uCIAiC0EN0Sq9NqVTKpEmTOmPXQg9QbXHy+tJ9Idf5/fD5xuKQ605EldXJWyvyQ67z+Pws29OFT5F9vkBwoy0rnodm8VRb6FmW76mmpN7OWYOO/SS4oyUeKGa6p1LU9RCEI9VYnby2ZG+b6z9ZV4zf78frcSPZ8E6b26nWvIijsarN9YIgCIIgBHRK0OPee+/l1Vdf7YxdCz2A1+ejqsnZ5vqiOhueI+q4nPgx/DTa3W2uL6mztWv/J8TngYbCttdbK8Hb9lgFoTu8t6qAftE60qP1XX7sWKMauUzCHlHMVBBa8Xh9VFvbvoYW19vw+Pz4PG4UlqM8RLBWgld0whMEQRCEY+mU/uLTrAABAABJREFU6S1//OMfOffcc0lPTyc7OxuFIji1+quvvuqMwwpdRKuUMzw5jN/yakKun5oZg1zavniaWiEjy2wgt40nxeP7HV8Btw4hV0LGGbDnp9DrE8eCsutvLAWhLfurrSzbU82tU9O7pY6STCohIUxDbqW1y48tCD2dTiVnZFIYv7ZR92ZqZjQKmRRkWhwpp6Leuyjkdp74sUjU7ZtKKgiCIAgng07J9LjrrrtYsmQJmZmZREZGYjKZgv4JvY/F4aaqyYHF4caoUfDnswYgDXEvFalTMiUzqt3Hi9KrePi8gSHXJUVoyI7v4vdR5lmgDVEXRSoLFDJVG7t2PIJwFO+vKsColjOhK4ODR0gM05BbITq4CMKRDGoF90/PCnkNjdApgzqqyLLPC90hTCrDP/UvqLS//1podXharuuCIAiC0Jd1SqbHBx98wJdffsm5557bGbsXupDF4Sav0spLi/PYW2UlI0bP3af3JzVKy0c3juOh+TvYX9MMwKSMSP5x4WASw7UdcuzhSWG8OWcUj32bQ2mDHakETh8Yw9/PG0SsSd0hxzhuYclw/c/w3b1QsCKwLKo/nPciRGV07VgE4SiaHG4+31jC2YNiUco7Ja59XBLDtfywoxy/3y+6dgnCEfrF6Jl303gemr+dfdWBa+j4fhE8edGQoGuoPDwZ97U/IvvhfqRFKwMLo7Nwnf08RP6+a4/V4WZvlZWXF+exp8pKWpSOu0/vT5bZgKGLix4LgiAIQlfolKBHREQE6enpnbFroQu5vT4W5lRy32dbW5aVNthZtqeaF64YznlD4/j0lgk0OdzIpRLCtEpMHfiByaBWMH1QLMOSwrA4PChkEiJ1KvQd0A73d4nKhCs+Alsd+D2gNoG+64tECsLRfLa+GJfHxxnZ3fveTAjXYHF4qLY4iTF2cZBSEHo4jULG+H6RfHJz4Boqk0gI1ykwaYJbyEqkUhSxA3HM/ACpox6/z4dfbUIdFvu7juvx+liSW81dH29uWVZSb+e3vBqenTmUi4bHo5R3YTt4QRAEQegCnXL3+Oijj/LII4/w7rvvotV2zFN/oetVNTn5+zc7Q67729c7GJsaQUK4hmhD57akNBvVmHvK7BFNWOCfIPRAXp+fd1cWMCE9knCt8tgv6ESJYRoA8qqsIughCG2INqiO6xqqNkaBsf1TR6ssTh76envIdY8u2MnE9MgOy9YUBEEQhJ6iU4IeL7/8Mvv27cNsNpOamtqqkOmmTZuOaz9PP/00X331Fbt370aj0TBx4kSeeeYZsrKyWrbx+/089thjvPnmm9TX1zNu3DheffVVBg0a1KHfU19Q3mhnV1kTUqkEtULG6n21qBUypmVF02Bzs3JvDSNTwhgYayQuTEOt1YnV6Qm5L6vTQ2FtM59tKGZaVjRJEVqi9J0b/Oh0tjpoKjtUsDTzLDAmhK7lcTi/HxqLoWQDVOyA+GEQPxLCkjp/zIJwmIU5FZQ22Ll9Wvdn2sUY1ShkEvIqLUzKaP/NmiD0dl6vl8I6O5uKGsitsJAdZ2RYkol+HdVhyeMMXMPyl0FDMaROhugBYIxr2aTW6qLJHvq6bnN5qbY4g4IelU0O9ldbWbm3lliTmlP6R2E2qlErZNBQAuWboWwLxGRD0hgwJkI7C5kLgiAIQkfrlKDHRRdd1CH7WbZsGXfccQdjxozB4/Hw0EMPMX36dHJyctDpdAA8++yzPP/887z33ntkZmbyxBNPcOaZZ5Kbm4vBIKqaH1RQ08xV/13DPWdk8ktOBYt3VbWse+an3dwwOQ2lTMr1720g3qRm3k3jkYaqsnYYi9PDS4vzeGlxHlMzo/nXzKG994lucw0seQo2vH1o2a+Pw6jr4LSHQXeUm7bKnfDeueBoOLRMGwFzf4CY0MVYBaEzvPVbPgNjDR13E9UOMqmE+DANeVWig4sgAOwoszDnnbVBQYdwrYIPbxzHoPYW5/a4IH85fHxloM06wG//DtSemj2/JQh/rHiE7LDrflmDnRveX8+uckvQ+revHc2U8Dqk758buHYepDLC3O8gdiiIOj6CIAhCD9IpQY9HHnmkQ/bz00/BLULfffddYmJi2LhxI1OmTMHv9/Piiy/y0EMPcckllwDw/vvvYzabmTdvHrfcckuHjKO3a7C5+OMXWwnTKmhyuIMCHge9vSKf164eiUElp6zRwV++3Ma/LxtKhE5JXbOr1faROiU2l7fl62V7qvlheznXTkztnUULy7cGBzwO2vguDDwv0LI2FEsFfDIrOOABgayRz+bAtT+AISbkSwWhI20raWBDYT33nZHZ3UNpER+mYU8bbacF4WRSUNvMHfM2tcqyqLe5+cPHm3ln7hhSInW//wCWcvj06kMBj4Nq8mDRY3D+S6DSEaFTEm1QUW1xttpFmFbRkrHpcHv5v8V5QQEPCEyha6guQ7r4tuCAB4CzKRB0uXExGON///ciCIIgCB2s0ytCWq1WfD5f0DKj8fcVaGhsbAQChVIB8vPzqaioYPr06S3bqFQqpk6dyqpVq9oMejidTpzOQxf8pqa+3VaxrtnFhoJ67p+eyRcbStrc7qcdFZw6IIYFW8tYm18HfnjxiuFc/956PD5/y3ZyqYSHzh3Iuyvzg17/1op8zhkaR4yhF2R7WKvAVhtIB9aEQcHKtrdd+TJEZQW2l6tAGwX66MC65mpoKAzeXqaAUXMh/XRoKABXE+iiA4VPT3In27nXld7+LR+zUcWolBDtLbtJQpiGxbsqu3sYAuLc624NNhcl9faQ6/ZVN9Nob2fb2LJNgetZKDnzYeqfQBuF2RDBi1cM59p31gVd12VSCS9cPpyYA/VFaq1OvtpcyoVDorhphBajz4JPpmZtlYTRZgnU5sHYmwLXOa8LZEooXAXr3wpcX0XQQxAEQehBOiXokZ+fz5133snSpUtxOBwtyw+2LvR6vUd5dWh+v5/77ruPyZMnM3jwYAAqKioAMJuDuxSYzWYKCwtb7eOgp59+mscee+yEx9BbOT2BoJNeJafB3jpr46B6m4uEA8UHAawuL2PTIvjpnil8tLaQnLImMmL0TM828/7qQraVNAa9vsHmxnvYh6geye+H6l3w+XVQvTuwTKmDSffA1D/Dsmdbv8ZeB1vmwdKnAl9HD4CZ7wamrrgdwdvKFHDJW5DzdSADxO8LpPn2PxvOfQ5MCZ353fV4J9u511UqGh18t72cq8clH3NaWldKCNNQb3NT1+wiQte9hVVPduLc615219E/9zjcvqOuP6Yjsy4O5/MErncrXkB66duMSU3l53unMG9tETtKGxkQZ2D2+BSSwrXIZYH5Lx6fnyemx3G24wcM818GdyBgkxo3DE/ma3DFPNg6L/g6l3EGzHwHPI62xyIIgiAI3aBTgh5XX301AO+88w5ms7lDpjvceeedbNu2jRUrVrRad+T+DwZX2vLAAw9w3333tXzd1NREUlLfLTpp1CgwauTsLGtibGoE324rD7nd2LQIFuYEnsqGaRWYNArUChkZMXoeOmcgDo+X7SWNXPPOOtze1sGNiemRGLqrnezxaiyGd88Be/2hZa5mWPIknPNvMA8K1Og4XMpEKN1w6Ovq3fDeOXDL8kAGh1R+KKV41HWwc34g6HGQ3w97fgSPPRAsOVZh1D7sZDv3usqHawpRyqRMzYzu7qEEiT8QRN1XbSVCd/K+73sCce51r2iDCrlUEpRdcZBKLiWyvUHBxDFtrwtPDQRFyjbD++ejvGER6dHxPDBjAA6PF5VchkIWXOzDpJZzlnQthlXPBO+rfCvyeTPhotcD17qD/H7IWxh4EHDhK+37XgRBEAShg3XKHeq2bdvYuHFjUJeV9rjrrrtYsGABy5cvJzExsWV5bGygT31FRQVxcYeqk1dVVbXK/jicSqVCperlnUZOgNmg4oGzB7A2v5bbpmUwNSua5Xtq+GlHBS5v4OlSjEHFgFgjz/2yB4C/nj0A82FFSeUyKXqZlJRIHWajulWarkIm4f7pmehVwZ16epzitcEBj8OteQ3G3QI//uXQMrUJ+p8FH80M3tZeD0VrYMC5MP4OWPVSYHn6aYEnX6HsXxqYDnO0oIezOVAfRCINBFRkPTyIdIJOtnOvKzjcXj5aW8jUzGi0yp71fokzqZFKYG+VlTGpIujRncS5173CNApmj0/hvVUFrdbdMDmNCH07gx7GBEidAgXLW6875X5Y+5/A/zeVQs1uMMW3XNdDjtdb+//snXd4k1X7xz/ZSZume29o2atskL1B9nYgICgIiFtf9XWg6Kvi+rlQmaIoS0QcIMhG9t4FShfdeyfN+v3xdIUmhWor6/lcVy6a5zzPOSeh6XPyPff9veHA+/bHKkgRNhBc/ARfq6rE7amICqEgTdgQUDqB5tZJuxMRERERufuol7piHTp0IDEx8R/3Y7VamTNnDuvXr2f79u2Eh4fbtIeHh+Pn58fWrVsrjpWWlrJr1y66du36j8e/U5DLpHSN8EKrUnDfogO88cs5VHIpiye3J9zTiXtb+vPB+NbM++UsIR5OfH5/FINa+Nm4uJcT4Kbhh0c6M6ZtIAqZ0N4hzJ2fZt1DuNc/MGH7t7h61HFb9hXwaiIIDhKJULJ27FLY8l8hfLdaX4dBroH2U6H/G6D1FXKb7Z1bjqMQZIsZMi/CL4/D5x3hq+5Cqk1+cu1en8hdx6+nUsgpNjKguWOh92ahkEnx1am5LFZwEbnL8XJRM717OC/f2xTvMt8MP52aecObcX/HENyd/qHoofWGkZ9DlzmgKqtc591YiMi4egTSzlSem3rGfh9VMekFbw5HZF4ShJZrkUgBCRxbAcsGwWftYNUDkHgYSsW/AyIiIiIiN4d62RZcvHgxM2fOJCkpiRYtWqBQ2O7+t2rV6ob6mT17Nt9//z0///wzLi4uFR4erq6uaDQaJBIJTz75JG+//TaRkZFERkby9ttv4+TkxP33O9htvwtJyilhwlcHSM2vzLNde/QqO6MzWDuzC15aJYUGEz880hmVXHrdsrPBHk7MH9WCp/s3wmIFF7Uct3+6YPu38G3muE0XCJ4N4YmTwnOpHL4dVen9Ua2v5lCQDF/3FPw9+vwX3EJrHt9RlEf2Ffi6l5BqA8LicPe7EP0bPLAOdP72rxO561mxL47WQa74u2quf/JNIMBVI4oeIiJAkLsTUzqH0q+pD2aLFZlUQoi7BplM9s87L86Gra+DqRgm/wp5VwUj7T0fQNZlm1MtnpHX3/GSqYToDEeRke6htukt5XSdC/s+gRMrK4/F/wVL+8P96yDSQSU0ERERERGReqReRI+MjAxiYmKYOnVqxTGJRFJrI9OFCxcC0KtXL5vjy5YtY8qUKQA8//zzlJSUMGvWLHJycujUqRNbtmzBxcWlTl7L7Y7FYmXTmRQbwaOcjEIDG08mMbtXBFp17dJSNAo5ge63Vij9DRHeExROYCyu3tbzeXANEqI8QMhRHvQubH4eMqJtz1U4CX3t+QD0eUKqS8IBwRC1YV+I2Va9/4AocPaqftxQBDvfrRQ8qpJ2BlJOiKKHiF1OJuZyKimPZwfUTSphfRDgpuZYQu7NnoaIyC2BQiEj3Etb9x3nJ8PZH4WfLWYhymPfp9XPc/JA79kUp+v1p/WDro/Dtjeqt2ncwTNCSHGpilQGDXrDtyOqX2O1wu/PwMN/CGkxIiIiIiIi/yL1kt7y8MMPExUVxf79+7ly5QqxsbE2/94oVqvV7qNc8ABBTHn99ddJSUlBr9eza9euiuouIpCvN/KrA+NSgN9Pp5L7T0vl/U2yi0o5mZjLyz+d5unVJ9hzKYP0gnp2fdcFwuRfQOtTeUwqExZ3TYYKgkdpkSBybJ8Px7+FjjNgwsrKKA6tD0zeKAgf0b/b9n94keALEnpNepV/Gxj3jeDTcS2GPLi02fGcT60Gyz909he5I/n+YAJeWiVRwW43eyoOCXDTkJxbct3qFSIiIrUnPV/P0fhsCi9VMXm/vBV8mkHLsZUiPoBrMMkj1rIp4QYiS2QyiJoEbafY9qELhIc2Cqmgod1sr4kYIHiGOCInTtgkEBERERER+Zepl636+Ph4Nm7cSERERH10L1IL5DIpTkrHCxwnpcyud0d9k11k4IMtF1l5MKHi2PrjSXQIc+ez+9vamKjWKTI5BLaDR3ZBUZogcOgCwNkHVFrBgC16E6yfLuxMAZxZJ+xMPbgezCZBuND5Q1G6IHxUpbQI1j0M3Z6E3i8LfTh5CvnW9gQPACRC2VxDgf1mtRtI60WfFLmNKdAb+flkEkNbBdxSZWqvJcBNgxWIzSyiWYDuZk9HROSOIS1fz5yVx0gvNLCuqxMV8SNWK/w8Gzo+CvevgZIcSt0asDdNxbyNWczq5VNTt5UYSwSR/77VglihdILiHCFdxqMBTPhW8P0oyhDucy6+EL+/5j6lt7jZuYiIiIjIHUm9iB59+vTh5MmTouhxC6BVyXm4Wzj7YrLstj/cLfym+HHEpBfZCB7lHI7LYfOZVB7qElonpY7tIpGAa4DwuJbCNPhpRqXgUU5BKvz5GoxZCuqyL27OPsKicvN/bM8tLRSiRKZuhrAu15+Psw+0nyaUzbVH24eu34fIXcevp1IoNVnodYuVqb2WANfKsrWi6CEiUjdYrVY2nU7hcLzguZHnFYW3VCaktoBgqH3wSzj4JcbIoXyme4pP/kpEKoFO4Z7XH8BQJKS2nFknPJeryoy6rcI9dM5RwQPLyQNoUnmdX0uQKYVzryWs211dsl1ERERE5OZRL6LHsGHDeOqppzh9+jQtW7asZmQ6fPjw+hhWxAGtg9wY3MKPTWdsS8v1buxNxyplJAv0RrIKS0kr0KNVyvFyUdVLxIXRbGHF/jiH7d/si2NIS/8Kh/s6pShT2JUqzhIiL5y8wLnKAvDqUaHEnj0u/ylcVy56SCTQbCSc3QCJB2zPbTcVvCJvbE4yGUQ9CBd+hZSTtm1dHwf3sBvrR+SuYs2RRFoFueKpvbXLkGrVctw0CmIyRDNTEZG6ILvIQE6x0ab87WeHCni1///hseVxW9HexY+kDv9h+fdCJbB3xrTCR3fN3wxjiRCxUZAipHtq/QTR5NyGynNMhsqfrVa4sksQPa5F6wujF8G6qbaVzJy9YOhHoHG78RdakCZEVOrzhYgTZ+/aXS8iIiIiIlJGvYgeM2fOBOCNN6obYNXGyFSkbvB2UfHmyBY83C2ctYcTMVutjG8fTENvLV5lwkJGgZ4Ptlxk9ZHEivVSkLuGJZPb08jXpU6jLiwWKwV6B8ICUFRqwnJtpEVdkJsAa6dAUpWytaHdYPRXgoGpyQgl9iNiAGGhd60govOH8d9A+jk48YMQ/hs1CTzChXDfG0UXIIQQp56G02tA7Sr04xYi7oyJVONKRiHHE3KZ2+cGhbWbTICbWMFFRKQuSMop4YlVxxnaOoCiKj45G87moFFE8tjEHXjGrEdTdBVr5ADyfTux9FAx93cKYXTbIALcNDgpqyz9SnLh9FqhNLupzFNL7Sp4XznaAADBi8oeCjVEDoDZB+HUGsiKgYi+gvG3W/CNv9DMy7DqPqGUeznNRsLgd0UjVBERERGRWlMvoodFNF285fDSqvDSqugQVv0LtNFsYeXBBFYdtnViv5pTwn2LDvLL490IdKu7cpgqhYyRUQHsvJhht71/M1/cnOo477coE9ZOtRU8AOL3wsa5MHaJsPhzrWFR5hkBKjvh+S5+wqNBb1vDt9qi8xcekf3/WT8idzwbTiTjpJTRLtT9Zk/lhvB3VYuRHiIi/5Cc4lKeXnOCI/E5BLpr6B7pxfpjlcahP5zIYu0pCZ0bDOXJvhG0D/fEHZg33Op44yL1NPz+rO0xfR7EbAf/1tWjD8sJ7+V4okon8GoklHEvT4epDfnJ8N1oyI23PX5ug7CZMPAtUNyaJbpFRERERG5NbsOaoyKOyCsxUqA3IgHcnZQ4qW7svzejwMCSPbF227KLSolOzb9h0SO/xEihwYjBZEEmkeDmpESnqS5gdAr3JMzTibgs29KxOrWc6d0aoJLfgLt8bSjKhKQj9ttitkFhBlzcLPhxNLkXLvxme45EAkPeF4zaTKVCyK3FLBiQlpehrSuhQhQ8RGrAarWy4XgSHcM8UMpvD4PbADcNf13OxGKx3tKmqyIi9Y3VaiUt34DRbEEpk+KjU91wJGVWYSkHY7MB2HQ6lcWT23M0LoeJLXV0DVJgQcKFPDmebu54aFVkFhjwcrHfv9VqpSgvG+ed72B39CNLhXSU78dX+oSUEzkA3IKqX6PPK6vOIhEiFJXON3Y/K0gV0mdkCiG1JjehuuBRzvFv4Z4nwD30+v2KiIiIiIiUUWeix6pVq5g4ceINnZuYmEhCQgL33HNPXQ1/V2MyW7iUXsj8387x1+Us5FIJ97by55kBjQnxcLru9QaThQKD4zDWKxlF9GnisLliDvHZRWQXGVm6N5at59IwWax0a+jJf4c2I8JHi1xW+QXN303Dykc6s2JfHKuPJFJqsjCouR+P9428oTnXGn1uze2FaZByQthJGvZ/EBAFx74Vjge2g86zwKuxsAO1/ws4ulSo1OLbAgb9DwLaCtVfRETqmVNX80jILmZS59tn0R/gpkZvspCcV0KQez18vkVEbgOyCg1sOpPK/227REaBAT+dmqf6R9K/mR8eztc3FC80VJaXLzVb+OV4PFvu90Tx58tID+4GqZyWTUeS5Pc09y0+i7NKzitDmxEV4oaLunLzIbuwlK3nU/Gw5NA/54r9wXITIHYvTN8Of86DhH1ClEWXOdBitG01MrMJsi4JKTIx20AihabDoe+rQpUXRxRnw6WtsP1NyEsU+uzzqmCa6ghzKRiLHbeLiIiIiIjYoc62CRcuXEiTJk149913OX/+fLX2vLw8fv/9d+6//37atWtHdnZ2XQ191xOfVcyoL/7ir8uCH4XJYuXnE8mM+3IfSTkl171eLZfiXkM6SWM/l+v2kZBdTFxmMXO+P8amM6mYLIInx96YLEZ+8RfxWdUXKYFuGp4d2IjNT/Rg2zM9mT+qBeFezvWzE6x2c9wmkQhu896NhR2tn+cIZWu7zILhn0BwJzj0FVjNsGYK7P9UEDwA0s7AN8PgqoMoEhGROub30ym4ahQ09b99KqH4V1RwKbrJMxERuTkUG0ws2nOF/244Q0aBYAqamq/nhR9P8+2BePTG63ud6aoIF2qFlCfbKVAt7480brdw0GJCfnYdoRvH8vUIXy6lF/LQ0kMV0SEAeqOZbw/E8cKPpzmdbsTsWcOOhkQiVGMZtxwePwaPbIfOj1X31MiNh8V9BbNvq1W4j579CZYMEMQTe5hK4cT38NOjguABgsn45udr9rFSOAkRJCIiIiIiIrWgzkSPXbt28f7777N9+3ZatGiBTqcjMjKSli1bEhQUhKenJ9OmTSMsLIwzZ84wbNiwuhr6rqOk1IzeKERmlBhNfLHzMnpjdR+VtHwDuy9lUGo0U1zqOJLDR6dmdm/75YX9XdVE+NQcwaA3mvn1VDIX0wpIyzfYabfw5a4YSuzMQSGT4eeqxt/1GnO1qlitgshgqt73DSNXC0Zq9mh8r+BY798GVGUCj1QOmZeERdyRJXDPU4Kz/dWD9vvY/ILgfi8iUo9YrVZ+PZVC+1B3ZLdRmoi3VoVCJiFGNDMVuQ3RG02UlP4zA/bMQgOLHaSRfrHjcoUQUhWL2YyhOB9TqdDm5wzvDmtIpI+W8a298D72aaX5aFXyk/HPOkDzshLRb/xyjtQ8YQMko8DAFztjAPj2eDYZHZ6xP2G5SqgqJpWBxhVcAwWx49p0FZMe9n9euRFQlaIMOP9r9RLwAIWpsPPt6seNJYII4u1AjOn4qJACIyIiIiIiUgvq1NNj6NChDB06lKysLPbu3UtcXBwlJSV4eXkRFRVFVFQUUuntkYN+K5KWr+dIXA6rDycglUiY1CWUSB8tey5lOrxm85lU9KUm/ryQzuioILo29MT/Gn8OmVTCqKhA8kqMfL37CgaTIKA0D9Dx2f1RFbu0jijUm+jTxIf3Nkc7PGf3pQzy9SY0joQNR+QmwsVNcP4X0HhC55mCQVptK5oUZQjXylVweWuZuZpU8O9oNUFIW9nzPkxYKZTZS9gH8fuFcN7xK8CnBRxf4bj/jAuCHwg+tZuXiEgtOJucT1JuCZO7ht3sqdQKqVSCv6tGNDMVua3IKNBzOimfb/fHY7JYmNA+mPZh7vhd555oj8zC0ooIyGsxmCxkF5USXJbaabVYMGXHYznzE6r4XVg6zcAqkaI5uowJJgODu43DHNAB5QYHJqOAZ/wmogLacDY5n4TsYi5nFJGWb8BJKau4x+cUG1l4Vs4TQ77GY8cLUJIjXOwaJJSdvZFqK/o8IaXFERd+hbaTKjcUyinJsS+UAGyfD1N+g1+fgoT9wjGZAto9DF1mg/z6qUAiIiIiIiJVqRcjU09PT0aMGFEfXd+1pOaV8OiKI5xKyq84tvNiBh+Ma4Wbk4J0O7tEAK4aBccScvnrchZ/Xc6igZcz303vRMA1woenVsXs3hGMax9MXkkparkMT60SD+cacmvLKDAY+eFQIq52DEvLcdMokdd2Zzo7FpYNFiIsyjn3E3SdC92fBk0tKlfIVbDuYejwCLR/WMgLlinhyk7h+LjlwngA66cLxqcVY24QTEydvGruX1rHFWdERK5h67k0nFUymvpfP+XsVsPfVS2WrRW5bUgv0PP8ulPsjK6sMrbnUibNA3Qsmdy+1sKHWlHzho+qiimxKT0axTeDBWFg0DtIj30jGG2XobuyA6t3Uxj0jlDlxA5GtTf5BkFkkUklFOiNPLj4IM8NbMyEDsGsLqvW9s3RLC5kBfLMoI34yYvwc3NG6eItVBK7EaQK4V6cE2e/3cnL/r2xJt+O4iwh/WXi91CcKYgjajfQ+gqVYURERERERGqJGHZxm/Dn+XQbwaOcJXvjmNwlzOF1g1r4saPKou1KZhGrDiVgMldPh1ErZIR4ONEy0I1IX5cbEjyKDSY+3HKRjSeSGdTC8SLp0R4N8NRev78KSoth5/9sBY9y9n0C+XaO14SztxC1se8T+GEirHlI+Pfgl+ARLuQdt54IBz63FTzK+f1ZCOkkpL3Yo9XEyiouIiL1xB9nU2kT7Ib8NoyYC3DTcCVT9PQQuT04fTXPRvAo52xyPpvPpmG1l7JRA55alcMqaA29nfHQCtEL+oIsZJufEwQPrQ9o3GwEj3IkGech8RCE97DbZ3rTSWyLFny++jfzZc9F4b624I9ohrXyp+oexMG4PMb/kMj9v5aQ49LoxgUPEKIuuz7uuL3LY6BQ27nOS/ALsYfWt+ye7SFEdgZECfdpUfAQEREREfmb3H4r57uQ7KJSVh60X77tXEo+PjoVA5r5Vmub1i2c01fzKLymMsuqw4kk513f4PRGyCkxsulMKgUGE2eT85jWLbzaOYOa+9Kjkbedq2ugJBvOrnfcfn5j7frT+Qu7RmpX2+NOHjDySzjwBYR1E5zkHZF+TogIkV5TTtenGfR8HhS1D3kWEblRknNLuJBaQLuQWkQ43UIEuGnIKDBQoDde/2QRkXokv8RITEYhv51KZmd0OonZxZSaKj07SkrNfLvfQclUYOWBeLKKSms1pq9OzdcPtUOnsRXO3Z0UfPFAO3xcBGFAWpKLNG6P0Nigd/Xy6VU5+YOQ8nEN+V2e5+cEFUWlZhp4OXNfh2B+Op5U0R6TUUTba/6O6NRyvp7UDl+dHYHieoR1hxZjqh+/50mh6pk9nL1gzFLbKjAASi3ct6p2wouIiIiIiMh1qJf0FpG6xWK1YjQ73lVa9lccH45vw+zeEWw5l4pUIqFrQy82n01lyd6YauebLFaOJ+QilUj+cflIq9WKuWzH64udMUzpGsbSKR04FJtFqclCnyY+NPXX1S7KQ+gYLI7NV/+WqalvS5i5V9gdSzsrGJcGthUWXZN+FnbWrNUjYCrIiYNOM2DOEbi8TYg2adATvBuBi7hAE6lfdkZnIJVAyyC3mz2Vv0WAq/Bl6kpGEa2D3W7uZETuWjILDXy09SIrD1ZWFVHJpXx2f1u6R3qhVsiEe67F8b3AZLHWOtIDoKmfjt/ndud4Qi7nU/JpHqCjTbA7AW6VQoPVWsUwVSoDcw0iocUo3MMe2QHRm7EqNZgiBhGTryHtQhELxrZCKpXw1JqTlFSpDmOxWvn8gbYcicvmbHI+Tf11RIW4OYxEuS5aHxj8npB6Gr1JSB1tPBh0AUKkiiO8G8GjOyH5OCQdE8xLQzoLniLXGqaKiIiIiIj8A0TR4zbA3UnJyDYBvL/lot32MW2D8HNVYzRbOJ6Qi5NSRrCHE9/si7N7ft+mPmw7n86aw4l8/kBb3Jz+vimYTq2gR6Q3uy4KYcDL98Xx3YF4Wga6IpdJGN02sPaCBwgRGRH94dIW++1N/0b1H6kU3EKER8uxtm2eDQTRI7QbxO+1f31EP6FcnkcD6Nig9uOLiPwDdkSn09jPBa3q9vyzXVm2tlAUPURuGjsupNsIHiAYic787ihbn+pBA28tzio549oFV5SBv5YRbQLw+Bv3TalU2GgIcndiWOsAu+dYVK7g3xpSTgpm2l0ft5veAkDz0YKw4B4KgW2RAAqgoasR87kLvLc5mozC6hsE9zT0wlen5t5WAdzbyv48ao2zl/AIaFO761yDhMffuaeLiIiIiIjcIGJ6y22ATCphVNsgu7swkT5aOoS5E5NRSHZRKV5aFTujM7BaoW2IW7XzPZ2VDGsVQKnJzMSOIaTk6UnKKcFo/nvl+HQaBS8NaYqTsjLlw2SxcjwxF28XVYVharHBRGJ2MTEZhaTmlVx/l0ytg/5vgtK5elvTYYJwUddo3GHwu/bTVFqOEwSP7DgwFNT92CIiNVBqsvDX5Uxa3aZRHgAapQxPZ6VYwUXkppFRoOezHZfttpktVjaeTK543rmBJ038qhsG+7uqGdcuCJnsxpdPafl6LqUVcCE1n+TcmlNLC6Su6Ad+IERL5MQK96OAqOonan2E8q0yW5PQ4lITeSVG7usYwguDGldLY5nYIRgf3d/YiBAREREREbmNqZctQ7PZzPLly9m2bRvp6elYrgkT3b59e30Me0cT6KZhzYwurDuayPrjScgkEiZ0CKZrQ08mLT1EfFYxKrmUkVGB/N/EKF7+6TRvj25JSp6eVYcSKDGa6dXYhwHNfCnQGwlw0/CfH09RVGpGq5IzrVs4D3YOxdul9ouhht7O/Da3G4t2x7LrYgZuTgoe6d6AeyI88XBWkZxbwnt/XODXkymYLFZ8dSr+M7gpvRt71xxl4hUJM/bA/s8Erw2NG3R5HBr2qj/TUO8mwph//Z9Q2cXJEzrNFNo+ay/822gQDPwfeITVzxxERK7heEIOxaVmWt/GogeAv5uamHTRzFTk5mAyW0nN0ztsr1pdyM9VzbKpHfjtVAo/HErAaLYyMiqQCe2DCLzBtFCLxcKF1EJe/fkMR+KFcrCNfV2YN7w5LYN0OKuqVzUpNJi4XOBKl4f/QLLvE+FeNPAtrOnnkRz/VkjtbD4a2k0RIjyqkJxbwoI/LvBL2b3W20XF3D4R9GnizZZzaTzWsyEdwj3+UXSniIiIiIjI7YjE+ncSU6/DnDlzWL58Offeey/+/v5IrsnN/Oijj+p6yH9Efn4+rq6u5OXlodPpbvZ0asRstpBdbMRiFXZ+n15zqto5PRt5ExXixsd/XmLdzC4cuJKNQibhaHwOvjoVxaVmfjyWVO26SZ1D+M/gpjj/zfB5vdFMfokRuUyKh7OwqMoo0DN52WHOJVevPPN/E9swok3g9Ts26kGfK5S9c/b8W3OrNcYS0OdBYQb8+DBkXpNa5BYCUzeD6w3MX8Qht9Nn72bywZZolu+L48sH2yG9jXPdl/0VS2xmEVuf7nmzp3LXczd+9nKLS3n4m8Mci8+12/7e2FaMbx9sc8xisZJVZAAkuDspkNciwiM2o5ARX/xFfomtP5VcKuHn2ffQPNC12jWZBXrkZ9fhtu05aDFaEOL1eZARjaVhH4yhvVB5BFWL8MgoMDB12SHO2LnXfji+NX2b+OAqih0iIiIiIncp9RLpsWrVKtasWcOQIUPqo/u7GplMireLiqs5xfx3w1m75+y6mMHkrmFIJcLPp6/mcTwxl2Gt/BndNogxC/fZve77Q4lM797ARvTIKzFiMltw1VRf7JWazBToTSjlUlzUCtQKGWqFjGKDicxCA04KGfHZxXYFD4B3Nl2gU7gnfq7XcYtXqEHhV/M5dY1CI5TNjf5N8P/ITYSzP0Fp2U5gboKQcy2KHiL/ArsvZdA8QHdbCx4gVHDZEZ2OyWyp1ZdHEZG6wM1JyYuDmzLuy/3V2jyclXRtWF1Ul0olyKVSrFhr9fkzGE38fialmuABQgroZzsu879RLXC7pjS8pyUbyV/zwVgMx7+zncv5jcimbsXsGYbBYKLYaMZJIcNJIcFalMnwxhqMZivRabYpmO9tjqZrQ0+qSywiIiIiIiJ3B/UieiiVSiIiIuqja5EyCvQmiksd+3Ck5pXgqlHw49GrfDutE4k5xazYH096gQGLg9ges8VKbrGRUE9h1+hIXDaL98ZSqDfRv5kvEzoEE+zhhMlsqehv76VMvFxUzOzZgKb+OtLzDSzcGUN0WgG9G3njXUP5u5Q8PcWlNVRouZkYCoV86pSTkH1FSLUZsxhOr4UzPwrnXNkFTURhT6R+KdAbOX01j4ftlIO+3QhwE76UXc0pIczLjl+PiEg909Rfx6KH2vPKhjOk5gupLm1D3Xh3TKtq1cxS8/TsupjOdwcSMFusjG4byJCW/hVeVfYwmoT74/GEHPbF2DdCBTiRmEu+3lRN9JAYi6AgxeF1hoQjXJY24qvdMWQXlfLBAE9UCb/ic/ZHHpUrGd1xChec2jF9fSJ6o5BanJqvr3G9ICIiIiIicqdTL6LHM888w//93//x2WefVUttEakb1ApZje2RPi6sebQLLmo5uy5l8MKPpwG4r2PNBqAahYysQgOvbjjDprOpFcej0wpYeTCejbO7UVhqYvQX+ypK4F1KL8RssTKhfTDPrjtJecJUscHEjJ4NHY6lkktRym/B3V5zqeCW/+O0ymMZF+DCrzDsE8hLFMreuteDmaqIyDUcjsvGYoXm/rf/Pm152dqYjEJR9BC5KWhVcvo19aFl4D3k643IpRLcnZW4X5P6kZqvZ+Z3RzmRmFtx7Nxv+azYH8+qRzs7FD7Op+Yz7sv99G/mi7fWsejv7aJCYS/aSaYUHuZSu9dJXAOZ+d1R0vL1/DE1nMCfRkHe1Yp2r6RjdA7uyhfD3+XhHxOBsnutGFklIiIiInIXU2eix+jRo22eb9++nU2bNtG8eXMUCtvc0/Xr19fVsHctns5Kukd6sedSZrU2HxcVQR4a/F01JGYX8+rPlWkwidnFRPpouZRevYJCEz8XPJyVxGUV2wge5eQUGzl5NZfdlzIIcNPYVGGY1DmUlzecpqpDTHKengA3NRqFrEIgqcrYdkF4/Z1ytvVNQRr8+iTIVeDVGCxGQfTQBQmRHp1nQ9o5aDr8Zs9U5C7gwJVsPJ2V+N4BFRfcnZWoFVJiMgrp29T3Zk9H5C5FIpHg56quMbXyaFy2jeBRTkJ2MRuOJzGjZ0NkUttNnewiAy+uP43BZGH7hXQ+ntCGDSeq+2cBPNK9Af72hBOtD7S+H44tr96mdCZB2ZCUvHgGNPHE7/IPNoJHOfLEfbRum0AjXxcuphUyOioQr79hUi4iIiIiInKnUGeih6ur7S7kqFGj6qTf3bt3s2DBAo4ePUpKSgo//fQTI0eOrGhPS0vjhRdeYMuWLeTm5tKjRw8+/fRTIiMj62T8f4v0Aj3xWcUci8/B31VDVIgbfjo1CgeREDqNgv+NbsnkpYdtxAcPZyXfPNwRf1dhMZWcV4LBVFk956vdMSwY25oXfjxFShUX+0A3De+Pa40VKxtPVl+kKWVSnhnQCG+dCi+tirHtAmnip2PV4UT+OJuKXCaxm7u8cOcVPhjXmmfXnbQJr20X6sas3hHXjVipN8wmKEiGpGOQGw8BbYUUFhc/KEyDLrPBrxUkHwOlFhr2gYJUSDggPJ++DS5tAakcwrqDi6/98roiIv+Q/TFZNPXX3RFRc1KJhABXjU2VDBGRm4nBaCYtX8+R+BwyC0vpEOaOv5uG9XbMvstZd/Qq49oH4e1iK5rkFRs5W+ZhVVxqZvelTJ4Z0IiPtl60SSt9sFMIXlolO6PTaeTrYhs1otBArxcg/TxcPVh5XKlFP2ENL28WqsAMj1TicvBHh3P0vPAdq0e/xKV8H0JDwjBbrMRmFrE/JhO5VMLQMFDnRCNNPwfh3YV7WdxekKshvAdofUFdvWRvnZOfDOkXIO00eEaCX0twDYI74O+diIiIiMitQ52JHsuWLaurrmwoKiqidevWTJ06lTFjxti0Wa1WRo4ciUKh4Oeff0an0/Hhhx/Sr18/zp07h7Pz7fElNDm3hOnfHOZcSqX5mEouZdnUDnQI9XAofAS5O/HDI51IyC4mOrWAIA8NkT4u+FfZvbp22ZCWb+Cln07z3MDGAOTrjejUCqzAoyuO0NjPhRCP6uX4Phzfmh8OJ/C/TRcqjkkl8MrQZihlEiTVRhI4lpCDXN6AJZPbE5tZRGZhKZG+WlLz9OyOTmdo6wBc1NXL9tUrFjMkH4dvR1YakwJ4RsCkn0DtKkRy7Hynsm3bPOj5ApQWwMox4NkQBi+AtVOEPoZ9As1HgUr7774WkTuaAr2Rs8l3hp9HOf5uoughcmtgMJrZezmTmd8dxWiuVCXah7rzzIBG7LyYgdmOCZZE4uCed80X9e8OxDO8dQBLJnfgSmYharmMQHcNey5lcv/ig1itQmTmD490pqFPlXuHLgAmfoc1N5GCuKOUqHxId2pIkcWXo4mHAQQRRVJDyopEivuxz+mQeg7ThO9Zc8LCfzecwcdFxbqx3jitGC8I+UPeF8rCn7lGQOn/JrR9SCgVX19kxcCK4bbRKhp3mPwr+LWov3FFRERERO466iXJs0+fPuTm5lY7np+fT58+fWrV1+DBg5k/f3619BmAS5cuceDAARYuXEiHDh1o3LgxX3zxBYWFhfzwww9/d/r/KsWlJj7YEm0jeAAYTBYeXn64wmjNET46Ne3DPHigcyg9G/kQ4Kax2REOcNOguSaa4mpOCU+vOclXu2LwcVEx75dzPLPmJMl5eg7H5dAx3NbBvktDT84m5/PXZVtTNosV3vj1HKPaBmEwmXF3qi5etAjUcSW9kPsWHeTjPy/x84lknl59knm/nOPFn85wNafkht6nOiU/GVaOtRU8ALIuw75PBL+O8xtt26xWQQRp0EvYicuKgf2fQ9tJYLXAxjmC14eISB1yPCEXixWa+N05JUUDXNVcTi+kHqqli4jUitR8PTO+tRU8AI7E5/DH2TT6N7OfgjWhQ1BFWfaquGoUtA6yjXrdeDKZqcsP4+Gk4vczKUz75ghL9sZWpIKmFxiYtfIYmYUG2860PkiC2hEdNIZhW7QM+zaBtCIjbmX32Q0XDRQ0He/4xTUeAld2IUk7hWznfC4kpGO1wn96eBH85wxB8PBqJNy/rhU8ALa+Iph41xdFWbB+evX0nJIc+H68cJ8WERERERGpI+pF9Ni5cyelpdVNuPR6PXv27KmzcQwGYZGgVldGNshkMpRKJXv37q3xuvz8fJvHzSKrsJSNJ+3f3PVGC2eT86odzyw0cDGtgPMp+aTklWBxVI4FYRfpndEtqx2XSyXM7hPJJ9suk1dirDheaDARm1nEq0Ob8uWD7Vg8uT1vDG/O2qP2v9BbrXAoNovDsdm8MaIF16Q4M6J1IKuPCIua9AIDMRmFNv4eqw4lXPfLj9VqJSNfT35GEqUp5zAlnxYWRH/3S1N2DOhz7be5+MP+Tx1fe+E3iOgv/HxlO4R2rWw7eXsIbTeTW+mzdztwOC4bnVpeYQB6JxDoriFfbyKryL5Ro0j9cLd/9goNRuIyiziXnE9CdhF6o4n9MVmYHNw/1xxJZHKXUOYNb86XD7Zj4YNteXdMKwY082VYqwCk19zscotLydcbeWNEC5yU1dM2A9zU/HU5y27kSHRaAVmF9j8PYZ5amvoLouc3++J5eUhTpnX04fnOziijJoK7nSiw8J5gMlRUgZGe/ZFxTQVPj7aeJiF1BqDFaDix0v4bBnDwKzAbHbffIEazhaTcEs4l5xOTUUhOcSkUZwrppfbITxLSTEVEREREROqIOq3ecurUqYqfz507R2pqpRmm2Wxm8+bNBAYG1tl4TZo0ITQ0lBdffJGvvvoKZ2dnPvzwQ1JTU0lJcVzy7X//+x/z5s2rs3n8E4xmS7VdpqqkF1Tu/pgtVi6k5vPU6hNcTBOiFLy1Kt4c2Zxukd5oVdX/O5VyGf2a+fLr491YvOcKVzKLaBXoyoNdQnnn9wtcSLWNMJFIBFf5S+kFvLs5GoPJwmf3R9X4BSWnyMikLiFcSitkzYwurD2SyPnUAhr5utC7iTef7rjk8NqUPD1mqxW5g/zd4lITF5JyCCq9gm7rHMi8KDS4+GEd+jGSsO61Tykpqm7+WoFKB0UZNV+r9RF+tlqFVJly7BjKidhyK332bgcOxWbT2M/ljvDzKCewzL/gcnrhrWlkfIdyN3/2UvP0zP/tHL+fTsFiFTyqXhjU2Ebwv5bmAToUMinL98URm1kEgJ9Ozdujq4sa8VlFPLv2JIfjcvi/CW34fnonfjyWxKmruXi7qBjfPvi6FhV6O2bfINyPF4xtxbGEXFYeiKeLj5ExV5chXbcanDxg6MdYs68gOfezYL7dfJRQ/eX3Zys7MRtRIHhuySxVIkqud78rTBWqyMj+fgpqTpGwsfP+H9EUGIQ5tA11Y/UwJ2rsVX93iXIiIiIiIvVLnUZ6tGnThqioKCQSCX369KFNmzYVj3bt2jF//nxeffXVOhtPoVDw448/cvHiRTw8PHBycmLnzp0MHjwYmcyxQeaLL75IXl5exSMx8ealJTipat7FbR3kVvFzUm4x47/cXyF4AGQUGpj53TGiUx0vEJxVcloEuvLOmFaseLgjrw5rjq+LmnxD9QXf0Jb+nE3OY/Ge2AoD1OjUAtqGuDvsv0WgK8UGM8//eAqpVMIbI1uw4uGOvDWqBX46DR3DPBxe26+ZL3Kp41/D2Iwi3I1p+Pw4qlLwAChIRbLqPqGqSm3xaeq4LScWQu9x3B7UAdLKquFo3IXdtHIaD6n9XO4ybqXP3q2O0WzhZGIujXz/BTPBfxE/nRqpBBsDZpH652797OUUlfLcupP8eiqlwky01Gzh850xNAuwXwZaKoGn+jfigcUHKwQPENJhpn9zhLis4opjKXkl3L/oIIfjcugR6UVMZiFjvtxPUm4JXRp64qVV8fyPpzCYLNUiIctRyCS4OzuWAHx0aga18OOzMQ3x2/8m0hMrwWKCwnRYdT+S6N+xDn4PfFsIaZcbH7e9N7kGk1wsrIkKpbpKESP9HAR3cvzmRQ4ARXWPr9qw62IGr208WyF4AByLzyWmUCGkitpDIgHXutsgExERERERqVPRIzY2lpiYGKxWK4cOHSI2NrbikZSURH5+Pg8//HBdDkm7du04ceIEubm5pKSksHnzZrKysggPd2z8p1Kp0Ol0No+bhZ9Ozcv32v8S3jbErWJXFOCPM2kUldrfDVrwR7TNrlVGgYHUPD3FVRYaaoUMNyclSrkUd2cl/x3SDIkEZFIJA5v7Mm94c6Z1D2f1YdvF8OrDiTzSvYHdnapgDw1uTgp+PJ7E1K5hBLs7oZIL46jkMrRqOc8MaIzczmrPx0XFPQ09q3daRpHBxB9nUvGM+xVKi6qfYLVi3fF27XeEtH4QMaD6cfdwaDIMejxnf2fL2VswV0s+LhjI3fuB4Hg/+D1o8wCEdK7dPO5CbqXP3q3O+ZR89CbLHSd6yGVS/Mp8PUT+Pe7Wz15WkcFuaffsolIK9EYa+1aPFOzRyJtd0Rk2lc/KsVjh8x2XySmLfryUVkhSbgnNA3Q80S+SAFcN93UI5lBsNl/uusKqw4nkFhvZfCaV+zuGEOCqZk6fCN4a2YJZvRriq1MxrVs43naingoNRlLzSsgq8/twMeciv/BT9RcZvw/J1cNYk44J1Vf6vgb3fggdHwG1K6Z+b7D8tOAPtuJ0CYVtZwjXnVkPUZOEyJCqeDeBAW9Bo0GC91VpkZBSWphe43t9LWn5et7fEm23beGRQkxdnrJ/Yev7hPutiIiIiIhIHVGn6S2hoaEAWCzVFwr1TXnJ3EuXLnHkyBHefPPNf30Of5dukV4sfLAtb/12nqs5JajkUsZ3CGZ2rwi8XISFkNFs4WBslsM+zqcUUFJqxmAys/VsGov3xpJXYqR7pBeP94kg1MO5WhWYJv4urHusK6VGM7+dSuHzHZd5c2SLagu99AIDG04k8cnEKD7dfomLaYXIpRL6NfPlvg7BPLvuFAGuaj6e2AZvl+oLtwZezqyd2YX/bjjD2eR8pBLo19SXl4Y0JdDd8S5SUakJzAZ02YccniNJPSUsyNS1WMA7e8KIT2HfZ3B0qXD9gLeEMnl/vFjmnP897Hi7UuCI6A+dZsBvz4B/axj+GZxaXWkA13Lc3/cYERGxw7H4HOQyCWGet0cVqtoQKFZwEfmXSMs3OGx787dzbJh1D1/uimHD8WRKzRbCvZx5qn8j3v7tvMPrzqXkU6A34u6s5GxSHu+OaUVOcSnPrztNdpGBDmEefH5/FF/tvsK+GOG+/f2hBDY/0Z3ujbz5cmcMcVlFNPTW8r9RLWns54JGWbkc0xvNxGYW8dHWixyOy8ZLq+KxXg0Z4Z3j+D6z/U14dDfWK9uRHFoERekQ1AHLAz9y2ejNa0M9efePC6w6nk7X4RPo1csH7aGPYc8HMH4F7F4glGgf/J5wzzu8BPZ+AEEdofMsOLYCUk/BPU9AZH+hpO11MJgsDs3Kfz6TxaOdxtF8iAfseldIs1G7CmO1f1j4WUREREREpI6oU9GjKtHR0Xz66aecP38eiURCkyZNmDNnDk2aNKlVP4WFhVy+fLnieWxsLCdOnMDDw4OQkBDWrl2Lt7c3ISEhnD59mieeeIKRI0cyYICdnfxbFFeNksEt/Gkb4k5JqRmFTIKXiwqVvDJFRyGTEumj5c/z9ndaAt00WKwW/rPuDDujK3N0fz6RzOYzqfw8+x6a+NsKA05KOX4uKsYs3F9RJUaCENp7rdfa5jOpnE/J55OJUaTklWAFdl/MYOZ3xygxmukY5u6wbK1KISMqxJ1vp3Ukv8SETCqE8mpVNecJq+UyzFI5Ja4RaNhm/yS3YCGPuba4+EHfVwUhw2KG02tg7WShLfk4pJyEdlOFcn76PMiNB0MhDHxL2AX7dpRwrJx9n8DZ9TB1szAnEZF/yLGEXMK9nFE6KFl9OxPgpuHgleybPQ2RuwB3O1VWyskvMSEF5g1vzty+kZjMVpxUMnQqOcEeThyMtf87GuSuQVPm69GloSef7bhsc2/eci6NHdHpfPFAW2IyCknLN9CvqS97L2fy5q+VYsqR+Bwe/uYIb41swZi2QajL+jybnMeErw5UmKzmFBt5es1Jes4IxmFsZMcZWHcvQHry+8pjF/9AenkbHmPWY/X05cPxbcgtNmK2WLE4NYKosWDSC2km960Rojr++giOLq/Sx2a4vBVGL4aU4/DzbGg2UogkcXYcqQlC2o6bk4LcYvveKZeKVDRvPw2aDAGjXriXa/1AVm9LUxERERGRu5R6WU2vW7eOFi1acPToUVq3bk2rVq04duwYLVu2ZO3atbXq68iRI0RFRREVFQXA008/TVRUVIU3SEpKCpMmTaJJkybMnTuXSZMm3Tblaq/FV6cmzMuZwLIUkWsZ2y4YmYOk4GcGNCIt32AjeJRjMFl46/fz5F9j2mY2W1h/LKlC8Ijw0ZKar+fhe+ynBglGqgXM/O4Yj313jB8OJVZUYhnXPtiukWpVPJxVhHk5E+zhdF3BA0CnUdC3qR+FzR8Qdp7s0fM/gpnb30GuLBMorMJOEwhRHJH9hUXg9jfhh4nCDtSvT8Evj0PsHqGcbVXBo5y8q3D2J7gJkU4idx5H43OI9K6lSe9tQqCbhtR8PQX6f14ZQkSkJry1SiJ97H+OejX2xkOrQqOUE+TuRJiXMz4uatRKOQ91CbWb0tnMX8dLQ5ri7SJ4cUklEv48n45UAu1C3enVyJsAVzW+OjUHYrJ4un9jJBKY2CGYBX/YT/V4Z9MFUsruw1mFBl7+6YzdqjJ/xJmxBnWo3oFMgTWsm63gUY6LPz5xG/GSl+CiVhDsIbxOnZNa8M3wbChEN2q9wGywFTzKsZhh1zvQYbrw/NwGocLKdfDRqnikewO7bU5KGe1C3EEqBV3ZPFyDRMFDRERERKReqJe7y/PPP8+LL77IG2+8YXP8tdde44UXXmDcuHE33FevXr1qLGk6d+5c5s6d+7fnejsR6K7hq0nteOKH4xXeHjKphFm9GtIxzJ0vd8U6vHbPpUzS8vV8tuMyFouVe1v54+OiYuPJZCJ9tDw/qDGxmcWcTsqjmb+OHx7pzII/LnAsIRcQvDs+ntCGU1fzUMgkFRVnVHIpLw1pQmNfl4pUnDoj7ypt8g9hLc7GMvIrpL8+AcYyAzmpDEvP/yC1twCsLTmxENwRuj0lRHnkJkCDXuAeBn++LgggA+YLCzKLWTCKc8TptRD1ADjVvAMmIlITmYUGknJLGNM26GZPpV4o9yqKySiiTbDbzZ2MyB2Nt4uarx9qz7Tlh7lSxZS0TbArz/RvxPJ9sQxs7oe/qwadplKMD3LXsGBsK/674Qx6o4WoYDfm9o3kSmYhaw4ncjGtkM4NPDifks+4Vh7MaeeEW8Jm1MUpSDr1w+oWjPXkD0hLvOg6rS96jRS90b4gXmAwkVVoINzLmXy9sVpVtXLe3Z3BvY9+ieuGyZB2prIhciBkXVMlTesDA/8HhgJIOoLs2FJoOkIQOJQOUuYSHaeSkhENuip/j2K2g38rx+cDMpmU8e2DiEkvZP3xSpHE3UnB0ikdCHBzYGQqIiIiIiJSx9SL6JGamspDDz1U7fiDDz7IggUL6mPIuwK1QkaPRl5seaoHiTkl6I1mwr2c8dSq0KrkaJSOA3cUMgnRqQV8vfsKAIv3xvLh+NYEeWh4qHMYT64+YWOEqlZIWT61I/pSEy5qBTGZRTz+/XFaB7ux8MF25BSVIpVI0GnkZBQY8K/rxUvWZVg2BGlhmvA8vCeM/hqrXIPFakXi1RCps0/ty9XaQ+UGHWfA6klCqG85Tp4w+mvQeMDx7+DSFugyu+Z0GrkKJI4rB4mI3AgnysTGCAc71Lc75V92LqUViKKHSL1isVgpKCllevdwPLUqsgoNeLuoScotJrfEyKpDiXy49RLPDWzMg51DcNUI6TAezioGt/CjbYg7GQV6rEiYuuxwRXTjdwcT0GnkbHg0ioHWK7itml3pt3F8mRC5MPg9WPUAwZY3MI79lr6Rrmy7lGd3nnKZEFYilUiQSOxbd+SVGNmdpmHYpJ+gIFUwF9UFgC4QyaU/Kk9UOMHIL4WytdlXKo9vfwvGLIXGg+xXTnFUTaWcqqEvyhur6uLtoua14c2Z0yeCuKxiXDVy/F01QhUnR+VsRERERERE6ph6ET169erFnj17iIiIsDm+d+9eunfvXh9D3jUoZTIC3Z3sGoAOaO7H+1su2rkKBrfwZ+dF29SXRXuu8OaIFvxn/WkbwQNAb7Qw5/tjzB/ZgvX749Ao5STn6UnOS2XTmVRcVHKsQKHBxG9zu9XZ6wOgOBt+ngPlggdA7C6I3YXE2QfZo9vBtQ59MzSusPJJW8EDoDgL/pwHo74UBA8Q8ps7zYSE/fb76jQTNG51NzeRu5ITibm4aRR4aR37EdzOqBUyfFxUopmpSL2Tmq9n6vIjZBWVopBJcFbJKdCbMFusNPLVMrVbOG/9dp4Ff0TTu4lPhegB4KxS0MBbgUYpY/hnf1UIHuXkl5hQFKXjtnlOdZUiKwZOfA8txsCJlSh+nMJ/xu+wK3r46dS4Ownjujsp6RbhZbfijFQCrYLdQOssRHJUjbQI6SykgVot0Gq8kKZSVfAAoW39NJhzFDzspLEGtgOpTIhovJawbpB0rPJ5g97Vz3GAq0aBq0ZBgzs0XU9ERERE5NanXkSP4cOH88ILL3D06FE6dxbKeB44cIC1a9cyb948Nm7caHOuSN3gp1PzVL9Ivtkfz6PdGxDhq8VgtKBVy/B0VvLZjsv8PLsrVqDUZMVVI8dotnI5vRCJBMa3D6Z/M18MRgtKuZSrOcWoFTI2n0nl2+mdOHgli7isYiQSGNzSn4HNfZFLJVisVi6lFQilcJ2UNiHCf4vibMeigsUIJblCxRVDIWjchdJ2taneUpWSXGHHrCTHfnvqKSisIhZlxYBcI6S/XNlpe254L2FhKCLyDzmemENDby0Se6YCdwiBbhoupdkP4xe5O0jL15NVaKDUZMFTq8LbRYVaUbeRcmn5erLKyssazVYbU82LaYWEelRuIPx4LJGQfo3ILCwlp6gUJ6UcL62SzAIDGQXVq8A08tWiSdoniAn2OL9RMAA9sRLMpYSa4jjxXAc0xhzBIFvtRqpJS4ZJhcFoJjWvBF+dmnnDm7PyYDxdG3qhr3I/dtMohdK2JTmC15Q+X6hy4uwliCD3fgi/PgkN+8C6qfbnZDEL91d7oofWB4Z9IpiVVsXJA+55En4qK3U74G3BcFREREREROQ2oV5Ej1mzZgHwxRdf8MUXX9htA5BIJJjNdnYURP4WOo2CKV3DGNLSn2fWnuTUZmFHSSKBx3o04NHuDZm18jhJuUIJOZVcypIpgifGWyNbcCYpn5nfHq0wUIvw0fLRhNZ8M60jwe4a/m9iFEfjcwjy0LDnYgYzqpzb0FvLq0Ob8vW5KzzeJxI/V/XffyHmUvvHnTyEBeSvT8PVstxjiQSajYJBb4OLf+3GyU+GzS9BZN+azzMW2T7//RkY+Lawmxa9WTjWbgr4tgCX65fxExGpCavVyqmreQxqfmd/qQh013C8LI1H5O7CYrFyPjWfmd8dJTFbuB8pZVLm9Ingwc4heDjXnT+U3ljzGqPcn0oqgd6NfHjr1/OsPpJYUcGsZaArLw2xX3VOo5ChKHUgmAOYq0RQKrXIPUJx+XUGstgdFYeDIgbgO+RDen15HItVwuf3tyXU0wmT2cqj3x7FXDaRRr5aFj7YFid9Gmx8HGKqVDRr2BeGfypElQR1gJJs+9Ea5RQ7mLPSGZqNgIAoOLIc8uKhQR8I7QJHl0HTYUI5WbcwULs47l9EREREROQWo16qt1gslht6iIJH3WMwW3hs5TFOXa0MobVaoX24Jw8uOVgheIBQ1SW/xEj/pj5czSnh+0MJNo7xl9MLeXj5EYLdnTiekMuIz//iQmo+55Lz+faA7bkxGYU8u/YUnRt4Mv+3c/+sKoPGVdhxupaeL8DWVysFj/IXd3Y9bJsPhqLq1ziiJA9+ewbO/QRaX+za9AOo3YRHVUwG4dptb0DUJBi7FCL6ioKHSJ0Ql1VMgd5Ewzs8FDzIXUNSbgnFpaabPRWRf5nkvBLu+/pAheABUGq28OHWi3YrkP0TAtw0OLKOcKlScWxQCz92Xczgh8OJNiXbTyflYTRbkNvpJCajiKKAexwP7ttcMMYG6Po4kj/+YyN4AMgub0Gx6RlWTmpKeoGBRXti+OFQAt/sj68QPECISjlw9grWn+fYCh4gPN/4uCB0+LUAr8bg3djxvMK6Om5TuQjzHvwujP8OOs8UqpoNek+IJPFvLdyjRURERKogkUjYsGHDzZ6GiIhD6r02mF6vR63+B7v+Itel1GQmu6gUixXS8/XkFJXyWM+GtAh0xWAycym9ECtW3hvbCpPZypZzqWw5m4bJYmXtkavM6t2Q6d8csdt3RoGBS2mFXEwr4JOJbQj3cmbS0krRoV2IO4/3jcDHRUWpyYLZYuVMUh6peXoK9SZ0TnJKDBaMFgsSwGIV9AVPZxVKuQPNTesPA9+BHx+uPCaVC9VTqjrWV+XUKujxLKjsl9utRnEmRP8u/HzxD0G8OLai+nn9XhfGVbuBPveaRgn4NAXZP0znERGpwqmruQCEezuosHCHEFTmS3Q5vZBWQW43dzIi/yqHYrPJ19sXuz7cepFuEV746Opm3eDprGR69wYcuJLFuHbBeGqVJOWWsOpQIqPbBrL2SCIAo6OCePyH43b7+OVkCo90b8DCXTE2xwsNJoqdArGG9UASt9v2IokEuj8riOOANfQeJDv/Z7d/2aU/COpbRLcIL2b3jmTSkoMEuKqZ0DGECG8tRaUmNhxPoqufFcnO7fZfaMw2TIXpmJU6VC6+MOR9WDG8mteItWEfSjT+mPVGXNSV9y5DcR4SfT5WQOLkQY5BhtlqRa0oxcNZKZR4FxERuSVJT0/nlVdeYdOmTaSlpeHu7k7r1q15/fXX6dKly78yh5SUFNzd3f+VsURE/g71EulhNpt58803CQwMRKvVcuWKYKb1yiuvsGTJkvoY8q4lJbeEBX9E0+/D3Yz64i/SCvS8O7YVJxJzmfPDMX46dpXejb1ZsS+euT8c56WfTuOiVrDoofZ4OCvZEZ2O1Uo1I9OqXEwrwGKF59ad4mpOSUVO9HtjWvL68GasP5bEqC/2MfbL/Sz5K5ZP7osiObeEbw/Es+l0Gq9tPEt0agH/+fE03d7dTp/3dzH/t3M2USc2SKUQ0Q8eXC/sOAF4RggRFo6wmKC0FqaI+ipmckeWCLtX/V4HXaBwzLsxjPpaKGe7ZjI8sBaajRTEF7lKEEke3gxudWioKiICnLqah69OhU59Z4tp5WVrL6aJZqZ3G1UjEa/lak4JRrMDj4y/gVYtpH2OaxfEl7timLXyGKsOJfJEv0gifbQcistmWrcwAtw01YxKy1l79Cr9m/vy/rhWBLkLv7cNvJx5d0xLfr9i5ESH99B3e0HwmAII6ggTf4BzG6C0gLxur2C5XlUvQz7NA3Sk5JXQq7EPL93blO3n05j9/THe2XSB1kFu+KpquAcCGRkZvPHLOa7mFENgW5i6GQLbC41OHpR0f4ljUW/T9sPjPLn6BBfTCjCWGjGmRSPbMAvlZ61RfdEeNv0HF0MKM749ykNLD3LgShaFBjEiS0TkVmXMmDGcPHmSb775hosXL7Jx40Z69epFdnb2vzYHPz8/VKq6S00UEalr6kX0eOutt1i+fDnvvfceSmXl7kDLli1ZvHhxfQx5V5Kar2fK8kMs2hNLocGEVCLBS6tm5rdH2X8lC2elnOndGzBpySF2XczAYoXiUjOrDyfy5q/neH1YMwCyiwy41mA+6qVVsepQIgaThRKjGTcnBRM6BBPh48LD3xxh48lkDCYLJouV30+nMmnJIQLdNbio5Lyy4QwTOgQz87uj7CybQ4nRzIr98Tyw6AApeQ6ED42rkDLy0EZ48gw8tAE8Ix2/GVI5KGuRDqCuEp5rtQrpKud/gR7PwWP7oO0U2LMA/vo/SDkO344SShDOOQyPH4fBC8A97MbHExG5QU4m5hLmeWdHeUBlBRfRzPTuo1WQ4/SIIHcNClndLU2KDCaW74vjlZ/PVgjtMRmFzP3hOKl5enY+14sXBjVBp5GjqcFE1WyxMrZdMOsf68r2Z3oyqUsoi/fE8uHWS4z+7goz43tzePBvlD5+Gu5fjTmwI4kd/suOXut57EpXDMqad0AtSh1f7b6CUialdxNv5nx/nJNl4lB2USkLd8WQb625TKxBrmXlwQQmfn2A5GKZUNHlgbWUPn6aQ4N+YUZsD8auvILeaGHb+XQmfr0f8uJRLO2L/OKvQnqMSY/y5AqcVg7lu3GBnEnKZ+LXBzgU++99eRIREblxcnNz2bt3L++++y69e/cmNDSUjh078uKLL3LvvfcCQurJwoULGTx4MBqNhvDwcNauXWvTT1JSEhMmTMDd3R1PT09GjBhBXFyczTlLly6lefPmqFQq/P39mTNnTkXbtekt1+tv586ddOzYEWdnZ9zc3LjnnnuIj4+v8/dHRKScehE9VqxYwddff80DDzyATFa5iGjVqhUXLlyojyHvaIxmCyZL9Z2v8yn5RKdW7pKObBPIB1uiaRnoyrh2QbwwqDGrDgtixbVcySwir8REqKcTMqmUqfeE2R3b20VIQ8koFHaYfjqexAOdQhnXLog/z6fZdbTPKzGy9uhVziTnMaJNAN8fSkBvrD6HuKxijsbXYAIHgiu9W7BgUuriJ5iF2qPVBHC24wPiCCcvaDzE9ljSUcEvZO9H8MeLkHmpsq20EPZ8AImHwDUQlJobH0tE5AYxW6ycTc6/a0o7BrlruJAqih53Gx3DPWz8NKryVL9GdZbaApBZaGDxnit22xb8EY3RbEUpl+HjouahrqHVzvF0VjK1ayjhZUKks0rO+1uimffLOS6VlVy2WmHnpWzGfR/HtmQFOHkg03pSpPHjkZ+S2Hclh0uFKiyh9qt7mSMGciJbeD/0JgufbLts97xfYoxYG/Sx30d4b7bGC5EqV3NK2B+TJTQ4ebA5Ucb47+PZfTm7IttFIoH3RjRGsv8zMNj5DOZdRRm3g+GtBYPw1zeeJS1fX/08ERGRm4pWq0Wr1bJhwwYMBsfRYK+88kpFRMiDDz7Ifffdx/nz5wEoLi6md+/eaLVadu/ezd69e9FqtQwaNIjSUqG4wMKFC5k9ezaPPvoop0+fZuPGjURERNgd63r9mUwmRo4cSc+ePTl16hT79+/n0UcfvaMr1oncfOpF9EhKSrL7QbBYLBiN/8Dg8i4jNV/PlrOpzF55jKdXn+TglSyyCiv/oG06nWJzfsdwd6Z3b0CPRt5kFZVyKa2Q8e2DGdkm0G7/+69k8kz/RhjNZsa1D+bhe8JszNoaemv59L4oFvwRXXFsz6VMgt3VyKQSDlzJcjj3befTaeyno1WQW+Xiyw6/nEy2K+jYResDE78X3OnLkUig+Wjo8wqoarE7rnGFez+ARoNsj0f0g9hdjq8782PNaTYiIv+A2MxCSoxmGnjd+ZEeAMEeTkSLosddR4CrhlUzOlekioBQveWpfpH0blIL8foGSM4tsTEmrUqBwUROWTlbpVzKtG7hTOwQjFQCcqmEV4Y25fXhzUgvMPDyhtP8fjqFzEIDR+Ici/W/nErGUjZgiLuGRQ+1x0urZPqaGIrv/RxzeE+b880N+1M66H1m/ygIM1YrxGbaN+X+aE86mX0/wHyN8GEO7010p//x0d5KE9hfTiVjMJkxmi38csp2rdA13JXtU0PorbmMPGarw9eiubiRAY2EqJyE7GIxxUVE5BZELpezfPlyvvnmm4qIiZdeeolTp07ZnDdu3DimT59Oo0aNePPNN2nfvj2ffvopAKtWrUIqlbJ48WJatmxJ06ZNWbZsGQkJCezcuROA+fPn88wzz/DEE0/QqFEjOnTowJNPPml3TtfrLz8/n7y8PIYOHUrDhg1p2rQpkydPJiQkpD7fKpG7nHoxMm3evDl79uwhNNR212Tt2rVERUXVx5B3HKl5eqZ/c5gzyfkVxzaeTGZ0VCAv39sUT61KMBcrw1urwsNZxaSlB8kvqVyYfHswnlfubUaBwci28+k2Y3g4K+nS0BONQoZWreDZAY2Z3DWM9HwD+XojZouVjSeSSMgutrnu5Q1n+Hl2NxsTtGvRqeXojWYMJjMuarlDzxB3JyWy2ii77qFw3yrBiNRQCBo3cPa2TVe5UXQBMOpLKMoUPD6UzmAsEdzrC9PtX+PkAdLr5GaLiPxNyr0Owu8W0cPdiZ/zk8krMdaYYidyZyGVSmge4MqPj3Ulq9CAwWTBS6vCS6tEo6zbZcn1+qtqqO3joua/Q5sys1dDjCYLH2y5yOazqRXtf5xNIyrEjTdHNGfGd8fs9ufhpERatnngpFLQPcKLdTO7kl1cysUSKw2HLkJjzMGqzwO1K5lWHQcSLGQXC+KLTCpBKsGuUFNcaub3eAnqsHkM6j0ffUEOJTItfyaY+fD7eIpLKz1J3J2UQl9IcHeq/Gw18tXy0T0mfNf0hR7PC/c7B5jVbhSWVZCXSEDhqAyOiIjITWXMmDHce++97Nmzh/3797N582bee+89Fi9ezJQpUwCqGZp26dKFEydOAHD06FEuX76Mi4vt3wO9Xk9MTAzp6ekkJyfTt2/fG5rP9fobMGAAU6ZMYeDAgfTv359+/foxfvx4/P39/94bICJyA9SL6PHaa68xadIkkpKSsFgsrF+/nujoaFasWMGvv/5aH0PeUVgsVn4+kWQjeJSz/ngSEzoE46lVMSoqkC93CbtDD3QO4fWNZ20EDxB2jf636TwLH2xXTfS4r0MI3i6VYcROKjnORjOv/HyUC6kFuDkpeHNEC74/lGhzndkCiTnFDG8dwK6L9ssLTu4axs/Hkzkcl82E9sF8uSuGotLqJnH3dwqpfTibs5fwqAs07pXmc8nH4fdnodVE2PGW/fPbTxf8Q0RE6oHTSXn46dQ4Owj9v9Mo3+m/mFZAhzCPmzwbkX8bX50a3zpMZbE7hosKT2clWWURHVVp5Ku12TwA0KoUaFUKtpxNtRE8yjmekMvV3BJaBbnaNWSd2NF2p1IulxLm5YwqV8q9n+4l22YemQCseLgjMqkEs8XKvphM+jTx4c/z1YV3hUxCuJeWh5aeY12YO8NaBfL27xfQmyzo1HKclLIK4WNSl1DkUkHQeaBTKGuOXAXgpe5u+G6ZKEQsnt8o3O/+fM3ue2doO52FPwkld/s08cHdWazgIiJyq6JWq+nfvz/9+/fn1VdfZfr06bz22msVooc9ytffFouFdu3asXLlymrneHt7I5XWLjHgev0BLFu2jLlz57J582ZWr17Nf//7X7Zu3Urnzp1rNZaIyI1SL+ktw4YNY/Xq1fz+++9IJBJeffVVzp8/zy+//EL//v3rY8g7iswiAysPJjhsX3EgHqPZgr+rhhcHNwGgqb+O44m5ds83mq2k5Orxcal0VZ7bN9LuAia3qLQixz632MiVzCImdqheoWRXdDrNA3SMbBNQrW1Qcz/kUglz+kQwu3cETfxd+N/olix6qD19m1aGLj/eJ4IQj5qN2WpNforgzXHxD0i/AMWOU2uqceJ74VqPcAizk3vd9QnBzFREpJ44fTWPMK86/kzcwgS6aZBJJaKvh4gNBqOZxOxi9l7OZM+lDBKzix1WVrkevjo1X09qh1phu9xxc1Lw6X1t8dJWrzaQry/lh2vE/qqsOXyV+ztWD8N+sl+kTcpOVTILDdcIHpV8eyCeD8a1RiKBH49e5aEuYdX6kUrgowlt0Knl7HisOSuGaBjneo4TM/w58Uwb3hnTirdGteSrSe14c0TzCg8SgFBPJ2b3FlKOGzrpIT9ZaEg5CS6+0KBXtTkZ2s3gsiWA+KxiAlzVvDa0WY3RnSIiIrcWzZo1o6ioMlXuwIEDNu0HDhygSRPhO0Tbtm25dOkSPj4+RERE2DxcXV1xcXEhLCyMbdu23dDY1+uvnKioKF588UX27dtHixYt+P777+vglYuI2KfethMHDhzIwIED66v7OxqrFQwmxwu8YoMZqxV0GgX3dwyhd2OfitBYR6gVUsa0DURvsnBPhBd/Xc5k4tcHWD61g41poslqG1P70daLzOrVkK8ntWPv5UysVujUwIOMAgPbL6Rxf6cQBrfwZ8/lTCwWK/dEeOLprEIhk3AsIYf3NkdTWlZ+UCWX8vK9TRnZJoAIHxcC3DR1G9KefgFWjoW8KovVhn1gxBegu4GQufKStz/Pgf5vQLupELtbKFHboLdgouok7kaL1A+WMhPTkVH2PXjuROQyKQFuaqJTq0e1idydFOiNbDqTyisbzlSYcCtlUl4Z2pQRUYG1LuUslUpoHezGlid7sutiOtGpBbQNdadjuEdF2eRrMZmt6Gu4B+tNZjo38GTVo53ZdDoFF7WcYa0D8HNV46qxHw1hdGQsAmw9l8azAxqx7eme/HE2jf0xmSx6qD1xWUX8dSmTIHcnBjb3xWK1EkAGmh8nQdqZiuvVvi3oNeZbBn8TT0aBgQ/Ht7YRedyclDzSPZyhrfzRFV9jJr9xLvSfB20nQ+wuLCoXTE3Hki7xYs3hHBY+2JbWQW4EOHivREREbi5ZWVmMGzeOhx9+mFatWuHi4sKRI0d47733GDFiRMV5a9eupX379nTr1o2VK1dy6NAhlixZAsADDzzAggULGDFiBG+88QZBQUEkJCSwfv16nnvuOYKCgnj99deZOXMmPj4+DB48mIKCAv766y8ef/zxanO6Xn9Go5Gvv/6a4cOHExAQQHR0NBcvXuShhx761943kbuPehE9rFYrR48eJS4uDolEQoMGDWjTpo3oynuDuDspGdTcj2/22y/dNK59UEUesotGgYtGQXq+nhAPp2r+G+WEeTrz2+kUDEYLKw8kVAgRk5cdYt3MrhUhxm4aBb46FWn5lWadX+yMQa2Q0j7UnecGNuGhpYcwmS18PLEN4786gJNSRtsQd6QS+M+PyRQYTHw/vRNf7IypGAfAYLLw6s9n2TjnHpr66+rkvaogPwm+HQkFtoZtxGyH7fNhyAJQXmcHveV4IdrDpIdNzwtpL4FtwWwCfQE06Fnz9SIi/4DYrKK7ysS0nBB3J87ZSeUTuTuJzSzi+XW2BnylZguv/HyWZv462v2NNCi5TEqIpxOTuoTd0PkeziqGtvR3aMI9oJkffq5qwryc6dzA84b69HFRoZJL7VZT06rkaFUKAt01PNarchOiqb+OwS0qBfvC7DRU6x+1ETwASDuDeuOjfD7ya4YuOc+slcf448keRPpW5tO7OSlxc1JCrr/g41FescVcCptfFO53IZ2R9n4JpdaXYGB+UNANvTYREZGbh1arpVOnTnz00UfExMRgNBoJDg7mkUce4aWXXqo4b968eaxatYpZs2bh5+fHypUradasGQBOTk7s3r2bF154gdGjR1NQUEBgYCB9+/ZFpxPW65MnT0av1/PRRx/x7LPP4uXlxdixY+3O6Xr9lZSUcOHCBb755huysrIqyt/OmDGj/t8wkbsWidVqdbz98DfYsWMH06ZNIz4+nvKuJRIJ4eHhLF26lB49etTlcHVCfn4+rq6u5OXlVXy4bzYJWUWM+PwvcoptDUCb+LmwfGoH/Fyr77r8dTmTSUsOVjNAm9wlFJlUwtK/4uyOtWF2V9oEC74WVquV7RfSmb7iCNf+ZjzWqyEms4VFe2KZ1bMhV3OL2XgyxU6P0DPSizBvLd/sqz7m0Fb+vD+uNWrF3zAEzU8BYzHIlEI1F3lZaHL8Plg22P41MgXMOQLuYZS9SEEcMZWCRAoWE0ilIFUKu17XutmrXGD6NmFMiQycPQXTU5F/zK342btZ/HwiiSdWnWDRpPZo1XeHpwfAxhNJbDyZzOnXB1YYQIrUP7fiZ09vNPPculP8cjLZbnufJj58el+UY88bqxUKUu3fI2pJfFYRU5cd5so1lVS8tSrWzOxMuFftykobjGZWHU7ktY1nq7W9N7YVo6ICMVusZBYaMJmt6NRyPKw5UFpU9lq8Kc2MRfml43z34kf28cruQh5oriHARYavpweSa6MczUY4tRp+nl29g8ELhIgPxd97z0RERG5NJBIJP/30EyNHjrzZUxERuWnU6cr68uXLDB06tEJxbNKkCVarlXPnzvHJJ58wZMgQTp06RYMGDepy2DuSYA8nfp59D4v2xrL5dCpKuZT7OwYzul2QXcEDoG2IGxvndOPDLdGcuJqHr07F7N4RNPfX0edDx2VYMwoqozokEgmdG3jy02Nd+WDrRc4k5RHgpmFu30g6hrlTVGqmZaArPi5q3tl8wWGfV3NLaB9uf0cuIbsYvdFcO9GjJBeu7ICtr0JuAig0EPUQdHtSqMKSl+T4WrNRqMoCUJwNl7bCuQ3Q9iHY8wFcPSxY0zfoAwPfgqRRsHsBlBYIJW07PQbb3oALvwomps1HQ99XwE0srSVSd5xJysPHRXVXCR4AIZ7OFJWauZpTQojn3eNnIlIdvdFMfJb9cq0AiWX3DruiR3E2XN4Kf84TIv+UztDuYeg6B1z8aj2XUE9nlk/twKrDiWw4noTRYmVQcz8evies1oIHgEohY0SbABp4O/PR1ovEZhYR4aPlmQGNaRagI7PQwKfbLvPjsas82tGLhwMTYM88yI0HuRqiJiHvNEN4XaX23yO1qZD5mlVoNpZFLLqFwsD5EN6zssKZTAFNhgltO96CzGjwaAi9X4aANqLgISIiIiJyR1Knq+uPP/6Yzp07VzO6adKkCaNGjaJfv3589NFHFXWhRRwjkUgI8XTmlXubMbtXBBIJeGlVyGrYCdUo5bQIdOX/7ouiyGBCIZPiqVWRmF2MUmY/rBYgyN32i4azSk6bEHc+v78txaUmlHIpHs7CQsjdGc4l57NsXxxN/Fw44cA8tZm/joQs+6k2rYPdcFLWQvCwWuHyn/DjtMpjxhI49BWknYZxK2o2GFVqhYWixQznNsLm52HiD7DqfmFhWD5GzDZYdgwe3QUR/cBqERbSi/tUnmcxwek1kHQEpvwmCC4iInXA6aR8wjzvvgii0DKh43xqvih63OU4KWW0DnazWxUFoFWQq33Bw2KG87/AL3Mrj5UWwf5PIeMCjPpKiNCrJSGezjzZN5IJZWbe3lolTqq/70Pl5qSke6Q3LQNdK4R/NyclGQV6Zn57lJNX82gRqGOqfxzuv0yvvNCkh8OLkKScgr6vCemX1yKRIJFK0ZxYWnksNx5WT4Jx30DzkZXHNa4Q3h38fhDupXK16FclIiIiInJHU6fVW3bu3MmTTz5pt00ikfDkk0+yY8eOuhzyjkcpl+LurEAll2KuwQitKi5qBX6uGjzLXOld1HI+mtCGjmHuDGrhx8yeDXiir1BZ5eUhTfBzUDJQIgEnpbxC8ABIz9cz/7fz7IxOY1RUIEpZ9V8hmVTCjJ4N2XSmerk/jULKrO6hKEvzK6MvrkdBCmz5r/22+H3Crp5rEPi1sn9O18fBxV/oZ/s8aDZCCO8tFzKqUpIDZ9aDszfIlfDLHPvnaX2EMGpD4Y29BhGRGrBarZxNyiP8LvPzAMFHyFWjEH09RFDKZUzpEoZCVl3cl0klPNqjof0IwYIU2Pa6/U4vb4XC6veiG56TQkaopzOhns44qRQYjGZyi0spNVmwWq3klRgpNAil4otLTeQWl2I2299gKMfNSYmfq0bw2AASsks4WSb0PNVJh8feeXavk1w9KKRp2kuvbHwvEkdlJbf8V0gNvRaNuyDclwseJbmCfxVAaTEU5wiRkjVQarKQW1yK4W9W1xEREal/rFarmNoictdTp5EeCQkJtGzZ0mF7ixYtiI+3b84pUp3iUhMJWcUs3hvLxbQCmvq58HC3cEI8ndHcQGpIRoGe4wm5LNsXR0MvZ166tyk/Hk1iX0wWQW5OjG0fyMXUAg5cyaJdqDs+ZeJHar6egzFZfHcwHqsVJnQIplukF/6uGopKzWhVcuY/1IFzKfl8dn8U8387X2GgGuCq5r2xrQjxcOLrSe14bt0pknIFcWNUKx/+10eH6uB8SDwE7qGCIOEZCeoacspLC6sblFYl+YQQljvxe/jtaWGRa7UKKTCdZ0P7aUJIr6FAiNzwaQ5Hlzvu7/IW6DANSksg6Zhtm2sQDHlfKG276TlQOEPnWYLhqdbHfn8iItchMbuEAoPpripXW45EIiHUw4mzyfZ390XuLoI9NKyc3oln1p4kMVu4dwS6aVgwtlVFVFA1yv+2OyL9PPg2/0fzKjKYiM8qYvHeWHKKSpndO4ITibn8dioFtVLGpM6hyKTw9a4rdIv0Zky7QILdnW7IwP1YfE7FzwEak20FsmvJvAgdHoH9n5X5UcmFKI4mwxyneeYlOkyJEdqT4NIWOPk9tBwn3JMPL4aCZAjrAe0mg2sIyCqXjPqyssJL98ZyNiWfCB8t07uFE+bpjJMjzxUREREREZGbRJ3emQoLC3Fycrxod3JyorjYfsqDiC0ms4W9lzKZ8d3RCkPRU1fzWHv0Kksmd6BnI+8aTf8yCgy8tP40W8+n08DLmUmdQxn/5YGKaiqnrubx+5kUXhzchPXHk/j+YAIfTGiNxQqzvjvKsYTcir6OxOfQ1N+FpVM6oJJLeH5QY2avPEaBwURDb2ce6d4AH50KF5UcT62Sxn6CgNE1wov1s7qSW1yKQiYltOQ8ssXdwVTmIZJ8DM7+BEM/glYTHVdXkSmFhZ3FZL+9XGxwC4Yxi6EoUzCyU+mEXO5yIzuZSghfMeQLoc45sfb7c/YtG1Mq7ISVlC1IpXIY9glsfFyILikndhc0GymIIVpvh/8nIiKOOFP2hf9uTG8BIcXlcJUvfiJ3L0q5jI7hnvw4s2uZkbcVNydlRYUxu8jL/rY78GW3OHn+o7DWUrOZndEZzPnhGFKJhKWT2/PEqhMVgj7A/pgsejf2YUgrf9789TzL/opl/ayuRPi41NCzgK+uMprSKlMKIr2jCIvy1zr+GyGtRyoXBIv102H8CvvXyBTCwx55V2HFCMi6DFEPChEhvz9X2Z50DA4vgof/AD9hU8tqtXI4Lpspyw5XRKCeuprHT8eT+Oy+tvRv7oNS9jeMykVEREREROqJOk1vATh37hynTp2y+zh7trpruYh90goMPLv2ZLU1nMUKz649SVq+nZSLKlzNKWJ7dAYAU+4J473NF2zKx5bzwZaLjG8fzJ7LmcRlFnHwSpaN4FHO+ZQCtp1Px0kp58tdMRSUhfPGZBTxys9nmPHtUe5ffLDadb46NY39dDTQFCPb+Fil4FGVTc9DUbrjF+PsLYgK9lA4gV+LyudqV8Hfw6+lEElS1bnf2QsiBwjpK20ecDxel1mgUIOzD3SaWXm86TBBpMm3s5t2bgNkxzjuU0SkBs4m5+HhrKwId7/bCPNyJjVPT05R6c2eisgtgo9OTWM/Fxr76WoWPIAiuTvmiAH2G9WuGF3/mXl6Rr6BF348hdUK/Zr68sfZNBvBo5wd0el4Oqtwd1KQrzfx5q/nyS+pOT0EICrEHVVZGfosqyulTcfYP1GhEUxH934Eqx6ANQ8J3lRHlwtpL44iPVqMFe6j12Ixw8lVguAhkQj32b8+qn5eaRH88iQUCyV8U/P1PL3mZLWUW6sVXvjxFBn5du7zIiIiIiIiN5E6Fz369u1LmzZtqj2ioqLo169frfvbvXs3w4YNIyAgAIlEwoYNG2zaCwsLmTNnDkFBQWg0Gpo2bcrChQvr6NXcPDILDOTr7Uc2ZBWVUmAwkZJbwt5LGXx/MJ7Dcdkk55aQmF3MhuNJHEvIZfFD7ZnWLRwfFzVxDkxFS81CPq5OLSc6tYCVBxMczumHQwmkFxg4cMVxGPHey5mk5+upVgm5OBsyL9m/yGykOOkcVzIKybO3QFQ6Q//XwaeZ7XG5Gh5YCy43aCaq1sGQBUIEh0kPrcZXP6fXS0JoLwihvO2mCM73AA37CmZ5jjj+3Y3NQ0TkGs4k5TsO3b8LKI9wOZci+nqI1J5Ug4LYDq+DZ4Rtg9KZtOErOZZjv+IZQKnJTGJOMX+cTeWHgwmcTcojq9D2S3tagaHCt6NnIy82n3XsEbL9QjpdGgqmqbsvZdi/p12Dr07F9490YtGkduh0LsS0fBKLzzWpwnIVWcO/o0QbUj2V0tkbxn8Hvi2rl+j1bQF9/ms/krI4E06sFH52CxNMXx1Ey5B0RPD8ALKLSm2qvlWl0GAirRaiR1q+nkOx2Xx/MJ69lzNJsSMmiYiIiIiI/FPqNL0lNtZBusA/oKioiNatWzN16lTGjKm++/HUU0+xY8cOvvvuO8LCwtiyZQuzZs0iICCAESNG1Pl8bgVCPJwoNVkYvugvMqoszsK9nJk3vDmvbTxbsdCa2CEYd+ea3eYtVqAsU8ZSg1mqxWqtWPg5Ir/ExOM/HOPVYc1p5q+rks9cswlrfrGBsV/u596W/jzRLxIv7TULN9dgmLRBiKa4ekTw1ghsJwgeslr8GruFwuTfhH68GkHHGZB4UEhnadALtL62/iIufjBmiZATbS6t+XVYRCM3kdpjtVo5nZRHr8Z3b2qUn06NWiHldFIe90R43ezpiNxmWK0wfnUSHw1ZRiN5Gur0k5S6BJOpa85/t2fzaE/7f5tLTWYOxWYzfcUR9MbKSMjuEV68P751ZYSJzZ99SXVR32YuViRlN1Sr9Xp3PgGpRILJbOWJ1Sf4YHxrXliXwAdDvqCZOguntGOYtP5kubXkzd25jOvoRLPhv6DLv4gqJxqDe2PSNQ1xkwcTEtgAZh8W7pH5SRDUATwagIuvg8kiVCmDsvSgmg1Yb+zVgPUGz0vILuahJQdtNmW8tSpWPtKJRr7XTwsSERERERG5UepU9AgNDa3L7gAYPHgwgwcPdti+f/9+Jk+eTK9evQB49NFH+eqrrzhy5MhtLXp4aVVoVXK7IsMT/SKZ8e1RG8EDIDaziE+3X2Jy11A+2XYZgFWHExnfPpggdw1Xc6rvoChkEry0SvJLTET4uHBfxxCOOMitH9jcj+iUfNqHujs8p3WwG5/vuMyUpYfZOOce/N3Kdtg0HsLiK/tK9YukcgpdI8kuiufbA/F0buDJva38q5/n4is8QrvaHfuG0fkLj3KC2td8vtZbeBj10GIMHFlq/7yoB//ZvETuStLyDWQXld61fh4AUqmEME9nTieJZqYitcdVo8DdScnktYm4OykIcu9MTnEpV3PikUqgWYB9o+zUfAMPLz9SLfVzz+VMlv8Vx9MDIlHIZPjoVDgpZRSXmtkXk0n/Zn6sOWLfbLRXEx/e+f0CAPdEeOKquX6J29Q8PZOXHUJvtHAprZAG3loeWZ+ITiMn1KMTeSVGErKFKMzH+qgYvOwkns5O+Oq6kJavJ70gjsa+Wax8pBNe7qFCaueN4OQl+Gntegdy4qpHU1bFvw1GpSsKwMNJiaezkiw76WhOStl105EAcopLeXrNiWpRqBmFBh5efph1j3V1WFlORERERESkttR5esu/Tbdu3di4cSNJSUlYrVZ27NjBxYsXGThwoMNrDAYD+fn5No9bDR+dirdHV4a3quRSHugUwleT2tHQW8uLg5vQr2n1aiGH43JoH+phc2z5vjjmDW+O3I7x6fMDm+CslPPttI408HKiW4QnzQJ0DGjmy2f3R/H5/W35alI7Hu8TQa9G3vi7aXj53qZ2q8dM7BDMvphMTBYrGYUGrlYNU3XxhRGf2zVTy+v2XxYfK6h4/sXOy2RfL7e/KAOyYiDtLKSehvzkms+/Hlar0Ef6eci8LKTjGEsgJx7Szgn/SoB7nrBfpSViAHhF/rM53AXcDp+9f5szZV/078ZytVUJ93LmVGLuzZ7GHcud/NnzdlHx3thWKGVScoqNnE7KqxD5nxvYuHrkYBkHr2TZ9boC+PZAPBkFwn3Ix0XF/JGCd9QfZ1MZ2sofb5fqfXZp6EmRwURGoQFnpYxXhzavJnoUFRWiz7iCPuk0+owrFBcXcuJqLq4aBd/c34Rpzaz8MELH2vuCaRWg43RSXkV1tEd7NGD7hTTGtPJiyQgfPuqt5LvxIfw2pQGfD9DgUhgrGHnfKDIZtH1QiIC0WiB6k62PVTlyNak9/keGWQsIXl3vjW2FPS/1+SNb4GPnvbmW7MJSjsTZ3zy5mlNCpoP0GZHbn9dff502bdrc8PlxcXFIJBJOnDgBwM6dO5FIJOTm5tbL/G41pkyZUiflZktLS4mIiOCvv/7655O6jejVqxdPPvlkxfOwsDA+/vjjf3UOtR3z2s9IXf0OVOXZZ59l7ty5ddrnrc5tX1fsk08+4ZFHHiEoKAi5XI5UKmXx4sV069bN4TX/+9//mDdv3r84y9qjkEnp08SHjXPuYeneWMa0C2LVoQRmrTyG2WJFq5LzUJdQ7onwYt4v52yuVcqljG8fXLETdTQ+hxFtAlg9ozPfH0zk5NVcgt01PNQ1jMOx2UxcdACpBAa38Ofle5uyaFI7Vh1O5IV1pygqNSOXShjayh+D2cJj3x2jka+WxZPb88fZVPbHZOGpVTK6bRDZhaW8s/lCxTyyCq8RLgLbwYy9WPd9iiTpMGaXYDLazGbdVR2rTqZVnJZeYMDoYCGKuRTSL0BJNux4Syh9C6ALhHvfh7DuoKplWKyhEOJ2w2/PVpqUBneC/m/Ar08KQohcDR2mQ9e5MO1Pwfzt3AYhT7rTLAjrJpasvQFuh8/ev83Z5Hx0ajmezneniWk54V7ObDqTSl6xEVen6++Oi9SOO/2z1zJQx+9PdGPR7iscS8jF303NrF4RNPFzwdlBCdWrOY6ryRUaTJjK0j2Vchn9m/myYfY9LNx5ma93X2HhA23ZeymTTWdS0Shl3N8pBDeNgo//vMj9HYN5oFOoTVUWAH1OEsq9H6E4uUIw9ZarKW37ML06zqH3tAaot/4HWcwWsFrpoHHn6+4vs7lZJ9aeK+b+jiG0DHQlPSWelrFr0Wz4BiIHQtPhsGO+EKkBEBAlbDB4NxX8q66HazBM/U0w+D61GtpPwzzxB6QHv0RSmEZJQGcyW0zj+W15vDlaSBOSSiV0aejJr493Z+HOy5xPLaChtzOP9YqgobczSvn1K7eUGGtOB70RA1iRW4N9+/bRvXt3+vfvz+bNm+t9vK5du5KSkoKrq+vf7mP58uVMnTq14rmfnx/du3fn3XffJTw8vC6mWWf83//9X40pdTfK119/TWhoKPfcc0/FsaoltbVaLY0bN+all15i9OjR/3i8W5XDhw/j7Hx7bTLV1e9AVZ5//nkaNmzIU089dcv9ztcXd4ToceDAATZu3EhoaCi7d+9m1qxZ+Pv7OzROffHFF3n66acrnufn5xMcHPxvTfmG0arktApy4/mBjXli9QkOV9kVKTSY+GJnDI/1bMiAZr5sOSeIBiq5lNxiI90jvdh7KYPkPD3tQ91pHeSKu7OK5gGuFOiNnE3O56nVJ8pKAgq+Hr+dTiHQTY1aKefT7ZcrxjJZrGw4kUxynp65fSN4d3M0k5ce4rtpnfDTqckqKuWjrRdJybOtKFNt51quAp8mcO/7nItL5pdzOaz5OYesojSb09qFuqN1sEglJ06IyPh5VoWTPCCIFT/cJ5TVC+lcuzc69bRwbVUSDwqu+MM/Ff416WH/Z4KL/YD50P1Z6DBNKBeocavdeHcxt8tn79/kTHIeoZ7ONouPu5EG3sIu8qmkXLpH3r3+JvXFnf7ZU8plRPi4MG9EC4oMJlQKmeP7SBkdwjwctoV7OdtENLqoFbQJduOjCW0oKTXjpJIRFezGmHZB/HYqmQ3HkmgaoGNAcz/OpxQw4vO/+HB8G4a3EYy2C/KyUG99BcW5HysHMelRHvoCubEIqc4PLv9R2VaSg9OWZxk89HNym3Rl/m/nmNnJh0l5nyM/tUoop97mfuH+VNWLI/k4LB0EM/feeJqLazB0eZyMiLFsOJHKjhNG2vm9ineQhONpJn5bnoSrRoGzqvL9cFLKaRag471xrSg2mNEoZTgpb3xJ6apRoJJLMZjsb3D4uYqpLbXFbLFyKDab9AI9Pi5qOoZ7ILMXjlPHLF26lMcff5zFixeTkJBASEhIvY6nVCrx8/P7x/3odDqio6OxWq1cuHCBGTNmMHz4cE6cOIHsFiq5/E/Enap8+umnvP7669WOL1u2jEGDBpGbm8uCBQsYN24ce/fupUuXLnUy7q2Gt/ftt76oq9+Bqvj4+DBgwAC+/PJL3n333Trv/1bktk5vKSkp4aWXXuLDDz9k2LBhtGrVijlz5jBhwgTef/99h9epVCp0Op3N41Ymo7DURvCoyrcH4nmwcwhP92/ES0Oa8saI5mw6ncyiPVeY0DEEF7WMpwc0wgIU6I2oFTKMZitzVx2vEDyq0ibEnUW77fhuAIdis4nwcUEulWCyWFm09wp6k5kle2OrCR69GnnbDf8FkCg0uHj68/2J7Go5wXKphCf7Rgo7c/o8KEgDQ1nqi8kA534WTEWjHhTEhw7ThTK15Wx9DYrtv1d2Kc6Gra86aMsSKs74Nq88dnyFkFojkwklcG8FwaO0SHifypz1b2Vut8/ev8HppLy7PrUFwN9VjUYh49RV0dejPrhbPntqhQzPMk+smjCW6onyKOXT4cE0t+P58d97m9q9h5nMVsxWKyazFZlMSkJ2Me9sjmbflSyW7I3l4z8vsftiBqPbBlGgN5Jclmaj0mehOL/e7lykJ7+DyP4Q2LbyoEIDbR5AY9XTL0i4d/cIBPnpNUJ7q/FweLF981FDPpzdIKRtlpYI94dix1XXhElIkTh58dvlEvbFZPHpX2m8ujWVn05lUmq28Nygxvi6VBciNAo5nlpVrQQPEFKSHuluf3dxaCt/PB2kJInYZ/OZFLq9u537Fh3giVUnuG/RAbq9u53NZ1LqddyioiLWrFnDY489xtChQ1m+fHm1c9555x18fX1xcXFh2rRp6PX6aucsW7aMpk2bolaradKkCV988YXDMe2lt+zbt48ePXqg0WgIDg5m7ty5FBUV1Th3iUSCn58f/v7+9O7dm9dee40zZ85w+bKw6ffLL7/Qrl071Go1DRo0YN68eZhMJpvrFy9ezKhRo3ByciIyMpKNGzfajLFx40YiIyPRaDT07t2bb775xmbu9lJ9Pv74Y8LCwiqeX5va0KtXL+bOncvzzz+Ph4cHfn5+dsWMqhw7dozLly9z7733Vmtzc3PDz8+PJk2a8OWXX6JWq9m4cSO7d+9GoVCQmmpbreqZZ56hR48eFc8XLVpEcHAwTk5OjBo1ig8//BA3NzebaxYuXEjDhg1RKpU0btyYb7/91qb99ddfJyQkBJVKRUBAgE3ahcFg4Pnnnyc4OBiVSkVkZCRLliypaD937hxDhgxBq9Xi6+vLpEmTyMx0nOZ3bapJTWNfS0xMDCNGjMDX1xetVkuHDh34888/bc5JT09n2LBhaDQawsPDWblyZbV+8vLyePTRR/Hx8UGn09GnTx9OnjzpcNxrfwc2b95Mt27dcHNzw9PTk6FDhxITE2NzTVJSEhMmTMDd3R1PT09GjBhBXFyczTnDhw/nhx9+cDjuncZtLXoYjUaMRiPSa0I4ZTIZFsv1XMhvHy5nFNo9rpJLmTeiOQlZxfx+OoUV++M4k5TPmHbBlJrMdA73YNUjXfi/bZcY9fk+Ji89xOYzKeSVGMkvcVyFpaaw05S8EtzKQs+3nU/HSSnn1aHNcC87ppJLmdQ5lHfGtMKjhnD9IHcN62Z2oXODyp22Rr5aVj3amQidGeL3wepJsLgvrHsYrh4VxAyvJoLYcPUwHPxKEEBGfQXNRpZN8AQYHYcsV8NUAqmO/9CQesq2DKLFLKTW3AqUFgvz+2mG8D79MBFitl9/cStyy5BVaCA1T39Xm5iWI5VIaOjtzPGEWoiWIiK1xGqxYMy8gmXLazh/O5hhJ2awqt1FfpgYhkouJcBVzcIH2laLAskvMXIoNovHvjvKqM/38cQPx7mYVlDt97VbhBcLH2xLocHIFztjmPX9MbaeS8VanO24HKzFDIXp0GgQjF0GEf1hwneCr9Tejwn5cyZre2bhryiqFDk8IyGlhntX3C7ITYDfnhHuD9+NETYNCjMcXuLlomLhA+0Y0zYIhUyIEPDWCn4pA5r5Iq3DqAG1QsaUe8J5YVBjdBp52TEp07uH8+rQZjdkACsisPlMCo99d6za5lNqnp7HvjtWr8LH6tWrady4MY0bN+bBBx9k2bJlNmH4a9as4bXXXuOtt97iyJEj+Pv7VxM0Fi1axMsvv8xbb73F+fPnefvtt3nllVf45ptvbmgOp0+fZuDAgYwePZpTp06xevVq9u7dy5w5c2r1WjQawXjfaDTyxx9/8OCDDzJ37lzOnTvHV199xfLly3nrrbdsrpk3bx7jx4/n1KlTDBkyhAceeIDsbGENFhcXx9ixYxk5ciQnTpxgxowZvPzyy7WakyO++eYbnJ2dOXjwIO+99x5vvPEGW7dudXj+7t27adSo0XXFboVCgVwux2g00qNHDxo0aGAjUJhMJr777ruK1KC//vqLmTNn8sQTT3DixAn69+9f7T366aefeOKJJ3jmmWc4c+YMM2bMYOrUqezYsQOAdevW8dFHH/HVV19x6dIlNmzYQMuWlZ6GDz30EKtWreKTTz7h/PnzfPnll2i1QmRoSkoKPXv2pE2bNhw5coTNmzeTlpbG+PHjb+h9vN7Y11JYWMiQIUP4888/OX78OAMHDmTYsGEkJCRUnDNlyhTi4uLYvn0769at44svviA9Pb2i3Wq1cu+995Kamsrvv//O0aNHadu2LX379q343bkeRUVFPP300xw+fJht27YhlUoZNWpUxXff4uJievfujVarZffu3ezduxetVsugQYMoLa3cbO7YsSOJiYnEx8ff0Li3O/WS3pKWlsazzz7Ltm3bSE9Pr5aHZDbfeGnPwsLCCtUVhLK4J06cwMPDg5CQEHr27Mlzzz2HRqMhNDSUXbt2sWLFCj788MM6ez03G0cO5i8Oacq6I1fZf6UyzePbA/H8fDKJRZPao5BJGPXFvgqjtqTcEmZ+d4zVj3auiNa4FoVMIlSuc7A283BWUqCvFEwW7ozhz6d7MKiFH8WlZlRyKd4uKtR2jE6rIpFIiPR14csH25FbbMRsteKqUeClssLJ7+HXpypPzkuES1tgyu+QGQ3b36zeNuJzyIkVokJqU8JWIhfK3uY4KLfs4i8IKVVRam+8//ok8SB8N7pyEZyXCN+Ogj6vCGZ0qltkniIOOZMsmEk28BZFD4AIHy17LmcKZT/v8nQfkfrBlHUFxdJ+UFIpVrhsfZpOYT048uRCihVe+F6TWmEwmvnlVDIv/3Sm4lhSbgmXM4qY26dSFA/20PBg51AeWXEEo9lacd4jK45yce51Uk0kUtjxtuB9NfRjQagwly1O8xLxSZqGZfJvlecXZwleVkUORAz3cNjyXzi/saIP1jwEbSdDv9fByX5qj7+bhjdHNufJfpEYTBacVTJ8XdR1KniU46VV8Uj3BoxoE0hxqRm1QoqPixql/Lbej/tXMVuszPvlnN0iwVYE//V5v5yjfzO/ekl1WbJkCQ8+KFSuGzRoEIWFhWzbtq0ivfzjjz/m4YcfZvr06QDMnz+fP//80yba48033+SDDz6o8JEIDw+vEBomT5583TksWLCA+++/v8KsMjIykk8++YSePXuycOFC1Orrp0pdvXqVBQsWEBQURKNGjZg1axb/+c9/KsZv0KABb775Js8//zyvvfZaxXVTpkzhvvuE9Oi3336bTz/9lEOHDjFo0CC+/PJLGjduzIIFCwBo3LgxZ86cqSYK/B1atWpVMY/IyEg+++wztm3bRv/+/e2eHxcXR0BAQI19GgwGFixYQH5+Pn379gVg2rRpLFu2jOeeew6A3377jeLi4gpR4dNPP2Xw4ME8++yzADRq1Ih9+/bx66+/VvT7/vvvM2XKFGbNmgXA008/zYEDB3j//ffp3bs3CQkJ+Pn50a9fPxQKBSEhIXTs2BGAixcvsmbNGrZu3VrxO9WgQYOKvhcuXEjbtm15++23K44tXbqU4OBgLl68SKNGjWp8zTWNbY/WrVvTunXriufz58/np59+YuPGjcyZM4eLFy+yadMmDhw4QKdOnQDhM9K0adOKa3bs2MHp06dJT09HpVJVvEcbNmxg3bp1PProozXOGWDMmDE2z5csWYKPjw/nzp2jRYsWrFq1qsLjsnwttWzZMtzc3Ni5cycDBgwAIDAwEBB+P+qjAuutRr3cWaZMmcKxY8d45ZVXWLduHevXr7d51IYjR44QFRVFVFQUIHxYoqKiePVVISVh1apVdOjQgQceeIBmzZrxzjvv8NZbbzFzph0H8tuUIHdNNTd0dycFbhqFjeBRTn6JiZUHE9hzyb4z/Y/HkiryjK/lREIu/Zr62m3z06kpKTXb5OBO7xaOh5OSADcNET5agj2crit4VMXNSUmYlzMNvbWCw35hGvzxkv2TjcWw83/Vj1utsH0+dHwEuj0FWvvzt4uLr+DPYQ+pDBr0FKJOyglqL5T5u9nkp8DGOfZDm3e+7XghLHJLcSYpD2el7IaqHdwNNPTRklVYSnJe9fBnEZF/Sqm+EHa/byN4lCON240y51I1wQMEc+03fz1X7XhOUSnhXs50KvNOmNQ5jI+2XqwQPKpSovSwTZWsSmA7oRIZQNJRwWfKjjG2NPko+JctuE+thnYOvhBKJNDkXrhox1Ty2DdQkFr9eBWclHKCPZyI8NHi76qpF8GjHLlMWrF+CHJ3EgWPWnIoNrtahEdVrEBKnp5DsXUfARodHc2hQ4eYOHEiAHK5nAkTJrB06dKKc86fP1/NG6Lq84yMDBITE5k2bRparbbiMX/+/Grh+o44evQoy5cvt7l+4MCBWCwWYmMdbGghpBhotVqcnZ0JDg6mtLSU9evXo1QqOXr0KG+88YZNn4888ggpKSkUF1dGE7dq1ariZ2dnZ1xcXCp29aOjo+nQoYPNmDV9oa4NVccF8Pf3t4kmuJaSkhKH4s99992HVqvFycmJDz/8kPfff5/BgwcDwve5y5cvc+DAAUAQFMaPH19hBBodHV3tNV37/Pz58zbmqQD33HMP58+fB2DcuHGUlJTQoEEDHnnkEX766aeKNKJyf5WePXvanfvRo0fZsWOHzf9TkyZNAG7o96emse1RVFTE888/T7NmzXBzc0Or1XLhwoWKSI/z588jl8tp3759xTVNmjSxSfc5evQohYWFeHp62sw7Njb2hn/nY2JiuP/++2nQoAE6na7CiLR8HkePHuXy5cu4uLhU9O/h4YFer7cZozy6qerv9J1MvUR67N27lz179tSqJJUjevXqVaNjrZ+fH8uWLfvH49zKhHo6s3RKB6YuP0xGWRm3tqHu/BXjOGftj7OpvDumld22n08ksX5WV5JySjhY5Ubo6aykQ5gH97byJz1fz8kqufW+OhWLHmrH7O+PVxwb0tKPCR2DkcnqcJFSkCKE9F6LSgeFqWBx8McoPxl0QbUTPMppNBDaPwxHKm/UKDQw5AM4sqxSWPBqBGOWgLNn7ceoa0pyIO+q/TaLGTIvgsfd4cZ8O3M6KY8wL9HEtJyIMjPT4wk5BLppbvJsRO40rMU5KKN/cdguP70aIntXO56Wr0dvrBSYVXIpzw9qjJ9Ow6YzqfRs5M1jvRripJTx9u/n7fb9wu/JfDJmBcpV4yC7im+WVyPo/TKsn1557OImCO0Kp9bYdrLvE0z3/4h8/TTIuixENnaYDkeWVIZnypSCAffJHyojRa4ldjf4NnP4PojcPqQX3JhAfKPn1YYlS5ZgMpkqdotBCN1XKBTk5OTg7u5+3T7Kw/EXLVpUsTNezo2aiVosFmbMmGHXi6EmU1UXFxeOHTuGVCrF19fXpqKHxWJh3rx5dquYVBUPFArbNCyJRFLxmuxFLF77fUYqlVY7ZjRev3JRTePaw8vLi9OnT9tt++ijj+jXrx86nQ4fH1ux1cfHh2HDhrFs2TIaNGjA77//zs6dO21ez/VeY/n8rj2n/FhwcDDR0dFs3bqVP//8k1mzZrFgwQJ27dpV8aXcERaLhWHDhtk14vT396/x2uuNfe17DPDcc8/xxx9/8P777xMREYFGo2Hs2LEVKSPlr72mNZ3FYsHf39/mfSznWi8URwwbNozg4GAWLVpEQEAAFouFFi1aVMzDYrHQrl07u34iVY1cy9Npbkdz179DvYgewcHBdV5a526nRaAr62Z2ITG7mOQ8Pe1C3Ph2v+McLIVMitnB/4HBZOF8cj7/N7EN51MKuJJZiKeziggfLRdS83B1UnBfxxBm9VaSmF2Mt4sKq1Wo4vLR+NYk5+lp7OeCt1aFe12X2ZQ6+JW0WkB6nRxfqdzuDt510XpD39eg8yxhh03lAt6NQaoUFqQtxoB7GLgGgcs/dwyvE65XivB675XILcHpq3m0CXa72dO4ZXBzUuKrU3M0PoehrWoOxRURqT0SkDn+22iV2b+fXZsW8N7YVqw6nMj+GNtIy/8MbsJDXUJZYefevPlcGie7h9P8gV9QFSYiSzkBugDBg+mnGbZeTFKFIF5fS1EmeYnnuNhjBSGybOT58ahb9EbRZjpFV0/h5uKCwr+58Do31BDt6uB1itx++Ngxl/0n590oJpOJFStW8MEHH1SEypczZswYVq5cyZw5c2jatCkHDhzgoYceqmgvjxoA8PX1JTAwkCtXrvDAAw/8rbm0bduWs2fPEhERcf2TqyCVSh1e07ZtW6Kjo2vdZ1WaNGnC77//bnPsyJEjNs+9vb1JTU21EQFOnDjxt8d0RFRUFAsXLrQrUvj5+dX4OqdPn87EiRMJCgqiYcOGNlEbTZo04dChQzbnX/samzZtyt69e21+B/bt22eT8qHRaBg+fDjDhw9n9uzZNGnShNOnT9OyZUssFgu7du2yW5Gzbdu2/Pjjj4SFhSGX/72vtI7Gbtu2bbVz9+zZw5QpUxg1ahQgWDBUNQdt2rQpJpOJI0eOVES8REdH25jutm3bltTUVORyuY1h7Y2SlZXF+fPn+eqrr+jevTsgBBtUpW3btqxevbrCKNURZ86cQaFQ0Ly5gyjEO4x6ET0+/vhj/vOf//DVV1/9rf9QEfuEejoTWsXwcFTbIJY7ED6GtwkgIct+uJJEAu3CPPBz1bDrYgaL98RSoDfx8YQ2hHlpeeu38xyJz0Ell+LhrCS/xEhRqRmdRs6mJ3rQroYyf/8Yra9QjUV/TQWH0kKhRJ9CYz8SxDMCMqKh0YDqbTeCxk14eEXaHtf5QXAHe1fcXDQegjCTEV29Ta4Gz4b//pxEakVOUSlJuSWMigq8/sn/JoZ8oexl2llhR7ogVfj8WczClyW1q/BlzTMC/FoKjzr8EtXIR8sRB9WqRET+CVKtF6Ut70N5yH5lCHPrB+wuinx0anRqOfl6E+1C3bmSUVRN8AB4Z9MFlk/twOrDidVKsUolQj9Onp6YdT4Y4g+j+mmG3fuZtdkIJL89Ve04EglFni25f2kcKrkUdyd/coqTMJgsNPb155upHfB11Qj3z4Z94fKf1fsACO9u/7jIbUfHcA/8XdWk5unt+npIEMr/dgyv23Xbr7/+Sk5ODtOmTatWTnPs2LEsWbKEOXPm8MQTTzB58mTat29Pt27dWLlyJWfPnrXxZXj99deZO3cuOp2OwYMHYzAYOHLkCDk5OTZlth3xwgsv0LlzZ2bPns0jjzyCs7Mz58+fZ+vWrXz66ad/6/W9+uqrDB06lODgYMaNG4dUKuXUqVOcPn2a+fPn31AfM2bM4MMPP+SFF15g2rRpnDhxoqK6Tbnw0KtXLzIyMnjvvfcYO3YsmzdvZtOmTXVeXat3794UFRVx9uxZWrRoUatrBw4ciKurK/Pnz+eNN96waXv88cfp0aNHRRXN7du3s2nTJhth5bnnnmP8+PEVZp2//PIL69evr6h6snz5csxmM506/T979x0eRbU+cPy7Jbspm2x6r0BC6L13RUGUIijFBoKIvaPX370WVNSrXhBRRFEBUVA6KALSQXoLvQUIgXTSNnWz7ffHQMKS3SSEFJKcz/PsA8yZnTm7yTAz75zzvl1wdnZmwYIFxXkavby8GDt2LOPHj+err76iTZs2XLp0idTUVEaOHMnzzz/PnDlzGDNmDJMnT8bb25vY2Fh+++035syZU+5oobL2bUuTJk1Yvnw5gwcPRiaT8c4771iNsGnatCkDBw5k4sSJfP/99yiVSl555RWrESv9+/enW7duDBs2jP/+9780bdqUxMRE/vrrL4YNG2Y1NcaW69VYvv/+ewICAoiPj+df//qX1TqPPvoon3/+OUOHDuWDDz4gODiY+Ph4li9fzuTJkwkODgakIE6vXr3KHVFTX1TL5MlRo0axdetWGjdujKurK56enlYvoWqEeDozpnNIqeXBHk7c19KfDqHupTKgezg7sGB8ZwwmM8euZNEp3JOPh7fiy9FtcVYpcFEpaeKrQS6TRoQkZReSVyQ9cdIVGIlPl0qAFRpMXM7I59iVLM6l5JCRp7fbz1RdIaeTdRxPyCYhMx+jjTwjxVwD4MHvpaRuN1KopIDI0FlS1OZGDk5SYjbPcNDcISMxqpvGF4Z9Bw7OpduGzKzcNB+hRh1PlAJ7je6IcrUWKZfA5g/h98dg22dShSSFSiqjGTUQmg2RbqR8mkrlo8+tl8o9/zYG/pkOaaerpCeRfq6cTNSRX2R/Xq0gVIaDyhFZl2ekkXs3MbV5hBynEK5k5KO/qYKZn6ua6aPaIpdJ5VSXHbKeWujjquZf90Uz54mOyGUy5j3ZmftbWQ+tfm9wCylvFaBwUCFrOUwaPXhzP5oPJ9+jmc1s4rl9ppCv8kIhk1FokM7PhQYzCpmMtwZGk6s3kaIrlAKTAz+VHhTc7O73xfmhHlHIZbw3WJqqdPOA+uv/fm9w8ypPYvrjjz/Sv3//UgEPkEZ6xMTEcOjQIUaNGsW7777LW2+9RYcOHbh06RLPPvus1fpPPfUUP/zwA/PmzaNVq1b06dOHefPmFecpKE/r1q3Ztm0b586do1evXrRr14533nmnQtMb7BkwYAB//vknGzZsoFOnTnTt2pVp06bdUrLHiIiI4ryGrVu35ttvvy2u3nI9gWWzZs2YNWsW33zzDW3atGHfvn3FSUGrkpeXF8OHD7c53aE8crmccePGYTKZrEZrgJSbY/bs2UybNo02bdqwbt06Xn31VaspQMOGDWPGjBl8/vnntGjRgu+++465c+fSt29fQJrSMWfOHHr06EHr1q3ZtGkTf/zxB15e0lTyb7/9loceeojnnnuO6OhoJk6cWFyOODAwkJ07d2IymRgwYAAtW7bk5ZdfRqvVlqrsaUt5+77Z9OnT8fDwoHv37gwePJgBAwaUGhEyd+5cQkJC6NOnD8OHDy8uTXudTCbjr7/+onfv3owfP56oqChGjx5NXFwcfn7l/98sl8v57bffOHjwIC1btuTVV18tTpZ7nbOzM9u3byc0NJThw4fTrFkzxo8fT0FBgVVAbdGiRUycOLHcfdYXMks1zEMpr8xURbIx1ySdTodWqyU7O7vKo6vVLT1Xz7mUXObukkZrDGoVQMtAN77ffoFwbxeGtw9mw8lktp+9SqSvC493C+fNpUc5fDkLAJVCzujOIUT7u/F/K46hlMsY1i6I7o29eGPJEW4u8DL7sQ50jvBg3s44vtt+ofhpVssgN2aOaUeEd0nFEJPZwonEbF5cdJhL10aduDkq+ff9zbivZQBu9krSGQog8xLs/0G6kQpsB+2fAPcwMBsg4yLs+156Cu3fClo8KF3cuQWC6k64gawhJiNkx0PMIojfDZ6NoPPT0gV9HancUpePvds1a2ssMzfF8sNY6Uap1iQfk3ICXD0nHUNBHcC3BTiVvpi1YrFIZTZTjksjQ/LTwbcZtH1UOmYrKT4jn7eWHWXhxC50b3wHJA2upxrysVeUEY85dguOp5ZhUWtIb/EkWzN9eHNtAiqlnBfvasKYzqF4upQkGC4wmLiSkU+KTioDmqOXgnLBHk5MHdaSz9af4cS1akxqpVR6NdLXle3n0niyezhhXi5W57zcAgP6zCvI4rbjGbscs8KJ9BbjkAe05KMtqTzfXo3PpTVoE7ZjcAkgtfk4Fl9Q0Sw8mCZ+rizce4lTSTlE+blydzNf5u6MY8uZVII9nJgxuh2tA91wyL0CJ1ZC7AbpgUKXSdIILSf3mvy6hRqw7ngSU/44aZXUNEDryHuDmzOwZeVv/oWqNXXqVGbPns3ly5drfN/Hjh2jf//+xQkub8XEiRNJSUlh9erVFVr39OnT7Nixo7JdFWrAmjVrmDx5MkePHq301KC6plqCHnVNfbj40xtMmMwWnNVKzGYL+QYjKoUClVKO2WyhwGBCV2hgzPd7iLMx7eXp3o24kJbLxlNS9udHOoeSX2RiZUyC1XobX+vNjnNXmfJH6Uz2AVpHlj/XnQCtNEwqPiOPgV/uIL+o9NzkX5/qQo8m5dzQmAzSE2WlE9w8RM1kgKICqcKKUlXmPO16z2ySAkVKdZ37HurDsVdZz/5ykEvp+bzzQC0lFCzKkYKHsZtBGwKR90g3RJUJwFjMkHoGLmyRSmMGtYfOk2w+yS6P2WJh0oKDPNUrglf6l11uTqi8hnzsXZeVlcm32+NYdvQqV3Otk37+d0QrRnYMKTX/XVdQxOtLjrDhpHSu/HpMOz748ySpOaVHO344tCUjOwajtlHRbO+FdEZ9v4dof1d6hTlTaIL1Z7LQFRr44YmOPP7TPrpEeNDGV0V6oYV1pzLI1Rv544UetAp2x2gycz4tly/+PsuW06lWJejVSjnrX+lNuLcLmM1S5TOFSjpXCvWWyWxh38UMUnMK8XWVprRUR5laoeJmzZpFp06d8PLyYufOnbz44ou88MILFZ4iU9Xmz59P+/btadWqVYXWz87OZv/+/QwZMoRVq1bZLIn7xRdfcM899+Di4sLatWt5/fXXmTVrVnGZYuHOtHjxYsLCwkolEK7Pqi20c/78eebOncv58+eZMWMGvr6+rFu3jpCQkAaTMKU6WSwWUnV6CgwmlAoZ/m6OxRdWcrkMjbrk5lcul+GiVnIqSWcz4AGwcG88Hz/YsjjosezQFb4c3dYq6DG0bSBOKgUzN8fi6aLi4Q7BNPV3JafQyOojiRy8lElsSm5x0OPvEyk2Ax4An607zbzxnfFwLuMiTOFQ+ibebIa8VOlGX60BdQVHduSng7FQShBnoxRgnSZX1JmRHUKJo1eyaRfqXjs7TzkhTWEx5EmJeoPal55SditkcvBrBr7RkHJSqj6x6nloPRJajbylYJxcJqOpvyt7LlR9iUVBuNH+RD3f7Uq02TZ9wzn6NPXF3806AaSbk4o37m3KvosZjO4UikIuIzVHT6cwdya0dUHrKGd/YhFzD2bw1aZz3NPcF3+t9Xzpyxn55BQaeef+ZoRpZbT2tGC0gJ+7C19uvsDBS5l0Dvdkz4UM9txQ6KVHYy8Cr1U1yi8yMeWPk+yykVtEbzSz4nACr/SPRCaXi/NDA6GQy+jW+A6oLicUO3fuHB999BEZGRmEhoby+uuv8/bbb9daf251pP3QoUPZt28fkyZNshnwANi3bx+fffYZOTk5NGrUiK+++koEPOqAkSNH1nYXaly1BD22bdvGfffdR48ePdi+fTtTp07F19eXo0eP8sMPP7B06dLq2G2DcTW3kITMQr7ZEsuOc1dxVikY1SmEx7qGFV8Q2XI+LdduW67eiPKG0rN6o7n4CYG7swMTe0YwsmMIuUUmOoR5MLJjCD/tvMi8XXF4uagYeW3/Z1J09IrywWy2cOCS/WSE51JzKTTYDojYlZMMx5bCnm+kTPfhPaWqK95R4GAnM3nhtaSMG9+H1BPgHgp9/gWN+t0ZpWeFBik9V09CVgEj2tdCEtMza2Dvd9Kx0GlC1Q51l8nAvwX4RMH5LXD0N4jfA33elEaTVFAzfzeWHLyM3mhCraxY2UJBuFXHE7LttiXrCiky2s4/FeblzO+TurEqJpFjCdksHBNO8/SNuO/+DvLT6RzSjYdHvs2U3UarhKbJ2QXsj8tk5uZztPTX8HE/DYrt/8Uhdh0o1Uxs/ThjX5nA1H+yePeB5rzyewznUnPRqJU80S2Msd3D8bqWF6SgyMTZlBy7/Y+5nIXeaMbRxigTQRBqxvTp05k+fXptd6PSbJVVvdnixYvLXUcQ7gTVEvT417/+xUcffcRrr71mNW+sX79+zJgxozp22WCk5+q5kJbH2J/2U3AtaFBgMDFr63m2nE5l7vjOpZ5MXRfqaSPp5TWODnLMN8x0UsplhHg4s+n1PqiVcgK0TijkMsy6Qoa3C2LSggPF+T4Sswv5cuM5+kb58MYAaTi6XC6jRYAb644n29xfmJczKuUtPFnOTYOVz8H5TSXLYjfCha0wfj0E28h2bDbB2XWw/IYkPVfPwbIJ0PNV6PWGeAIm1Ipj1262GvnU4O+fxQwHfoITKyCsOzQdJI0Sqg4KB6mSkl8LOLYY/ngZur8EjfpW6O3NAlzRG80cuZxd5VUHBOG6SD/789q9XFQ4KGxPDUjMKmT4rF14OKuYP7oRodteR3WxpFKK4uIWAi5t54OHVsG1oJ2u0MCSA1f434azuDkqWTUmAMe5d4H+WuCiKA/1nhk4xK7j3yOX4OqrZdHErhQYTCjkMnxc1Tjc8GBC7SAn1NO51LSc66L9XVEpqiVXvSAIgiDUOdVyRjx27FhxDeMb+fj4kJ5eeiimUHGZeUX8+E9cccDjRqeSczh+xf6Tq3BvF/zc1DbbRrQPtgpQDGoVwILdcTjIZQR7OBeP+pDLYMamc6USnAJsPZtmtXxwm0C7F12v9o/Cy8V2X2zKircOeFxnNsLat6TpKzfLSYJ1b9ne3s4ZkJdW8f0LQhU6diUbF7UCX9dbOAZuh9kE/0yTkho2Gyy9qivgcSNtEHR9Hnybw/bPpYSplvJHeIV7ueCiVrDr/NXq76PQYLULccfN0fazn2f7NsbPtfQDhHy9kRmbzpFfZCIzv4gg2VWrgEcxswnfHf/BWyGNsLyao+frLbEAvDMwAuXO6SUBjxvIr57BIekgAN6uakI8nQl0d7IKeAC4O6t49R7bOW+UchkjO4YgF/kcBEEQBAGopqCHu7s7SUlJpZYfPnyYoKBaGM5dj2QVGPjnnP2b9VUxCdjLTRugdeKXp7oQ7mU94uO+lv50ifBk7bWgR9+mPgxqFcDSgwnEpuURn5GP+Vo0I09v4nSy/SG1+y6WzMMP9HDk5wmd8XAumc/voJAx+d6mdAq/xae3tgIe1yUcsHnxSEGmNA3GFotZqv4iCLUg5koWjb01pZIkVguzCXZ8Dhe3QZtR0iiPmqRUQauHIfoBKeiy9VMw2X46fZ00UkzLP+dE0EOoPgFaJxZO7EqAtiS4IZfB2G5hDGsXZDNooCs0sOW0lPsq0leD8dwW+ztIikFeJJ2bUnP0xVNdugUocDi/3u7bHE8uwWQs+xgBaBWk5d0HmqO+YdSkm5OSH8d2IsjD/lRXQRAEQWhoqmV6yyOPPMJbb73FkiVLkMlkmM1mdu7cyRtvvFGqxrNwa2QycFYrybOTINTNyaHMG6lIX1cWT+rG1dwidIUGPJxVpOoKKTKZ+d/DbdA6OXDgUiYv/3aYIpMZvdHEA1/tYNHErrQI0qKQy5DLsDnSA8DVsSTAoVIo6BTuyZqXepGWo0dvNOGvdcJbo8JZdYu/eo5lVBdQOIDMxlNreTnJE1X2p/sIQnWxWCzEXM6iVxOfGtiZCXZ8AZd2QptHpOkmtUEmg/Ae4OQJRxfBhveh/ztSZSY7Wga58fPuS+TqjWjUDaOcmlCz5HIZLQLdWPFcD67m6skrMuLv5oiXiwqNo+3zh0wmw0mlQFco5esoUpYxRU2uLB5RdeN0ziIzUnl1WyMUAZNaa/ucdhN3ZxWPdAnlnuZ+JOsKcZDL8HVzxNdVbZWjSxAEQRAaumo5K06dOpXQ0FCCgoLIzc2lefPm9O7dm+7du/Of//ynOnbZYPhrHRnaNtBu+8iO5ScL9HVzpHmgG10beeHh7MC7q08wYf4B3lx6lCfn7eebLbHojWY8nB3QG8zoCo28uOgwaTl6PFwcGNDC3+Z25TLoctP8e4VcRqC7E21C3Okc4UWop/OtBzwAGt8tlWS1pcUIcLaRlNTZC3yb2X6Poxbcbr2cpiDcrqTsQtJzi2jsW8HKQ5VlMcPOryDuH2hdiwGPG/k1gw7jIO00bPqwzBEfLYO0GM0W9tioTiEIVUUmk+GvdaRlkJYuEV6EebnYDXgAeDo7MKZzKABnUnLQBfQoKfMsk0mlYa9rPgycpHOTl4sKn2tJSBcczaew3Xi7+zC2G4fi5jLtdjg6KAjxdKZTuCdtQz0IdHcSAQ9BEARBuEm1nBkdHBz49ddfOXv2LIsXL+aXX37h9OnTLFiwoMIncsE2DycVD7YLonlA6ZEPE3s1wkEh4+LVXLILDBXanq+bIzPHtEOjVlJkKskyr1LI+WBoS374R5oCcuFqHhl5ejRqB94e1MxqOPB1n45oja+bGiwWyE6A1NNSBYfL+6ScHGbbmfDLlZMChnwYuQAeXSIlIb1eAtM9TKoMkXwUruyH7CtgMkptGh8Y/oMU4LiRwgEeng+uAdK62Vek917aBZmXoMh2WV9BqApHLmcB0Lhak5haYP+PELsJWj0kVVS5U3g2gg5jIfUkbPuv3Rwf/m6O+Ls5svVsag13UKj3ivIgM076P//KQekcYK5ANbG8NORXT/NMWBJ7ngriw3sCmHe0gJz7v4ehX8OoX6Q/H10CvSfD3e8Ul1UP8XDiqzHtUCvlzN0dT0H0cEyBHUrtQt/haSyejaEwG9LPS0HLpKNSjqpbVai7YRtHQFeJbQjCbZLJZKxcudJu+9atW5HJZGRlZdVYnwRr48aNY9iwYbe9naKiIpo0acLOnTtvv1N1SN++fXnllVeK/x0eHs6XX35Za/25E8ybNw93d/cq296xY8cIDg4mLy+v0tuo1jHDjRs3pnHjxtW5iwbHWa0kzMuFrx9px4lEHRtPpqB1UtK7qS8H4zIZ9NU/AAxtG8i/BzXD104llxs1D3Bj7cu9+PtkMofjswj1dKZzhCffbb/A8QRd8XoGkzSnJdTTmWXPdufgpUw2nEwmyMOZ4e2CCHR3wlkplyqkHPwJ9s2REo0CuPhIQYvgTqC4hV+7q+dg0WhIjy1ZFv0APLYC8q6Cewj8PBSyL0ttTh7w0FwI7SaVsfVrAZN2SJVe4v4B/5bS0zdtsNS3+F2wdLyU/wOkp3R3vwdtHwFnUTVCqHoxl7Pw0qjwcFaVv3JlHf0dTq6UftcD21XffirLM0I6xg4tgP0/QeeJpVaRyWS0Dtay+XQqFoulZvKfCPVffjrs+wG2f1ZyfnL2gpE/Q0hn65EaN8qMhyVPoEg8jALwBx4J70XGgFk4mdWw+FEpeAIgk2FuNxaj3JHrW5PL5bQP1fLXS71YezyJz3dm8+bQ+ThmnkZ1YjEmBw2Wto9i0YbhKDfD2n9JU8Gu5+jyCIcxv9kfvXiz3BTY9BHELCjZhnvotW00LxmdIgi3ITk5malTp7JmzRoSEhLw9fWlbdu2vPLKK9x9990V2kb37t1JSkpCq9WWv3IFvP/++6xcuZKYmJgq2V5DMGPGDLv5AG/F999/T1hYGD169CheduO5W6PR0LRpU/7v//6P4cOH3/b+7lT79+/HxaWaR/OWIS4ujoiICA4fPkzbtm2rfX/h4eG88sorVoGfUaNGMWjQoCrbR6tWrejcuTPTp0+v9KyRagl6WCwWli5dypYtW0hNTcV80xP+5cuXV8duGwyNWonGR0MjHw09m3jx4Z+neHHhYauKLqtiEvFxVfPmgOhyS8PK5TJCPJ25v2UAMfFZ7L2Ywayt563WcXNS4ulScjEY6O5EoLsTg9vcNNVGlyQlHd3zrfXyvDRYMAye2yPd8FSELhEWPFgS0Lju9J/SKI0O42B2D+u2gkz49SF4bi94N5Eu7DzCoNME6XWjq+fg14dLLnxBGm7/97/BNxqa9K9YPwXhFhyKzyTStxpHeZxdLwUTmvSH0C7Vt5/b5RMN0fdLwRnPRtCk9AVy2xB3/j6ZQmxqbpnlRQWhwi7+A1s/tl6Wnw6/DL92fmpU+j15abBkLCQetlosj9uBV/4FZEvHlQTOASwW5IfmYXELxdTzJRRKaWSi2kFJY18NL9wVWbKuXwhE31My7NZkkJL9Hllo3YfMOJg/GJ7eKgXty2IywMF5cPhn6+VZ8TD/AXh6u/TAQBBuQ1xcHD169MDd3Z3PPvuM1q1bYzAYWL9+Pc8//zynT5+u0HZUKhX+/ranTVcng8GAg0M5ud8aiKoKOM2cOZP333+/1PK5c+cycOBAsrKy+Pzzz3n44Yf5559/6NatW5Xs907j41MDOdtqSGWPEycnJ5ycqjah9pNPPskzzzzD22+/XamZI9UyveXll1/m8ccf5+LFi2g0GrRardVLuDV5eiMJmQVcycxHd9O0lVNJOayMSaBbYy+mjWzDN4+0550HmtHEV8Ove+JJzSksc9tFRjMJmflcSMtFbzIztns4YGFs93C+eaQ9sx5tz4SeEXwwpCW+rmoy84q4nJFPYlYBhTbK5qJLgH3f296ZsRDOrqv4B8+8VDrgcd3hn+0P9zUb4cgNT8hssVgg5lfrgMeNtnxsv/KLIFSSwWTm2JVsmvhU0w18/B7Y/TWEdoXGd1XPPqpSaDcI6iD1OSu+VHOLQC2ODnL+PplSC50T6hOz2UJ+RhJsnWp7BaMeTv5huy3vKiQeKr1cG4IsI9Y64HED9d6ZGLOTS87hGfnlTz3NSYa9s+30Iw3SzpRanF1QxJWMfBIy88nXG6VRHru/sb2N/AyMycfRFZRfHUaoY8wmuLgDji2V/qzIlK3b8NxzzyGTydi3bx8PPfQQUVFRtGjRgtdee409e/ZYrXv16lUefPBBnJ2diYyMZPXq1cVtN09vuT4sfv369TRr1gyNRsPAgQOtqkJu3bqVzp074+Ligru7Oz169ODSpUvMmzePKVOmcOTIEWQyGTKZjHnz5gHSiIPZs2czdOhQXFxc+OijjzCZTEyYMIGIiAicnJxo2rQpM2bMsOr79akfU6ZMwdfXFzc3NyZNmkRRkf1j6PpnWLlyJVFRUTg6OnLPPfdw+bL1Ne0ff/xBhw4dcHR0pFGjRkyZMgWjseS6VCaT8cMPP9j97gBWr15NZGQkTk5O9OvXj/nz51t9n++//36pp/1ffvkl4eHhpT7jdX379uWll17izTffxNPTE39/f5vBjBsdOnSI2NhY7r///lJt7u7u+Pv7Ex0dzezZs3F0dGT16tVs374dBwcHkpOTrdZ//fXX6d27d/G/58yZQ0hICM7Ozjz44INMmzat1NSJb7/9lsaNG6NSqWjatCkLFiywan///fcJDQ1FrVYTGBjISy+9VNym1+t58803CQkJQa1WExkZyY8//ljcfvLkSQYNGoRGo8HPz4/HH3+cq1ftV5e7eXpLWfu+2fnz5xk6dCh+fn5oNBo6derExo3WZdHDw8P5+OOPGT9+PK6uroSGhvL99yX3XhER0sPldu3aIZPJ6Nu3LyCNQLnnnnvw9vZGq9XSp08fDh2yPrfZOk5A+j3r2LEjjo6OeHt7F4/U6du3L5cuXeLVV18tPubA9vQWe9sA+OWXX+jYsSOurq74+/vzyCOPkJpqPb15wIABpKens23bNrvfX1mqJejxyy+/sHz5ctauXcu8efOYO3eu1UuouLireby59Ci9P99Cz/9u4fmFhziTnIPxWv6Nyxn5zBzTnnAvZ95bdYLnFx5i4d54JvVuxIj2QRQa7OfRSMoqYOpfJ7l72jbu+t82nvv1EJn5RXwyvDWJmQW8uOgQzy88xPnUXFqHaDmWkM2T8/bR67Mt9PtiKx/8cZKErALrjZpN0hMpe5KPVfzDZ12y32bUS0EUe1KOSU+77DEVQcoJ++2ZF8veviBUwpnkHAqNZiL9qmGkR8oJ2PYp+DWH6MF1Y/i6TAbNh0jT0rZ/ViqxqUopp3WQO3+fTLazAUEo39VcPXN3XeTklfSyz09JMbaX2wlq4OonBeftKcjEYtLz1rJr5/DPtvDcrwc5nawrPoeXYiyEolz727xhqqfBaOZUko5nFhyk52db6PP5Vt5ecYwifYGUE8SOnMsn+fiv05xIzKbIWL03xkINObkavmwpjeRZNkH688uW0vJqkJGRwbp163j++edtDuO/+WZnypQpjBw5kqNHjzJo0CAeffRRMjLsP1jKz8/niy++YMGCBWzfvp34+HjeeOMNAIxGI8OGDaNPnz4cPXqU3bt38/TTTyOTyRg1ahSvv/46LVq0ICkpiaSkJEaNGlW83ffee4+hQ4dy7Ngxxo8fj9lsJjg4mMWLF3Py5Eneffdd/u///o/Fixdb9WfTpk2cOnWKLVu2sGjRIlasWMGUKVPK/I7y8/OZOnUq8+fPZ+fOneh0OkaPHl3cvn79eh577DFeeuklTp48yXfffce8efOYOtU6MFvWdxcXF8dDDz3EsGHDiImJYdKkSfz73/8us18VNX/+fFxcXNi7dy+fffYZH3zwARs2bLC7/vbt24mKisLNrYxqi0h5H5VKJQaDgd69e9OoUSOrAIXRaOSXX37hySefBGDnzp0888wzvPzyy8TExHDPPfeU+o5WrFjByy+/zOuvv87x48eZNGkSTz75JFu2SCXFly5dyvTp0/nuu+84d+4cK1eupFWrVsXvf+KJJ/jtt9/46quvOHXqFLNnz0ajka7TkpKS6NOnD23btuXAgQOsW7eOlJQURo4cWaHvsbx93yw3N5dBgwaxceNGDh8+zIABAxg8eDDx8dYPhv73v//RsWNHDh8+zHPPPcezzz5bPLpq3759AGzcuJGkpKTiGRY5OTmMHTuWHTt2sGfPHiIjIxk0aBA5OTlW2775OFmzZg3Dhw/n/vvv5/Dhw2zatImOHTsC0uyN4OBgPvjgg+JjzpaytgFSPpgPP/yQI0eOsHLlSi5evMi4ceOstqFSqWjTpg07duyowDdfWrVMb9FqtTRqZGOIqHBLrmTm89DsXVzNLbkR2HHuKsO+2clfL/ckwltD21B3PvjzJDtjSyocnE/LY/LSo3z8YEvcnGz/iFN1hTz18wFOJJbk7DiRqOPpBQeZ9Uh7TiXrisvSHk/M5mJaHk8vOIjp2kK90czCffHsvZjOr091wV97bQiTwgF8mkLqKdsfKqRTxb8AzzLywahcQFlGvpLgzqAsI2eCUi3lFzn3t+12n+bgIEraClXrcHwmCrmMcK8qnuuZGQebpoA2FFqNBHkdqt6gUEHrUbDnGzjyG7S3LmveMdyDWVvPk5xdiL+NBMqCUJacQgMzNp5lwZ54Pro3gI4+0VJST1vC7Ay1tlUdDKQcHt6RttsANL4UmJX8ebTkInBnbDrDvtnJmpd62U5m7OAkBQHtBVp8mxf/9VJGPsO+2YneKAVQjGYLq2ISeax5KJ2cveyWxC30asbGvSksP5TAny/1JEpMHavbTq6GxU8AN41u1SVJy0f+LAWXq1BsbCwWi4Xo6OgKrT9u3DjGjBkDwMcff8zMmTPZt28fAwcOtLm+wWBg9uzZxXkBX3jhBT744AMAdDod2dnZPPDAA8XtzZqV5LrRaDQolUqbU2YeeeQRxo+3rpx0Y/AiIiKCXbt2sXjxYqubWpVKxU8//YSzszMtWrTggw8+YPLkyXz44YfI7ZxvDQYDX3/9NV26SNNM58+fT7Nmzdi3bx+dO3dm6tSp/Otf/2Ls2LEANGrUiA8//JA333yT9957r0Lf3ezZs2natCmff/45AE2bNuX48eOlggKV0bp16+J+REZG8vXXX7Np0ybuuecem+vHxcURGGi/uiRIIyo+//xzdDpdcc6XCRMmMHfuXCZPngxIN8f5+fnF3//MmTO57777ioNeUVFR7Nq1iz///LN4u1988QXjxo3jueeeAygebfTFF1/Qr18/4uPj8ff3p3///jg4OBAaGkrnzp0BiotubNiwgf79pWntN97Dfvvtt7Rv356PPy6ZFvnTTz8REhLC2bNniYqKKvMzl7VvW9q0aUObNm2K//3RRx+xYsUKVq9ezQsvvFC8fNCgQcWf96233mL69Ols3bqV6Ojo4uk1Xl5eVsfBXXdZjwD+7rvv8PDwYNu2bTzwwAPFy28+TsaMGcPo0aOtjpXrffT09EShUBSP0LBn6tSpdrcBWO2vUaNGfPXVV3Tu3Jnc3NziABRAUFAQcXFxdvdTlmq5Mn7//feZMmUKBQUF5a8s2GQymTmfmkvnCE/UN+XkKDCYmLP9IoUGE0VGM6eTcni4YzCPdw2jZVBJhPXrzbHFQYqbnU/LtQp4KOQy7or25Ylu4RyKz+SxrmHFbaM7hfD99gs2t3U+Lc9qO7gFQtfnbX8oRy00uoUh9+4hUgDFli7Pgqud/1wdnKDliPK33+ohaV1b7voPOLlXqJuCUFEHL2US4e1Sbp6dW5KbAn+/Ix1f7R4rqWxUl7gFQKN+cGwJZFy0amoX6oFSLmPtcVF5Qrh16blF/LJXekI2e38WqZ3/ZXtFtStEDrDZZHTywhLRt3RDTjIW9zBwtX2hp+v6Bml4WC1zUMjoG+nJmfgk9Ho9APl6Y8l0UU0A9Hzddh/dw8BLusnLLzIyc/O54oDHjb7cm0NRjzdsb8MtiHhFCFdziygymZmx8Rx5ejvTPIU7n9kE696iVMADSpat+1eVT3W5nvSyogmmW7duXfx3FxcXXF1dSw1dv5Gzs7NVIYSAgIDi9T09PRk3blzxE/AZM2bYfbp8sxufLF83e/ZsOnbsiI+PDxqNhjlz5pR6qt6mTRucnUsehHXr1o3c3NxS01VupFQqrfYXHR2Nu7s7p05JDwUPHjzIBx98gEajKX5NnDiRpKQk8vNLqgiW9d2dOXOGTp2sHyaWdUN9K27cL1j/DGwpKCjA0dH2g4kxY8ag0WhwdnZm2rRpfPHFF9x3332AFNSJjY0tnhL1008/MXLkyOIRRGfOnCn1mW7+96lTp6ySpwL06NGj+Lt++OGHKSgooFGjRkycOJEVK1YUTyOKiYlBoVDQp08fm30/ePAgW7Zssfo5XQ/2nT9/3uZ7blTWvm3Jy8vjzTffpHnz5ri7u6PRaDh9+nSp38kbfz4ymQx/f/8yfz4AqampPPPMM0RFRRWnnMjNzS217ZuPk5iYmAonJranvG0cPnyYoUOHEhYWhqura/GUnJv75uTkZHV83IpqCXo8/PDDZGZm4uvrS6tWrWjfvr3VS7DPaDITdzWP/204y7fbzuPkoGDWo+0Z2tb6Bn/7uTTy9EYKDWb+80BzsvINxKblMrBFAN8/3gE/NzWJ2YXkFdo+sHacK5mL1q2xFz+N7UiguxNnknVkFRjo2siL+1r607epDw+0DmR/nP1hiBtP3TDf3tUfwrpD/ynSReR13pEw9g/Q3kLyNFd/eGQJhHYvWaZwgE5PQdtHpRul4XOkp2LXeYRXfD/aUGld95IAD04eUplbvzuoxKdQb+yPyySqKpOYFmbD3/8BGdB+nFSxqK6K6CNVedrzDTdewGvUSloHa62elgtCRSVmFRSnd7qSWcCiBB+y+k+zLmXu2QjG/lkqQWhqTiGbTqXw/PI40u6ejiXqvpJpYzI5llYjSXWMIGvkCvC/4QbBwQlj7/9jt6oHhcaS3+X7m3mw+YkApmsXM+jYKzhsfp/chFPM2nSS5xceYvvZNK7mG7G0GY255+vSiMTrgtpjemyF9GAByCk0sivW9kiOneczOeh6F5Y+b1mPiAxsR/zg33j1r9Qb1r1KTmHFStwLd6BLu6Sk73ZZpFxrl3ZV6W4jIyORyWTFN5XluTkRokwmK1XkoLz1b6wuMnfuXHbv3k337t35/fffiYqKKpVHxJabp+IsXryYV199lfHjx/P3338TExPDk08+WWa+jpv7davt15eZzWamTJlCTExM8evYsWOcO3fOKnhQ1ndnq7LZzVVY5HJ5qWUGQ/nH/K3+zLy9vcnMtD1Cbfr06cTExJCUlERGRgavv14S2PX19WXw4MHMnTuX1NRU/vrrL6un/hX5jNf7d/M615eFhIRw5swZvvnmG5ycnHjuuefo3bs3BoOh3GSbZrOZwYMHW/2cYmJiOHfunFXeEXvK2rctkydPZtmyZUydOpUdO3YQExNDq1atSv1O3urPB6QA08GDB/nyyy/ZtWsXMTExeHl5ldr2zcdJVSQkLWsbeXl53HvvvWg0Gn755Rf279/PihUrAEr1LSMjo9KJYqtlesv1L/Wxxx7Dz89PlBq8BccTshn1/R6rpzfLDycwZUgLMvOK2H4tWOHu7ECRyczSg5f5/cCV4nV3n08n2MOJz0a05ukFBzCYLZjNFuRy65+Bj6t0MRXlp+GxLmFM/PkgRdfmGO+5kMGyg1f45tH2mMwWjiVko3VyIDPf9gHq63rTjZZXY+mJc/QgaYiu0km6mXELuPUvRBsCgz6ThhEb9VLQ4/Qa+KYTjPldKskZ1l0axitTgouX3adupSiU0hSXCX9DXjpYjNIwZk0AVCIrsCCUJUVXSEJWAQ93KKfyQkUZ8mHDu6DPgc6TwLGOD1FXKKHZENg/By5sg0Z9i5u6NvJi1tbzJGQVEORetdnAhfrNRW19mTP9n1RiIpvx0v1r8JTlEODphsrVR8rPcYO0nELeXHqUrWfS6BTuwY5UFUnuk7lv1Ns4GHPRK105nqXGM8eRmMt6/Jt9Rec+ZhQmPTlyVxad1HNvsB/Ka+fejqFuTGmdgffvTxQ/dZfH/YPmwBzGDl3EuFg1T/y0jzGdQhjXI4JfswcxeviDaMw5mBSO7EqWYYlVMMzViEatRCGXoXV2IC1Xb/NzLzujp/PQV5C3fZSMqynojEr2pMiY9nuq1Xs8nFUo6tJ0OMFabgWTPFd0vQry9PRkwIABfPPNN7z00kulbpKysrJK5fWoau3ataNdu3a8/fbbdOvWjYULF9K1a1dUKhUmU8VGtuzYsYPu3bsXTxMA20/vjxw5QkFBQfGN2549e9BoNAQH2z+fG41GDhw4UDwq4cyZM2RlZRWPEmjfvj1nzpyhSZMmFf7MN4uOjuavv/6yWnbgwAGrf/v4+JCcnGwVBKiOcr7t2rXj22+/tRmk8Pf3L/NzPvXUU4wePZrg4GAaN25sNWojOjq6OEfFdTd/xmbNmvHPP//wxBMl02N37dplNe3JycmJIUOGMGTIEJ5//nmio6M5duwYrVq1wmw2s23btuLpLTdq3749y5YtIzw8HKWycrfN9vZtayDAjh07GDduHA8++CAg5fi41ekcKpU0vf/m42DHjh3MmjWruJTs5cuXy0zIel3r1q3ZtGlTcZ4VW/sr75graxunT5/m6tWrfPrpp4SESA+tb/4ZX3f8+HEeeuihcvtsS7UEPdasWcP69evp2bNndWy+3krNKeTVxUdKDVe1WODjv07x1eh2xUGPZ/s2IU2ntwp4XHcls4A1x5J56e4oVsck8kT3MAK01jcK/aJ9+fGfi0zoGcHUNSeLAx7XmS1SkrQpf5ykRaAbIzoE88MOadi5Rq1E6+TA1Vw9eqOZwW1sBDNcvKXX7cpNhqXXSs2aCqU5qteTHa6YCJN2SE/nyivhVxZX/4oHSgShkg5ekp6AVEnpVVMRbP4IdFeg09NSsK8+8GoE/i3hwFwpv4JCCs52DPNErbzIysMJPN+v8heIQsPj56bGx1VNWk7Jjf6Wc5lsi82kV6Q300dG4qlRl3rf0SvZbD2TBsDY7uH83/Jj6AqNfLH9+hpZKOQy/nyxJ9M37r/p3dKxnpBr5qmeUhb9//T2wHvN46WnGZgM+Pz9PG/3W8JjS3LwcFHx3urj7LmQwc83bVYmS6RnEx80aiXeGjWv3B3J5KVHKTSa8HN1xGg2F+cAe6JrGAqVM6jCSMp354GZ/9j8fib0jCh+CCLUQRq/8te5lfVuwaxZs+jevTudO3fmgw8+oHXr1hiNRjZs2MC3335b4VEgt+rixYt8//33DBkyhMDAQM6cOcPZs2eLb3jDw8O5ePEiMTExBAcH4+rqilpt+3e8SZMm/Pzzz6xfv56IiAgWLFjA/v37i6tfXFdUVMSECRP4z3/+w6VLl3jvvfd44YUX7ObzAOlJ/IsvvshXX32Fg4MDL7zwAl27di0Ogrz77rs88MADhISE8PDDDyOXyzl69CjHjh0rrphRnkmTJjFt2jTeeustJkyYQExMjFW1GpCqa6SlpfHZZ5/x0EMPsW7dOtauXVtuwtFb1a9fP/Ly8jhx4gQtW7a8pfcOGDAArVbLRx99VJy75boXX3yR3r17M23aNAYPHszmzZtZu3atVWBl8uTJjBw5kvbt23P33Xfzxx9/sHz58uKqJ/PmzcNkMtGlSxecnZ1ZsGABTk5OhIWF4eXlxdixYxk/fjxfffUVbdq04dKlS6SmpjJy5Eief/555syZw5gxY5g8eTLe3t7Exsby22+/MWfOnHJLp5a1b1uaNGnC8uXLGTx4MDKZjHfeeafcERw38/X1xcnJiXXr1hEcHIyjoyNarZYmTZqwYMECOnbsiE6nY/LkyRUaxfHee+9x991307hxY0aPHo3RaGTt2rW8+eabgHTMbd++ndGjR6NWq/H2Ln3/V9Y2QkNDUalUzJw5k2eeeYbjx4/z4YcfltpGXFwcCQkJNoNTFVEt4f2QkJAqO5i2b9/O4MGDCQwMRCaTsXLlSqv26+Vxbn5dT+pTl2TmGbh4Nc9mW6HBTK7eiLNKweDWAfRq4s3yw6UDHtetOZrIXdG+/Lw7jozcIpKzCzkQl8HmU8mcS80hIbOAdwc3J9rfjcRs21VK1A4K0vOK2BF7lY5hnjzQOoBpI9sw9cGWPNkjnBmj27Hs2W74u1XTkHqjAQyFMHw23Psh9H4LHlsGAz6WEpnmZ0BuWvXsWxCq2P64DPzc1Hi6lJFgtyIsJtj+BaSehHaPV24E1Z0scgAUZliVD3VSKegU7smyQ1dsDmsVGo7sAgOxqbn8eSSRrWdSuZKZX2YFEj83R34c2xHNtREfMhk81SuCeU925sF2wRy5ksXlDOttFBSZWLBbqsoil4FcJkNnY6potL8rW8/Yn0O99Uwa3ho1a17qSaRLgf3kpLkphDlJOdBaB7uz54Lt6aQWCxxNyEKfmYg+bi8DLDs4MM6NEy9F8seAHP4ckMf6cWHMGtmMUK+S/AMhnk68NbB0fqy7on25t3nV3wwLNSis+7UpT/ZGVMvALUhar4pFRERw6NAh+vXrx+uvv07Lli2555572LRpE99++22V7+86Z2dnTp8+zYgRI4iKiuLpp5/mhRdeYNKkSQCMGDGCgQMH0q9fP3x8fFi0aJHdbT3zzDMMHz6cUaNG0aVLF9LT061GfVx39913ExkZSe/evRk5ciSDBw8ut4Srs7Mzb731Fo888gjdunXDycmJ3377rbh9wIAB/Pnnn2zYsIFOnTrRtWtXpk2bZvdm2JaIiAiWLl3K8uXLad26Nd9++21x9ZbrgZ5mzZoxa9YsvvnmG9q0acO+ffuKk4JWJS8vL4YPH86vv/56y++Vy+WMGzcOk8lkNVoDpNwcs2fPZtq0abRp04Z169bx6quvWk0BGjZsGDNmzODzzz+nRYsWfPfdd8ydO7c4L4S7uztz5syhR48exSMO/vjjD7y8pAdG3377LQ899BDPPfcc0dHRTJw4kbw86X4sMDCQnTt3YjKZGDBgAC1btuTll19Gq9WWGfS6rrx932z69Ol4eHjQvXt3Bg8ezIABA245NYRSqeSrr77iu+++IzAwkKFDhwJSvpTMzEzatWvH448/zksvvYSvr2+52+vbty9Llixh9erVtG3blrvuuou9e/cWt3/wwQfExcXRuHFju1NPytqGj48P8+bNY8mSJTRv3pxPP/2UL774otQ2Fi1axL333ntLx8iNZJZquIJcs2YNM2fOZPbs2VZ1oCtj7dq17Ny5k/bt2zNixAhWrFhhVUv65trOa9euZcKECcTGxla4goxOp0Or1ZKdnV3lkc9bcTpZx8Av7ZfhmfWoVJo2UOtIvsHM9A1nWXLQduBDrZQzbWQbnl94mOXPdufFRYfRqJVMHtiUN5ceJcpPw6NdwnBQyHnml4M2tzH7sQ7FbU39XPl0RCsmLThI6g1PzFoHuTH7cSkfSJUy6iHhMOh1sOo5yLshuBHSBXq+CkvGwbg1EFw6MZVQN9wpx15NGDRjB94aFc/2vZ2RChbY/Q2cXQ9tHwO/ZuW/pS46uUoqb/3QT6CScqAcvZLFJ2tPs+zZbnQI86zlDtZ9dfHYu5qr57O1p1l8w3nP0UHON4+0p0cTbxwdbD9xM5ktJGUXcPBSJj4aNauOJPL7/pIEhGqlnJlj2tErygcnBwX5eiMTFxxgZ2w6CrmMGaPa8sKiw6W22ypIS69Ib2ZttZ/MbsVz3Rn53W52POqG/+L77a6XOHId3X/O4NvH2vPsL4dsruPlomLLhDDclo2G9Bv26dkIBn0BK56GQh3mod8gbzoI1CX5g3SFBlJ1ejafTiG30MjdzfwI9nDCy8YoF6GOKa7eAtYJTa8FQqqhektDMm7cOLKysko9dC3LvHnzeOWVV8jKyqq2ftkzdepUZs+eXWaS1epy7Ngx+vfvT2xsLK6utzaqdeLEiaSkpLB6dflllidOnMjp06crXbpUqHv0ej2RkZEsWrSoVNLaiqqWkR6PPfYYW7ZsoXHjxri6uuLp6Wn1uhX33XcfH330EcOHD7fZ7u/vb/VatWoV/fr1q5Mlc92dHOyOmlDKZbQM0tI8UIveZGHSgoP0aGJ/+sjdzXzZce4qbo5KMvKKSMgq4MW7m/DGkiNk5BXxdO/GvLHkCAUGUxlPni3FT8ee7t2IZ385ZBXwADiaoOP91SeqPhGaLkEaur/qWeuAB8DlvXB0sVTaUlN+hFIQapuu0MDpZB3R/rd5cxmzEM6shRYP1t+AB0iVXExFVqM9WgZp8XVVs2hfzV/ICbXPYrGw7niyVcADpFGQTy84SJKdEYsgVScL9nBmaNsgrmQVWAU8QCrB/swvB0nMkkZbOKuVjOwgzSs2mS2olHIcHUpfLp1NyaF9mEep5dd1Cvdgy5lUDCYLSjc/aYSiLY7uyDTS07GLaXlWVdhu9HYfH9z+eMo64AGQcUEqWd39JTAVIV/xNGRZZ7x3c3Sgia+Gp3s35rV7m9ImxF0EPOqL5kOkwMbNo/7cAkXAowGYNWsW+/fv58KFCyxYsIDPP/+8uAxuTWvVqhWfffbZLeWgyM7OZuPGjfz666+8+OKLNtf54osvOHLkCLGxscycOZP58+fX2mcUaselS5f497//XemAB1RTTo8vv/yyOjZbrpSUFNasWcP8+fPLXE+v1xeXigPpidedwM/NkU9HtGL8vP3cXB32zYHReGuk4ERqjp5jCdlkFRjo1tiL3eetM7i7OSkZ3SmUSQsO8uGwFvy48yKeLioKikxk5Rto6ufKqSQdeqOZBbsv8ca9Tfn3ymPcPObHYrbw1sCmfLTmFGoHOck62xeVG06lkJ5bhKtjFZbKPL9FukDMs5Ng59RqeGojaEQujrrkTj32qtvBS5mYLRAdcBv5PM6ul4IeUQPq/+gmRzcpyfDJFdIFu8oFuUxGv6a+rIpJ4D/3N8Pd+TanCTUwdf3YS8vR862dERUms4V1x5J4tpx8L2k5er7eHGuzzWyBP44k8kr/KAC6NPKiWYArp5JyWHzgMi/fHcl/152xeo/eaMZFpWBE+yCWHUqwalMr5bx6TxQvL4oB4IpBi6LPR3hseLXUvjP6fUqiSaomc/RKNh8Obcmo7/aUyrV1V4gMNh+x/eGSjkDft6W/WyxSXpz7PgW5SMrdIDQfAtH3S1VaclOkHB5h3cXPvwE4d+4cH330ERkZGYSGhvL666/z9ttv11p/bjUYMXToUPbt28ekSZO45557bK6zb98+PvvsM3JycmjUqBFfffUVTz31VFV0V6gjoqKiiIqKuq1tVEvQo7aib/Pnz8fV1dXuqJDrPvnkE6ZMmVJDvao4mUxG5whP/nihJ19tjuVEYjZB7k68eHckLQPdcFZJP66sfClR2Sd/neLj4a3oE+XDqpgEcvVGejT2Zlz3cObvvsjMMe1wUsnZfT6dRt4uxaM0PF1UJF97KnYoPpNQT2e+e6wDC/fFE5uaS4iHMy/e1QQvjYrtu+L45tH2XLWTIR6k66sCQ9XWgCcnpexqFGYjKFSgFDc+dcmdeuxVtz0X0vFwtj+Sq1wJB2H31xDaVSrt2hCE94bL++DsOmg5AoC+TX1YfvgKv+2/zDN9GtdyB+uWun7smSwWu4F3gPNXc8vdhtFkJqWsbaSVbMNf68hP4zrx17FkFu6Np2WQlh/HduTHfy4Sn5FPEx8Nj3QJZdmhKzzXtzF3Rfsxe9t5MvOL6BDmwRPdwlh9JJG0XD0yGVzJMbI+rSVjH1qNz8FpKDIvYPKMIq3jq3x30oHuriY+GtaS/s388NQ4sOalnny79Tz74jLw1qh5rm9jXBXljHIy3VDaL/MCmAziprchkSsgoldt96LeuZ4Y9FaMGzeOcePGVXlfbJk+fTrTp0+vkX1Vh61bt5a7zuLFi6u/I0K9V2VBj1t5alRd84d/+uknHn30UavkNra8/fbbvPbaa8X/1ul0xSVyapuzSkmLIC3TRrYhT29E7aBA62Q9gsLvWolYvdHM64uP0DzAjYEtA3BUyjmekI1SLsNF5cDkpUeYNrItAMm6QsK9pKG1cel5DG4TWLy9lTEJbD6TwrC2QXRv7I3BaCLK35XcQiO/77/MplOpfDW6XRl9VuDqWMXxM/+WUsY5exy10tNgoU65k4+96rQ7Np1mAW6VK9+ddQm2fgLeTSF6cNnHRX3ipIXAdnBiBTQbDAoV7s4qujf2Zv6uOCb0jMBBIUptVlRdP/bUSgUtA904ciXbZntZ0z2vc1IpaBmkLa6kdLNekdYJ2AK0TozvEU7fpj5sOJHCysMJNAtwo29TXxKyCpi89Ch5eiPP92vC/a0DaB2sZdOpFI5cyWbGpnM0uzadzWKRzu2z96ax+JiKR9u9Q0SEjNgsC78uzSCn0MjjfVpwT/OSkYuRfq5MHd6SnAIjKqUcd2cVRam5IJODxUYmf5kclDdMV4noY/1vQRAEQWjAquxO1d3dvdwL+uu1mytaP/tW7NixgzNnzvD777+Xu65arbZbvupO4aJW4qK2/eNRK+X0ifJh21kp18XJJB0nk6Sg02Ndw9Abzfzwj1ReNuZyJq/fE4UZ0Dop6R3lzfazV3F1VBLk7kTCtTnMugIjP1/LVj/niY54a9SolXIebCcN2z18OZMuEZ7svVg6q/zTvRvhW9Ul74Law7kN0p8JNpK69Z4MGhtVK0wGKMwCuQM4uVdtn4TbVheOvaqWU2jgeGI243tGlL/yzfQ62DgFHN2hzUioQKbweiW8N1w5ABe2QuS9ANzX0p9tZ9NYHZPIiA63Uaq6ganrx56ni4q3BzVj9Pd7SrV5a1R0Di+/bLO7s4q374vmodm7bW6/e+PS25DJZMhlcDA+kw0nU0q1j+wYguLatU+guxMKuYwVhxOQyWBCjwh+3RtPrt7IztirDGzhz9rjyXz1j/V2xnQKwdVJOt/rCgwUmcy4OSpxcpBe11lcvDG0GInD8d8opcWDELtJ+rujFpoPtR0gzc+QgiZOHmIUiCAIgtBgVFnQY8uWLVW1qUr58ccf6dChA23atKnVftSEy1n5jOkciqeLij+PJmIwWXBWKRjdKYQof9fiQEb7UHc6hnuy7ngyey9msPv8Vcb3iGBE+2A++esUn45ozext59l1LSeIt0bFv+9vTucIKTGbq6MDb90XjZuTA99tu8DnD7fB11XN2uPJGM1SktOx3cN5oHUAeqMZlbIKL6C0wdCoD3hHwt7v4MwaMJuki7leb0CbMaC44dfXYoHMOGke87n10nrdX5QqvYhkp0It2h+XgdkCzQNucWSSxQTbP4eiXOj6HCirqTT0nUzjDb7N4PhyiLwHkBHm5UL7UHdmbY1lWLsgFPIGMvJFoGWgG9893oH3Vp0onurSOdyDT0e0JsijYhXEmgW48cPYjry78nhxufYOYR78d0Rrgj2cbb4nMbOAYW2DcHdyYFVMIkUmM44OckZ2DKF1sJa0HD2hXi4o5DLubx2A0Wxh+sazfP73GWaMbsu0DWeZvyuOL0e1RevkwIrDCeiN0jZGdwqle2MvtpxOxdXRgXk740jPK6JHYy/G9YggxMMJ5bURTWoXd4r6v0+Rkzuqw/PAWCj9v9BmDIT3gBWTpNFRQ78B7U2jeHKSpeDh3u+k97UcAa1HgntopX4WgiAIglCXVEvJ2qqUm5tLbKyUeKxdu3ZMmzaNfv364enpSWiodLLW6XQEBATwv//9j2eeeeaW91HXSvcdvZLFiG93MaxtEHc388N0LevpH0cTWX8imQXjO/Pa4iN8MrwVzy88RKHBeijshB7hPNE9HIPRjJNKQZHJTJHRgpuTEj9XR+Q33UQUGkyk5egpKDKiVMpJydaTlV+EyWxh1ZFENpxM4b8jWjGsbRBqOyUDK60gCwqzwZAPZgM4eoBrgHXAA+DqOfihvzTK40atHoaB/wWX8p8CCjWvrh17lTHljxP8eSSJGaPb3tr0lphFEPMrdHxSCv41VBkXYN8cuPdDCJRq1cem5vDOqhPMGN2WoW2DarmDdVNdPfYsFgspukJ0hUYcFHI8nB0qldQ2JbuQ7EIDSoUMT2dVmds4cjmTEd/uZmjbQPo39ytO+r3maBJ/HU9izYu9aB5Y8h2aTGZScvTkFBpxUStwUMjI1Zswmy24OztQaDCjKzSQmFXA6phE/LWO6K5NJ72Rk4OC5c91p9lNAVODPh9LTgqWonzkKmcUamfkBZnSyA0nz9Lnu5wUWPYUxG23Xu4aABPWg3vYLX9/giAIglCXVEsi0+3bt5fZ3rt37wpv68CBA/Tr16/439fnJI8dO7Y4udBvv/2GxWJhzJgxt97ZOsjP1RFfV0eWHLzCkpvK93WO8MDDWcWjXUL5enNsqYAHwI874xjTJYwmfhWrJOHooCDE05mk7AKGzNxJmo2kpu+tPkGPJt52n5RVmpN7+dNU9LlSub6bAx4Ax5ZAt+dF0EOoNTvOXaVF4C3m80g5DkcWQpO7GnbAA8AjQiq9eGJVcdCjia8rHUI9+N/fZ7mvZQAqZQOb9tOAyWQy/LVO+Gtvbzt+Wkf8tBUbPeXpoibYw4llhxJKVWlpG+KOh4t13i2FQk6gu/XIE78b/l5kNDF/Vxw//HMRpVzGd493YML8A6X2W2Aw8e6q48x5oqNVUMZB7Qzqm6bLufphV8qx0gEPgJwk2Ps93P2eSAouCIIg1GvVEvTo27dvqWU3XvDfSk6Pvn37Ut5glKeffpqnn366wtus6/y0jswd14lHf9hrFYCI8Hbhi4fb4qZW0q2xN9M3nrO7jT0X0mniqym13GKxkJhVwIlEHefTcmke6EaUrysB7k5k5hXZDHgAFBrMJGYVlAQ98tIg+wpc3CEFLcJ7SiXUVC639dlt7zwLzvxlv/3kamnIryDUsBRdIbGpudzX8hZKKxflwY4vwCMUGt1VfZ2rK2QyCO0mTXHJSQRXKQnzyE4hvL38KL/suVS5fClCvZOiK+RCWi5HrmQT7OFE2xB3/N0ci6eHVFaIpzPfPtaBcXP3kaIrOQeGejrzxcNtCHCRS9Mr4/dI00hCu4FnuHTOsyEzz8CqI4kANPHVcCzBdnJWgP1xmWQXGCpfotlshEM/228/thi6vQBuNnJkCYIgCEI9US1Bj8xM68zoBoOBw4cP88477zB16tTq2GWDE+XvyqoXenDxah7x6fk08dMQ6umM37WSmD6asi+Q7E2DP5WUw5g5e8guMBQvC9A6snBil3KfVMuvt+ckw8rn4fzGkkaZHIbNlurIq0sHW25fGX2TiafAQu3YfjYNGdAy8BYeS+//AQp10OHJhpe41J6ANlLp2tN/QaenAOmGs19TX6ZvPMuQtoF4a+pukk7h9l3OyGfsT/u4cDWveJmzSsEvE7rQJsT9tnO/NAtwY/GkblxIyyMuPY9IXw2hXi6Euing4nb4bYyUSPu6wA4w+hdplNLNZCVnLIuFcs+tt9dzWdnnQJn8dncgCIIgCHe8armi1mq1Vi9vb2/uuecePvvsM958883q2GWDFOjuRI8m3ozpEkqncM/igAdImei7XEtIakuXRqWneyRnFzBh/n6rgAdAUnYhL/8Wg6taib+b7eHAzioF/lpHKdnokd+sAx4gZYtfOUl6UlvVnDykTPX2lNUmCNVo29k0Gvu44HZT2Wm7Eg/Dub+h6SDp91qQKBwgqIP03RgLixeP7BQCFpi65lQtdk6obTmFBt5bfcIq4AGQX2Ri3Nx9xUlPb1eYlwv9on15skcEPSN9CPV0lqaI3BzwAEg8CDumWf2+XufpouKha5WHYtNyaRVkPyjavbEX2sqO8gApz0eHcfbb2zwKzuWX+xUEQRCEuqxaRnrY4+Pjw5kzZ2pyl/VeocFEao6enEIDLioljg5ycgqNWID3Brdk5He7ydUbrd7z8t2R+NgoMZuWoycp2/bF4dEr2ZgsZqaNbMP6mIuMbq7G0ZyPQeHEynMGmocHSmVrc1Nhzze2O2uxSDk22j4GDk5lV1XR50JeqvSn2k1aV2UjX0hBJuSlS3k7Wg6Xkj+e/rOkvcM4qRKMINQwo8nM9rNp9G9exlz7G5mKYPc34NkIgjtVb+fqopAu0hP1uB3Q5B4A3BwdeKRLKN9tv8DgNgHcFV3B71qoV9Jzi9hyJtVmm67QyMWreQTdkGPDZJaSoWblG3BQyPB0UeFV2ZFCl3aVDnhcd3gB9HgZ3EsqqeTpjVzN1fNQh2AupObwRCs1UcrLHJoUTKbJESUmzIU6si3OzD+m59m7o9FWNGhqj29ziBoojZa6kXuYlChZcZvbFwRBEIQ7XLUEPY4ePWr1b4vFQlJSEp9++mmDKClbU9Jy9Hy37TwL9lxCbzQjk8Fd0b480jmUl3+LoW2IlqXPdOPPo4n8E5uOj6uaib0iiPJzxc2x9EVOTqHRxl5ubDfRxddIV8ffkC+bL13oyeS80XwYltCPUCgV0oiOvKv2N5IVD+vegoyLMHwO+LUsPYRflwgb3oXjy6TtKRyg7ePQ9y1wvSE3QmYcrHpBugkCKV9ItxelKTTHl0nzlP1bgbNnBb9RQag6R65koSs00ibYvWJvOLZMCvR1f0nKYyFYc/YE7yg4vaY46AHQJ8qHvRczeHPpMda/4l75m1ehzioymSkr9VfGDbmocgoMbD6TypQ/TpKRVwRAtL8rX45qS1N/11tLOAygS7LfZiyUqo5dk5xdyGfrT7MqJpFnu/kyo10C6vWTpYcFgKdPU+g/BXZ+CumxtOr9JnJNBFCxpON2aXxh8FeQcBD2zpaqobUaCdGDxEMBQRAEoUGolqBH27ZSacabE5B27dqVn376qTp22eAUGkx8t/08P/xzsXiZxQKbTqWSmVfEC3c14dO1p7l/5g7+fqU3T/VqhEopx1ll/0ce4O6ETIbNi0dHBzlBLmYU2z6Fg3NLGixm5CeWS8lER/wojcYI6Qrxu2zvJLgT7JopBSzmDYJn/gGP8JL2/Ez44xU4t75kmckAB3+SnoTf918pJ4guCRY8KJWzvK4oD7Z9CgM/hZE/V0/SVEGooI2nUnFzVNLEpwI5bHJTpYSCYT3KHgHV0IV0kZ6ep8eCVxNAyofwdO9G/GvZUV5bfIS54zqVKrst1G8atRJPF1VxEONmUTdUKjuWmM3Lv8VYtZ9OzmHk97tZ82IvQjxvsQJZSGf7bV6NwUE6D2UXGHh31XH+PpmCt0bF443yUS8Za71+2hmptOzDc+HXh1FsfBccXaH9uNvP7+PqJwU5InpJ01AdtSK4KgiCIDQY1ZLT4+LFi1y4cIGLFy9y8eJFLl26RH5+Prt27SI6Oro6dtngpOXoWbD7ks22Q/FZNPHRcF9Lf94b3IKzKbko5DKrgMfVXD3J2QXkFJY8hfLWqHiove2nPs/3bYLWnCXdcNhyfrNUscXJA+790HbiNI9wqT0zTvq3PgdO/Wm9Tn6adcDjRkcXSfsAKdiRlyYlNXxgOtzzgZTsEGD751CQZXsbglBDNp5MoV2oR8VuwA/NB6UaGvcrf92GzCdaulm7aZi+h7OK5/o2YfvZNKZvPFtLnRNqi5+bI28OaGqzrW9TH3zdpNE/mXlF/HftaZvr6QqMbD+XVu6+MvOKSM4uICv/WoDFUSuNKLSl92RppIfJhCIvhYeilDzY2punOnrhve8z2+8pyoX4vVKAD2DrJ5CbXG6/7MnVG0jOLuBqzrXRLmpXqaKaCHgIgiAIDUi1jPQICwurjs0KN8jVG9EbzTbbgj2cCHR3xMNZxczNschlMKJ9MI91DUPtIGfnuat8vSWW1Bw97UPcef3epjTy1eDq6MCbA5sS7OHEj/9cRFdoxEej5qW7mzCodQDy7FNS+Tt7cpLBp6k0f3jsn7B2MqScALlSSszYYSysfM76PZd2QpdnQHHtV7GsqTFmk5TDgwhpvYfmwoGfpKksLt5SrpDOE6WRIob8W/o+BaEqxafncy41lwda26jccLP0c3BhK7R4EJS2EwUL18jlENwRLmyBjuPBoeSpfJsQd0Z1CmHm5ljCvVwY0UEM228oFHIZA1v64+ig4L/rTpOUXYizSsFjXcOY0DMCTxcp6FFoMHE6OcfudnafT+fRLravX7LyizhyOYtpG85yKSOfSF8N/x3RmoizfyO76z9wYoX0MuqlvDy9XoOkoxDcGXZ8gebQfO41FdEzcjC0eQZFfK79D5RyXNrG5b3SKDBDwS1/J3qDiYvpeUzfcI69F9PxclHzTJ9G9Gvqg7er+H9GEARBaFiqNOixefNmXnjhBfbs2YObm5tVW3Z2Nt27d2f27Nn06tWrKnfbIDmrFMhlYL5pKopMBh8Obcm4uftJzSmZxzxr63n+PJrE/0a24aUbhvZuPpPG1rNpLJzYla6NvPBxdeT5fk14uGMIRUYzagc5fq6O0tPqwnKG6V/PnaFyhvAe8MQqacRF5kVpHv7vj5cORvhElwQ8QHpqVhb1tWHKGl/4eXBJArmCTNj4npSs7a53xM2jUKvWn0hGpZDTOrgCpWoPzpd+n4M6Vn/H6oOgjnB+i5TUNGqgVdOQNoGk6Ap5c+lR3JwcuKeiSWSFOs/dWcWwdkF0bexFYZEJlVKOt0aFSqkoXsdBISfI3alUlZfrov1t584oMJhYevAKH91QJWh/XCZvLTvK792Ckf3+OLQYBg9+J41yzEmWkhL3fVuq7JJWksDdOeYnOLsKHp4PC4ZKwfxSHyYE0q9N3VS7gvLWq7ecTNLx8OzdGK9dJGTlG5i89ChD2wTy/pAWeLjcRkUYQRAEQahjqnR6y5dffsnEiRNLBTxAKmM7adIkpk2bVpW7bLC8NGoGtvAvtbxnE292nU+3CnhcF5+Rz94LGaVuxMwW+L/lx0i79h6lQk6guxPh3i4EaJ1Khuc7e0Pj/rY75NsMNDfdYLj4gHek9OTr4LzSAQ+5AtqOKf0ee0OFG/WVRnTkp8Pf/7adMf/sOghsJ/IiCLVq7fEkWgdrcXRQlL1i8jGpTG2T/rc/Z7+hcHIH76Zw5q9STTKZjAk9G9Eh3INnfznI+hOVnxYg1E3+bo6Ee7sQ6O5kFfAA8HZV8/LdkTbf56CQcb+dkVlXc/R8vr505bn9cZno/dpLQfaji2HJOFj8BKx9U0qOlZ9hFfAolp+O+fQaiLqvdJtcAZH3wsWt0r87TwJN6XN9WTLy9Pxn5fHigMd1aqWctFw9OXo71WYEQRAEoZ6q0qvsI0eOMHDgQLvt9957LwcPHqzKXTZYGrWSdwY3p0sj68okd0X72i3dB7D1TCqdwktXM7lwNY/sgnIuhJzcYciX0nDdG3lHwehF9gMNYd2h95vSxdx1alcY/RtoQ63X1fjCyAXSCJAbBbaHIV9LOUEKdVIWensSD0lPz66ehe1fwB+vwtn1UlUYQahmKbpCDsVn2TzOSon5BdwCwa9F9XesPgnpBOnnpalBN1HIZbx4VxM6hHnw3C+HWLz/ci10ULhT9Yry5pnejbgx1Y6rWslP4zoR5G57hGCyrtDudNK3NmdjeWyZFLC/UeeJcOoPu/2Qn/4TQ/vx1udFlQYGz4ADc6VzWMsR0PnpWy4pqyswciJRZ7XsjV6+bHzUizlBawjc+R7E7SyuGiMIgiAI9V2VTm9JSUnBwcH+yVmpVJKWVn6iMKFiArROfPtoe9JyikjIKsDLRYW7swPLDl2x+x5HBwVFdi7elBVJuKgNgTGLpOG7ugRpdIdrgJQZ3h5nL+jxMrR9BDLOg9IJ3EOlp1dKG78vnhHwxGopeVtOMrgFSfvRXLuolMmlC0Vbw4JBqtpyYRv8PqakFM3Bn6RqD4+vlIYOC0I1+etYEkq5jPZhHmWvmHwMko9Du8dtJ/4V7PNuWpLQtFvpJ/dKuZyX7opk7q6LvLnsKOev5vLmgGgUoqpLg+fpouaFu5owpksoF6/m4aRSEOzhjJ+rGqXC9nHoYGc5wOqjqUy+tw8hEzZI58SCTPBsLAVBzm+x3xEHRwyekcieO4Ay64I0WsQ1AIrypfPdPVOkbTiV8/+IDXK5zKoS29t9/XjEvBrX32eUrHRwDoT1hId+kPYrCIIgCPVYlQY9goKCOHbsGE2aNLHZfvToUQICxMm1Knm6qPF0UdP0hrnIY7uFM3npUZvrP9A6gJmbY0st7xLhibtzBZ8muXhLL/+WFe+oWiO9PCMqtr6rn/S6XpHlRs5e0PR+OLXa9nvDe8HsHqVr76bHwtZPYdAXoHKqeN8F4Rb8cSSR1sFaNOpy/ns9skga5eHbrGY6Vp8UJzTdCh0nWCU0LVlFxvgeEfi7OTFn+wVOJOiYMbotXhp1zfdXuKNoHB3QODoQ5lWxsua+rmrcnR3Iyi89GjLYwwmVSgluEaXPb52fhtN/lnoPAJ0m4uwVIiXi8m50U6OdKZ4V5OHswN3Rvmw8lYpGreSBED2uS2aUXvHSP3BipZRMXFRzEQRBEOqxKn28OGjQIN59910KCwtLtRUUFPDee+/xwAMPVOUuBRt6R/nQJaL00Pp7mvsSHeBGQpZ1JngPZwemPtgKd+c6kthMrZFK1LramOc84BPIumx/FMixxZBfRoUYQbgNCVkFHIrPomsjr7JXTDsDSUekPDXiZqNygjqBsUgKfNghk8m4v3UAb9/XjGMJ2dw3Ywf7LmbUXB+FesHXVc3XY9rhoLA+Vh0d5Hw1uh1+bnYSZ/s2h1YPl14e1BGaDa62Y9/V0YH/3N8cH42a3pFe+J7+xf7Ke7+DPDHNRRAEQajfZBbLzY/DKy8lJYX27dujUCh44YUXaNq0KTKZjFOnTvHNN99gMpk4dOgQfn53VkZ9nU6HVqslOzvbZhLWuihVV8jJJB2/7b+MQgaPdgkj0s8VZ5WCK5n5/LbvMpcy8ukT5cPdzXwJ9ij9pPSOl3VZGj58Zo00PLfjeHAPg33fw5aP7L/vpZiKjzgRqlV9O/a+2RLLV5vO8e2jHXBSlZHEdMtUKR9F91dEAtPbcehnKVHykK+Asm8gM/KK+GZLLKeTdbzaP4rn+jVp0NNd6tuxV92KjCauZBaw/NAVTibl0D7UncFtAglyd7I7LQaA3DQpv9SBn8BYAO2ekEYwulX/qNeEzHwuX82h64FXbCb+BaSqa8/sqpH+CIIgCEJtqdKgB8ClS5d49tlnWb9+Pdc3LZPJGDBgALNmzSI8PLwqd1cl6vPFn9FkBpk0x/1GFosFg8mCSlkPbriMRSBXltw8Xt4PP9qpMhPQBh5bLk3PEWpdfTr2LBYL/adtw1/ryAv9bFeIACA7AVZMkkpchnS2v55QvrTTUsnf+6eBT9NyVzeZLSw/dIUVhxPoGenNjNHt8GygpTvr07FXkyp97jSbpCmXiiqdVVwhlmNLkS2bYLux9Sh44Eup1LwgCIIg1FNVfvYNCwvjr7/+IjMzk9jYWCwWC5GRkXh43HoyLsE+k8lMWm4RZosFJ5UCDztTU+w9gZLJZKiU9eQpp/Kmz+4RJuX1iNthvVwmh/s+EwEPoVocS8jmfFoeD3UoJ1HuyRXSFK3AdjXTsfrMK0pK9HjmrwoFPRRyGQ93DKGpvyvfbIll8Mx/+HFcR6L9xU2/UDGVPnfKyylffYtScwoxmiwoFTJ8Xe1Mr7lGFtoNPBtBxgXrBgdn6P2GCHgIgiAI9V61Peb38PCgU6dOdO7cWQQ8qliKrpCZW2IZOGM73T/dzPh5+zl4KZM8vbG2u3Zn0PjCiDnQ923phkgmg5CuMGED+Leu7d4J9dTiA5fxclHROkhrf6XCbIjdKP0+3mIZSsEGuVwaLXNxG+h15a9/Tetgdz4a1goHhYwRs3bxzzmR50eoGzLy9Kw8nMBD3+6m+6ebeejb3aw8nEBGnt7+m7RBUkW0zk9LgQ6ZHJoOgolbwOPmJKqCIAiCUP9U+fSWuqguDfO9mqvnpUWH2HXeOhmfTAYLJ3alW3kJFBsSkwnyUsBili70nEsndxVqV1069spSaDDR6aON3N3Ml1GdQu2veGQRHP0d+rwllVYWbp8+F7b9F9o/AS1H3NJbCw0mZmw6y/EEHd882p4BLWwkR66n6sux15AUGkzM2XGB//19tlTb6/dGMbFXIxwdyhhRYtRDfrp0TnTUgtrV/rqCIAiCUI/Ug4QOdZ/ZbEFvtFNt5CYJmQWlAh4AKoWc9ceSyMgt42lPQ6NQSCVBtcEi4CFUqz+PJpGjN9K3qa/9lUxFUvnKwHYi4FGV1BoIaA2n14ClYv+PXufooOD1e5vSMdyD5349xKZTKdXUSaEm6Y0mzOaaf55juoVzeWWk5ej52kbJeYCvN8eSllPO+V+pLjknioCHIAiC0ICIoEctyi8ycjYlh4/WnOSZBYeYs/0ClzPyy7xY230hvdSyhzsG882j7Skwmnl9yVHm7bzIlcz86uy6IAg3+GXPJVoHa+2XrgS4uB0KsiCsR431q8EI7Qa5KVIS41uklMt5oV8k7UPdefaXQxyIEyVt6yKLxcLljHx+2HGBZxYc4sM/T3I2OYf8Gpj2qSswcDwhm/+sPMYzCw6xcG88CZkF5b/xFqXn6dEbzTbb9EYz6eKhhyAIgiDYVPNpxAUA9AYTm0+l8uJvh7k+wWjLmVS+2nSOJc92s5tYz8PZOg/AfS39ifR15an5B4qXbTmTypebzrH0me408dVU22cQBAGOJ2QTczmL1/pHlbGWBU6uAp8oKeeMULW0weAeKn3HoV1v+e0KuYwX74rkk7WneOrnA6x+viehXiK5Y11yNiWXh2fvQldYEuSYtzuOL0e1ZUAL/7KnfdyG3EIDvx+4zNQ1p4qXbTmTio+rmqXPdCPMq+pGdamUZX8GVTV9RkEQBEGo68RIj1qSmqPn9SVHuDmjSo7eyBtLjtidptK1kRfyGxLHP9wxhP+uO11qvax8A/9ecYys/KKq7LYgCDeZtysOb42K9mFlJGxOPi5VTggVozyqTXgPSD4KGecr9XYHhZzX+jfFUalg4s8HyC8SiaHrioy8IiYvPWIV8ACpQuzkJUfLn/ZxG1Jy9FYBj+vScvR8svY0uXpDle3Ly0VFsIeTzbZgDye8Gmj5ZUEQBEEozx0f9Ni+fTuDBw8mMDAQmUzGypUrS61z6tQphgwZglarxdXVla5duxIfH1/znb0F51Jz7A5TPZ6gIzPf9oWSr6uaL0e1RSaTLnLiruZhsjMdZu/FDLKubafQYOJyRj47zqWx/Wwa8Rn5FIiLekG4LVdz9ayOSaR/Mz8U8jLKWJ5aLY3w8I6suc41NL4tpWpNJ1ZUehMaRyWv3RNFXHoe7646XoWdE6pTVn4RR69k22wrMpk5k5JT4W1l5BVxNiWHTadSOByfSXJ2YZnrl1X55+8TyWTmVV3Qw8/NkW8f64BGbT1IV6NWMvuxDmVPrxMEQRCEBuyOn96Sl5dHmzZtePLJJxkxonRm/vPnz9OzZ08mTJjAlClT0Gq1nDp1CkfHO/vkrzfYDnhcZ7JTVMdJpaR/cz82vdaHvRcyKCwnaZrZYiGn0MBfx5J4d9WJ4kCLg0LGOw80Z1jbINycROlMQaiMBbsvIZPB3dF+9lfKTYH4PdBsiFRmSagecrmUL+XsWmg/Flx8KrWZEE9nxveI4Ntt5+kV6cPQtkFV3FGhqtkL/F+nN1QsuWiKrpC3lh5l69m04mUBWkfmPdmZpv62E38WlLFts0U6B1elFgFurHu5F7supHPsShatgt3p1siLIHfbI0AEQRAEQagDQY/77ruP++67z277v//9bwYNGsRnn31WvKxRozu/7nx0gBsyGaWmt4A0gkNbRiDCWaWkkY+GRj4aYlPtP8Fq7KNB6+jAxat5vLXsmFWbwWTh3VUnaBHgRodwUdlEEG5VQZGJBXsu0TvKB41jGf+Vnv5TqpoQ2K7mOtdQBXeEC5vhxEroPLHSm+kd5cORK1n8Z8VxOoV7EihuKO9obk4OBHs4ccVG8lCZDJoHlF+St6DIyIyNZ60CHgBJ2YU89sNeVr3Qw+bvQa9Ibz5da3ub7ULdq/yhglwuI9jTmZGezozsGFKl2xYEQRCE+uqOn95SFrPZzJo1a4iKimLAgAH4+vrSpUsXm1NgbqTX69HpdFavmuatUfF0r9LBGZkMpg5raXeYqslsISm7gItX80jMKsBHo+axLqGl1lPIZUx9sCXOjgrmbL9otx+ztp4nrway2wsC3BnHXlVZcvAyWflF3N8qwP5KhgI4ux6CO4FSzLevdkq1VMnl7FootD3doaLG94hApZTzr2VHsVTx0/raUJ+OvZv5uTnyyfBW2JphNqFHBN4adbnbuJpbxNKDCTbb0nL1XErPK7U8PVePo1LBwqe6MLZ7uNW0EweFjClDWuDhLI57QRAEQahtdTrokZqaSm5uLp9++ikDBw7k77//5sEHH2T48OFs27bN7vs++eQTtFpt8SskpOaflrg6OjCpT2NmP9aeFoFueLqo6Bvlw6rne9ApwvbIi/RcPfN3x3H/V//Q74utDPxyO/N3X+K5fo35anQ7ov1d8XJRcU9zX1a/0IO2Ie7oDWYuZZS+WLsuPiOfwgoO/RWE23UnHHtVwWAy8922C3Rt5FX2PPrzm6TAR2i3mutcQxfaXfrzNnJ7ALiolTzVK4Lt566y7JDtm+G6pL4ce/Z0CPNg1fM96dfUBy8XFS0C3fj20fY8268xrhUYbVFoMFFksj/t9EpWgdW6By9l8viP+7h72jYe+3EvCZkFzHmiAy0C3bi/VQB/vtiLaDtTYgRBEARBqFkySx16hCWTyVixYgXDhg0DIDExkaCgIMaMGcPChQuL1xsyZAguLi4sWrTI5nb0ej16fUk2d51OR0hICNnZ2bi5lT8Mtqpl5OkpMppxUStxdbR9cVZoMPHNllhmbo4t1TamUwj/d38zioxmDCYLLmpF8XaKjCY+WnOKn3dfsrndEe2Dmfpgy2or5ycIN7rTjr3KWnzgMm8uPcp/R7Qm1NNOaVOLGVZMAmcvaPtIzXawoTuzDi7vhYd+AkftbW3qmy2xHE3IYvPrfSs0YuBOVV+OvfLkFBrI0xtRKeV4ulT855WQlc/A6TvIsTPyceVz3WkbKlVoOpGYzZCvd5bKJeLnpmbxpG54a9S4qO/42cOCIAiC0GDU6ZEe3t7eKJVKmjdvbrW8WbNmZVZvUavVuLm5Wb1qk6eLGn+tk92AB0jl777ffsFm2+8HLpOeW4SXRo2/1tFqOyqlgrHdwlEpSv+olXIZk/o0EgEPocbcacdeZRhMZr7eHEvncE/7AQ+AK/tBlwjhPWuuc4Ikorf057Elt72px7uGYTFjsyxpXVIfjr2KcHV0wF/rdEsBDwBfV0ee69fYZluzAFeCrpWKzSk08MX6MzaTp6bo9ByIyxQBD0EQBEG4w9TpoIdKpaJTp06cOXPGavnZs2cJCwurpV5Vj6z8Irslbs0WSM2xX1YvxNOJX5/qYnWDFuzhxM/jOxNW1k2bIAilLDt4hfiMfIa3L6eqx4nl4B4qvYSapXKG8B5SEtm8tPLXL4ObkwNjOoey4nACey6kV1EHhTuNg0LOyI4hvNI/EqdrDwJkMugT5cOcJzri4ypNY8vVG9kfl2l3OxtOpWAup5qMIAiCIAg1645/HJGbm0tsbMmUjosXLxITE4OnpyehoaFMnjyZUaNG0bt3b/r168e6dev4448/2Lp1a+11uhqUNxqjrCdLKqWCThGeLH2mG1n5BixYcHdWlZ2LQBCEUgoNJmZsOkfXRp6EebnYX/HqGUg+Dm0frbnOCdbCe8LlfXD4Z+j5+m1tqk9TH7aeTeU/K4+z9uVeONgYOSfUfV4aNc/1bcxD7YPJ0RtxdFDg5aKyqsCikMvwdFGRa2caTKC7I3JbGVUFQRAEQag1d/yV24EDB2jXrh3t2knlHl977TXatWvHu+++C8CDDz7I7Nmz+eyzz2jVqhU//PADy5Yto2fP+jWk3PNaYjZbgj2cKjTX3NfNkSh/V5r6u4mAhyBUws+740jRFZZfKvL4cnD2Bt/mZa8nVB+lIzS+G2I3S0Go2yCXyRjfI4ILabnM2xlXNf0T7kgqpYJgT2eaBbgR4e1SquSsr6sjT/cuXXntulGijKwgCIIg3HHu+KBH3759sVgspV7z5s0rXmf8+PGcO3eOgoICYmJiGDp0aO11uJp4adR8/Ug7/G8KVng4O/DD2I4iiCEI1Swzr4iZm2O5K9qPAK2T/RV1CRC3UxppIL/j/4ut34I7gWsA7J0tJZa9DWFeLvRv5sf0jWdJzrY/nVCo/wa08OO+lv5Wy2Qy+OTBVgS6l/F/gyAIgiAIteKOn94ilIjw1rDiue6cS83ldLKOxj4aogPcCBIXWYJQ7b7ceBaT2cJDHYLLXvHYElC7QlD7mumYYJ9cDs0Gw77v4dzfEDXwtjY3smMIey9m8PFfJ/lqjPj5NlQ+ro5MfbAlL9zVhD0XMnBRKejayAtfVzXOIompIAiCINxxxNm5jglwdyLA3YneUT613RVBaDBOJelYsOcSYzqHonWyX2WJ3BQ4vxkiB4CijPWEmuMZAUEd4cBP0sgPZ69Kb8pFrWRM5xBmb7vAmM7pdGtc+W0JdZunixpPFzUtAm+vJLIgCIIgCNVPjL0WBEEog9ls4d1VxwnQOjGwhX/ZKx9bLOWSCOlSM50TKqbpfSCTw+6vgdurrNEr0ocoPw3vrDqOwXR7U2YEQRAEQRCE6ieCHoIgCGVYcvAy++MyGdc9HGVZVTtykuDcBojoDUpVzXVQKJ/KGZo/KFVzOff3bW3qxqSmP/5zsYo6KAiCIAiCIFQXEfQQBEGwI0VXyNQ1p+gd5U3LoHKGscf8CioXCO1aM50Tbo1fM2l6y97vIDv+tjYV5uXCgBb+fLnxLJcz8quog4IgCIIgCEJ1EEGPGpSrN5CiKyS7wFDbXREEoRwWi4W3lx9DLpPxeJfwslfOiIXzW6DRXaAQozzuWNEPgKM7bP4YDAW3tamHO4SgUSv5z8rjWCy3N2VGqH/y9UZSdIVk5RfVdlcEQRAEocETQY8akF9k5FhCNq/+HsOD3+xk4s/7+edcmrgYEoQ72MJ98Ww+ncpTvRqhcSwr57MF9s4BjZ80kkC4cylV0PYRyEuFHf+7rTK2TioFT/aIYNvZNJYfSqjCTgp1WZHRxJnkHP61/BgPfrOTsT/t4+8TyaTn6Wu7a4IgCILQYImgRzWzWCzsuZDBkK//YcPJVBKzC9l3MZPHftzH7/svk19krO0uCoJwkzPJOXzwx0nujvalQ5hH2Stf3A4px6HpIKlEqnBn0/hCq1EQvwcOzb+tTbUP9aBnE2+m/HGCpOzbGzki1A/HEnQ8MHMHq48kkphdyJEr2Ty94CDfbI5FVyAedAiCIAhCbRBX6NUsWVfIv5Ydxdbo58/Xn+Fqjnj6Iwh3El2hgUkLDuDn5sgT3cLLXrkoF/bNAf+W4BNVI/0TqoBfM4i+H44thZOrbmtTY7tJCW7fWHIEs1lMc2nIrubq+b/lxzCYSv8e/LQzjrQcEfQQBEEQhNoggh7VLCvfQKqdwIbRbCEuXSTBE4Q7hdFk5qVFh0nL0fNq/yhUynL+i9w3R8oN0fSBmumgUHXCe0iVdvZ9D2fWVnozGkclz/RpzM7YdL7bfqEKOyjUNdkFBs6k5NhtPxifWYO9EQRBEAThOhH0qGZymazMdoeySmAKglBjLBYLU/44yfazabx4VyT+Wsey33BpJ8RulEYMOJVT2UW4M0UNhNBusPtrOPVHpTfTKkjLkDaBfLH+DLvPp1dhB4W6RFHO+V4lzveCIAiCUCvEGbiaeTg7EO7lbLPNyUFBiKdTDfdIEISbWSwW/vf3WRbsucT4nhG0CXEv+w05ibBzhjStJahDjfRRqAYyGTQbDOG9YO9sOPQzULkpKiM7htAswJVnfz1IvBjB1yC5OzvQMdx2DiC5DNqFutdshwRBEARBAETQo9r5ujny5ai2qG8aJi+Twf9GtsHXVV1LPRMEAaSAxxd/n+HrLbE80jmUu6P9yn5DUS5s/AAcnKDFCOlgFuoumQya3ie9jv4OWz+pVDlbhVzGS3dH4uig4Im5e0nPFfmaGhp3ZxUfP9gKNxvVnt4f0gIfjTjfC4IgCEJtkFkstlJsNiw6nQ6tVkt2djZubm5Vvn2jycyVzAKWHrzMwfgsGnu78Hi3MEI8nXFWlVUKUxDqt+o+9spTZDTzn5XHWHzgCo92CeWB1oFlv8FYCBveg4zz0OUZqRKIUH8kn4BjS6Sfa9+3wCPiljeRoivk/dUn8Nc68utTXfC6Q290a/vYq68sFgtXMgv440giO2KvEqh1ZFz3CMK9nHF1cqjt7gmCIAhCgySCHtTcxZ/JbKHQYEKllItcHoJA7d54JWUX8MLCwxy5nMXEXo3oHeVT9huKcmHzR5B2Fjo+CR5hNdNRoWblpsKR3yD/KrR9BFoMB/mtBacvZ+Qz9a9TeLqo+Hl8Z0I8bU9xrE0i6FG9zNfO90qFvPyEyIIgCIIgVCtxJq5BCrkMF7VSBDwEoRZZLBaWHLjMgOnbibuaxzsPNC8/4JGdAH9NhvRYEfCo7zS+0PVZKcHpoQWw6gVIOMit5PoI8XTm/cEtyC8yMvjrf9hyJrX6+ivckeRyGc5qpQh4CIIgCMIdQIz0QDzxEoTaUpPHnsViYfeFdD5bd4aYy1n0bOLN2G7haGzMvy9mNsHpP+HQfHDUQttHxZSWhkSXBKf/gIyL4NscWj8MQR1BVrEb2ZxCA7O2nifmchajOobw5sCmd8x0F3HeEwRBEAShoRBBD8TFnyDUlpo49rLzDfx1PImFe+M5lpBNI28XxnQOpWVQGWVmTXq4sE3K76BLgtAuUnlT5Z1xwyrUIIsFrp6B2M2QfRk0ftDkHojoAdoQoOxEthaLhY2nUvl9fzwW4PGuYTzWNazWp7yI854gCIIgCA2FCHogLv4EobZU9bFnsVhIy9FzKjmHmPgsdp2/ysFLmZjMFtqEaBnQwp82we7Ibq64YjGBLhHSzkDiYbi8V6rg4dtMusF1C7jtvgl1nMUCWfFwZT+kHAejXhr1498GfAZeacAAAQAASURBVJqCZyNwDwEH28GMnEIDfxxJZNPpVPKLTLQPdeeuaF86hXvSPNANV8eaTXIpznuCIAiCIDQUIugBZGdn4+7uzuXLl8XFnyBUAVdX19KBBRsqeuxdySpk2qYLpOYWUWQ0U2Q0U2Awk1dkQldoLHMfTXycaRfshrsalJd2INfFg9GAzFiIvTwNFpUGszYUi8ql3M8gNEBmE/L8NOS5KWAqKmNFGRalCuQqLAolyBToZY7sc+jIsXT7x4eLSoFGrcBZpUCtlKNRK3m5bzhtgss/P1X1sScIQsVU9NgTBEEQap4IegBXrlwhJCSktrshCPVGRZ8eV/TY03YfjXuvx6qia4JQJ+Wf3U3aiqnlrlfVx54gCBUjRk0JgiDcuUTQAzCbzSQmJt5ylF6n0xESElLvn5SJz1m/1MTnrOixZO/Yq2s/i7rWX6h7fRb9rZjbPfbqs7r2O1QTxHdi7Xa+j4Z0LAmCINQ1ZZQtaDjkcjnBwcGVfr+bm1uDuFgQn7N+uRM+Z3nH3p3Qx1tR1/oLda/Por9V43bPe3XZnfozqU3iO7Emvg9BEIT6RRSQFwRBEARBEARBEAShXhJBD0EQBEEQBEEQBEEQ6iUR9LgNarWa9957D7VaXdtdqVbic9YvdeFz1oU+3qiu9RfqXp9Ff4XbJX4mpYnvxJr4PgRBEOonkchUEARBEARBEARBEIR6SYz0EARBEARBEARBEAShXhJBD0EQBEEQBEEQBEEQ6iUR9BAEQRAEQRAEQRAEoV4SQQ9BEARBEARBEARBEOolEfQALBYLOp0OkdNVEGqWOPYEoXaIY08QBEEQhIZCBD2AnJwctFotOTk5td0VQWhQxLEnCLVDHHuCIAiCIDQUIughCIIgCIIgCIIgCEK9JIIegiAIgiAIgiAIgiDUSyLoIQiCIAiCIAiCIAhCvSSCHoIgCIIgCIIgCIIg1EvK2u6AINSowhzISwV9Dji6gYsvqDW13StBaJhMRshNhvwMkCvB2Qtc/Wq7V4IgCIIgCEI9IoIeQsOhS4R1/4JTq8FiAbkCWj4M/d8Ht4Da7p0gNCyFOji7HtZOhoJMaZlXYxjxI/i3lo5PQRAEQRAEQbhNYnqL0DDkZ8Lql+DkKingAWA2wdHfYMM70g2YIAg1J+U4LH+qJOABkH4e5t0PWZdrr1+CIAiCIAhCvSKCHkLDkJ8GsRtstx1fBnlpNdsfQWjICjJh4xTbbUV50mgsQRAEQRAEQagCIughNAz5GfbbLGYozK65vghCQ2cogLRT9tvjd0v5PgShnioymknL0dd2NwRBEAShQRA5PYSGwVFbdrvatWb6IQgCKFTgHgbJR223+7UAhTg9CfXTmqNJvL3iKLoCI50jPJn9WAc8XVS13S1BEARBqLfESA+hYXDxgcD2ttsa9QUX7xrtjiA0aC7e0O9t221yJbQeXbP9EYQasjP2Ki8uOkTzADee69uYM8k5jPtpH0aTuba7JgiCIAj1lgh6CA2Dizc8PA8C2lgvD+kCQ78BJ49a6ZYgNFghXeGud6Qgx3WO7vDIEtAG11q3BKG6FBSZmLz0CM0C3Hjxrkh6Rfrwxr1NOZ6YzQ//XKzt7gmCIAhCvSXGDwsNh0cYPLoM8lIhNw1c/aQRIGKUhyDUPGdP6PostHoIsuJBqQa3QNAEiKktQr20cF88ydmFTL43GrlMBkATXw39m/kxa2ssj3UNQ6MWv/uCIAiCUNXE2VVoWDQ+0suvtjsiCAIqF+nlEV7bPRGEalVkNPPdtvP0ivTBX+to1TakTSBbzqSycO8lnu7duJZ6KAiCIAj1V61Ob/nkk0/o1KkTrq6u+Pr6MmzYMM6cOWO1zvLlyxkwYADe3t7IZDJiYmJKbUev1/Piiy/i7e2Ni4sLQ4YM4cqVKzX0KYTKMpktJGUVcORyFocuZXI5Ix+9wVS9O827CqmnIX4PpJ21X9XFUAiZl+DyPkg4BNlXwHytb0X519r2QuJhyE4As5iPLdQNWflFxKbmsj8ug7MpOaTnVlEFiYJsSI+Vjq2UE5CbYn/dQh2kn5fWTT4OOWWsezvMZtAlQGKMdLxmxknHryDUsA0nU0jN0XN/q4BSbV4aNV0ivPh1bzwWi6UWeicIgiAI9VutjvTYtm0bzz//PJ06dcJoNPLvf/+be++9l5MnT+Li4gJAXl4ePXr04OGHH2bixIk2t/PKK6/wxx9/8Ntvv+Hl5cXrr7/OAw88wMGDB1EoFDX5kYQK0htM7L2YwUu/HSYr3wCAWinn3/c3Y2jbILRODlW/06x4WDoBruwrWRbRB4Z9C9qgkmUFWXB0MWx4B4yF0jJnTxjxE/i3hkM/w9aPwVQktbn4wMifIagTKKuh34JQRZKyC3h72TG2nk0rXtYmRMvXY9oT4ulc+Q3nJMO6t+HE8pJlvs1g1C/g1eSmdVNg4/twdBFcv8HzjoRRC8EnqvJ9uJnRAIkHYfHjkJsqLVOooM+/oOOT0jEtCDVk0b54mvq52j3O+kX78uGfJ9l7MYOujbxquHeCIAiCUL/JLHfQY4W0tDR8fX3Ztm0bvXv3tmqLi4sjIiKCw4cP07Zt2+Ll2dnZ+Pj4sGDBAkaNGgVAYmIiISEh/PXXXwwYMKDc/ep0OrRaLdnZ2bi5uVXpZxJsO5+ay4Avt2M0l/71WzixC90bV3GejbyrsGiMdcDjush7YPgP4OR+rXNbYMGw0uvJlTBxM8zpVzLq4zqlGp7bA56Nqrbf9Zw49mpOTqGBN5ceZe3x5FJtLYPcmPdkZ7w16lvfcFEB/P0fOPBD6TaPcHhyHbhde7pt1MPmqbBrRul13QLhqU3Sn1Uh4yLM6loSuLzRqAXQbEjV7KeOEsdezUnRFdL1401M7NWIftG+NtexWCy8ujiGu6J9+WR46xruoSAIgiDUb3dU9Zbs7GwAPD0r/gTu4MGDGAwG7r333uJlgYGBtGzZkl27dtl8j16vR6fTWb2EmmM2W/j9wGWbAQ+A6RvOkp1fVLU7zbtqO+ABcG6D1A5QkAlbptpez2yEmIUQ0bd0m1EPJ/+oip7Wa+LYqz1Xc4tYd6J0wAPgeIKOq5Wd5pKXAod/tt2WGQfZl0v+nZsC++fYXleXKE15qSqn/7Qd8AAp8JKbZrutnhLHXu1ZeywJhVxGpwj71zYymYwuEV6sPZ4sytcKgiAIQhW7Y4IeFouF1157jZ49e9KyZcsKvy85ORmVSoWHh3XJUT8/P5KTbV/gf/LJJ2i12uJXSEjIbfVduDV6k4nTSfYvuOPS8yk0VPFFX2FWOZ261h9DAWSWUTow43zJU+ubJR2uVNcaEnHs1Z48vZGyxvWl51Yy0GjIL5nqZUtW/A3rFkjr25NxoXJ9sCUxpuz9lNXnekgce7Xnz6NJtA7WlluZpUuEJ1n5BvZcsJNrShAEQRCESrljgh4vvPACR48eZdGiRVWyPYvFguxaSbibvf3222RnZxe/Ll++bHM9oXqoFQpaB2vttkf6anBSVXEuFqcyRg/JZOB4rT8OzuDd1P66PtFSUlNbQrpUvn8NhDj2ao+roxK57f8SAfCpzNQWAAcXcHCy335jZRYHZ1C72l/XO7JyfbAlpLP9Np+m0pS0BkQce7UjPVfPwUuZdAovfwRrhLcL3hoVG09VU2JfQRAEQWig7oigx4svvsjq1avZsmULwcHBt/Ref39/ioqKyMzMtFqempqKn5/tuqRqtRo3Nzerl1Bz5HIZIzoEo1LY/vV77Z4o3Ko6kamLt5S01JbowVIyUpDyevT7t+31lGpoPRoubivdpnKBpvdVSVfrM3Hs1R4vjYqhbW3ny+gY5oG3q6pyG3b1g45P2W7zjgLtDf+nu/pD1+dtr+sRDh4RleuDLVEDQKWx3XbXu9L/CQ2IOPZqx+bTUhLddqEe5awpTXFpF+rBxlMpooqLIAiCIFShWg16WCwWXnjhBZYvX87mzZuJiLj1C94OHTrg4ODAhg0bipclJSVx/PhxunfvXpXdFapQkIcTvz7VhQCtY/EyN0cl00e1pal/GU+CK8vZU6rSElmS+wWZTEpmeN9n4HjDDYB/K3jwu5LRHyDduD2xGrSBcP8065sp9zAY+ydoxXBx4c6lUTvwr/uaMaRNIDcOgusV6c1XY9rh6VLJkQ9KR+j+InScAPIbRmgFd4ZHl0qBjusUDtBpAnR5VkoMfF1gO3h8pf2pY5WhDYFxf1qPNFFpYND/IKRT1e1HEMqw6VQKkX6aClckax/qwZXMAs6l5lZzzwRBEASh4ajV6i3PPfccCxcuZNWqVTRtWjKlQKvV4uQkDZfOyMggPj6exMRE7r//fn777TeaNm2Kv78//v7SxfSzzz7Ln3/+ybx58/D09OSNN94gPT29wiVrRRb72pOiKyQ9V4/JDJ4aFX6uapR2RoBUiYIsyEsDfY4U6HDxtQ54XGcyQk4S5F+Vbs6cvUtuyEwGqURn/lWQO0hPjG+8sRMqTBx7NS+30Eharh5dgQGNoxIvFxXuzpUc5XEjfZ6U1LQgSxr55OINznZKbxblS2VkCzKkKS8uPuBSTWU6c5KlRMVmg3Qcu/pLwZcGThx71c9gMtPugw3c19Kf4e0rNoq1yGhm4s8HeHNgU57qJaqBCYIgCEJVqNWgh72cG3PnzmXcuHEAzJs3jyeffLLUOu+99x7vv/8+AIWFhUyePJmFCxdSUFDA3XffzaxZsyqcqE1c/AlC7RDHniDUDnHsVb99FzMY+d1uPhrWksY+dqZa2fDJ2lNonRxYMEHkiRIEQRCEqlB2KvFqVpF4y7hx44oDIPY4Ojoyc+ZMZs6cWUU9E+zJ10tPic+l5iIHmvi54qNR4aSq1V8lQRBqSXJ2AYlZhSTrCgn1dMbfzRFv14aVJFQQbNl+Ng1XRyURXi639L7WQe4sOXiZQoMJR4cqTuotCIIgCA2QuFMVKiwrv4glB67w6brTmMxSwMpBIeP9IS0Y0iYQV0cxZFwQGpLzqbmMnbuPK5kFxctaB2uZ/VgHAt3LqOgiCA3Ajtg0WgS6IS+rbJINrYK1/LL3EgcvZdKjScNKuCsIgiAI1eGOqN4i1A1nUnKY+tep4oAHgMFk4d8rjnMhLa8WeyYIQk1L0RUyfv5+q4AHwNEr2fzfimPoCgy11DNBqH26QgPHrmTTMtB+eXZ7gj2c0Do5sDP2ajX0TBAEQRAaHhH0ECokV2/gmy2xdtvn7LhAocFUgz0SBKE2peoKuZSeb7Nt29k00vOKarhHgnDn2HshA7MFWgbdetBDLpPRPMCNnedF0EMQBEEQqoIIeggVUmgwk5BZaLf9cka+CHoIQgNSVlDDYoGCImMN9kYQ7iy7z6fjo1HjW8n8Ni2C3Dh2JRtdoRgxJQiCIAi3S+T0ECpEo1bSLtSd82m5Nts7hnvgohYJ125LfibkpULCIankZ0Br0PiBg8iNINx5ysrZoVbK0VQmx8/1UtGpp6SStgGtwC1QKmkrCHXI7gtXaRbgardKXXmaB7hhtsCBuAzuivar4t4JgiAIQsMigh5ChTg6KJjUuxErDydgNFtX3VEr5TzaJQwHhQh6VFpuKmz6EA7/XLJM4QAPfg9RA6QgiCDcQbw1anpHerP9XOkh+E90C7v1J9wmIyQcgF8fAn1OyfLw3jD8e3ALuM0eC0LNyM43cDophz59Kh+s83dzxMtFxZ4LIughCIIgCLdLTG8RKizMy5mFE7sQ5uVcvKyxj4bfnu5KiIdzGe8UynXub+uAB4DJAMvGQ/aV2umTIJTB00XFZw+15sF2gSiuVadwdJDzbJ9GPN270a2X2tQlwi/DrQMeAHHb4Z8vwaivmo4LQjXbF5eBBWm0RmXJZDKaBbixSyQzFQRBEITbJkZ6CBWmUiroHOHFkme6kZ0vzTP2cFbhXck5y8I1OSnwzzTbbRYLxCyCe96v0S4JQkX4a52YOqwVr/aPIt9gwkWlxM/NEZWyEvH0xENQZKcK1OH50P0FcA+5vQ4LQg3YeyEdb40KH1fH29pOswA3fvznArpCA26iJLwgCIIgVJoIegi3zNfVEd/bvJgTbmAxQU6y/faM82A2g1wMzBLuPM5qJaHqKjiVZF+232YoAJMY6SHUDfsuZtDUv/KjPK5rFuAq8noIgiAIQhUQd1GCcCuMRZB5CY4vh92z4MoBKR/H7XBwhsD29tsj7xEBD6F+yUmB+D3SMXRytRTwCOpgf323IHAQeW2EO1+e3siJRB3R/q63vS1/N0c8nR3YeyGjCnomCIIgCA2XGOkhCBVlLIJL/8Ci0db5BYI7w8j5UpWJynByh/5T4Me7peksN3Lxhkb9Kt1lQbjjZF+BhaMh5VjJMgdnmLgZfJtD6snS7+k/RSQyFeqEQ/GZmCyWKgl6yGQymga4sftCehX0TBAEQRAaLvH4WBAqKicJFo4qnVDxyr7bT7ToGw2PrQDPRiXLwnvBk2tFHgOh/ijKgw3vWQc8AAz58MtDMPpXaP4gyK8lQdX4wrBvocndNd9XQaiE/XGZuDoqCSqjpPOtaObvxokEHXl6Y5VsTxAEQRAaIjHSQxAqKn43mIpstx3+Gbq/WPkAhcoFGveDJ9dBYRbIleDsJY0CEYT6Ii8NTq6w3aa7AunnYejX0P896VhTu4LGX0zvEuqMA3EZRPm5IpPJqmR7zQPcMFksHIrPpFdk5UvgCoIgCEJDdltBD71ej1otKncIDUR2gv02QwGYDbe/D1c/6SUI9ZFRD2aT/XZdEqg10ksQ6hiDyczh+CyGtQuqsm0GujuidXJg38UMEfQQBEEQhEq6paDH+vXrWbRoETt27CA+Ph6z2YyzszPt27fn3nvv5cknnyQwsJJ5DYQ6JUVXyMW0PHZfSCdA60i3xl74uzmidlDUdteqT1g3+22ejaAgCzZ/DKFdwSMc0k5D0hHwbQZB7UEbAlX09E8QbldaTiFx6fnsir2Kt0ZNjybe+GkdcarOY1jtCho/yE2x3R7QWvrTYpFyfyTFSMeQTzNpFJUuEQLaSPlzlCLgLtxZTiXpKDCYaOp3+/k8rpPJZDT1d2WPyOshCIIgCJVWoaDHypUreeutt8jOzmbQoEFMnjyZoKAgnJycyMjI4Pjx42zcuJEPP/yQcePG8eGHH+LjI55I1FeJWQWMn7ef08k5xcuUchnfP96BnpHeqJT1NPDh2Rj8WkLK8dJtvV6HVc9DQSaEdIKf7pWG8l/nqIWxf5bc1AlCLUrOLuSZXw4SczmreJlcBjPHtOOuaF+cVNU081HjLyUlXflM6baw7lKVFpCSmc5/APJvqFrh7AkPfge/DIeB/4VGfUTgQ7ijHLyUiVIhI8K7aisNNfN3ZeG+eAoNJhzr84MFQRAEQagmFZoo/fHHH/PFF1+QmJjITz/9xDPPPMPgwYPp378/I0eO5IMPPmDLli2cP38eNzc3fv755+rut1BLCgxGpm84axXwADCaLUz65SDJuttI5nmnc/WDR36Hto+BwkFa5hEOQ76GhEPSjVq352Hje9YBD4DCbKnqiy6pxrstCDcqMpqYs+OCVcADwGyBFxcdJqU6j2G5HKIGwvAfSgIcSkfo+BSM+BE0PpCTDL8/Zh3wAOnfa9+Crs/C749KiYUF4Q5y8FImjb01qJRVm4OmWYAbBpOl1DErCIIgCELFVOhx3r59+yq0saCgID777LPb6pBwZ0vPLWJljO3cFgaThUOXMgn1dK7hXtUgbTDc/zn0eVPK4WE2w9InS0Z/eDWGlBO236tLgLxUUXpTqFVXc4tYuDfeZpvZAtvOphFexU+qrTh7QKuHILyHVLVFoQIXX3BwlNrzrkLGBdvvzbggHYOmIrhyQAo6CsId4sClTDqGeVT5dkM8nHFRK9h7IYOujbyqfPuCIAiCUN+J6i3CLTGYLBhMFrvtaTn1eKTHdQ7O4BEm/f3yPuvpLqZykpkW5VVfvwShAkxmCwUG+8lEU3WF1d8JmUzKy2GLsZz9Xz/G7OUFEYRakJhVQHJ2IVG+VZfP4zq5XEa0nxt7L6YDkVW+fUEQBEGo72456GGxWFi6dClbtmwhNTUVs9ls1b58+fIq65xw53FRKQj3ciYuPd9me8fwqn/KdUdz8ZHKy5qN0r9lcmm4vq0bN5kcXP1rtn+CcBNnlYIWgW6cSNTZbO8Z6V3DPbqJs7c0fcxWAFHhUDK1LKRrzfZLEMpwKD4TgEi/6qk8FB3gyrKDVygymqt8+owgCIIg1He3fOZ8+eWXefzxx7n4/+ydd3gU1frHP9treu8JCb33LlUUFRRUUERBRey9XL2/ey3X7rX3hmLveFUsqPTee4dAei+bbK+/P4YkJLsbEg39fJ4nD9mZM2fOLnMmO9/zvt/30CGMRiNhYWGNfgRnNrGhWh6Z0DXgvn5pEaREnMGpLYEwxMDAo0wZt38LA28M3Lb/LCmMXyA4iUQZNTx8UZeAhYS6JISQGXOSy8UaY2HQrYH39Z0Ju36QKiSFp5zQYQkEzbEpp5rYEA3hevVx6b9TfCh2t5ftBabj0r9AIBAIBGcyrY70+PTTT5k3bx4XXHDB8RiP4DSgX3oEH103gMfn7+JAqRm9WsGVA1K54Zx2RIecZdUUNEYYdpdUjnb587Drf3DBf2H8f2HFi5LZoiEaht4DPaZI7QWCk0z3pDC+uGEQj/20k91FtWiUci7vl8ytI7OIDdWe3MGp9TDkdsnodPl/wVwqCSEDZoM2QvL1GPOotE0gOEXYmFtJVuzxu79nRBvQqRSsPVRB3+PgGyIQCAQCwZlMq0WPsLAw2rVrdzzGIjhNCNGqGNEhhi43DMTm8qKUy4g2qs/cUrXHwhAjPZB1mQhuh1RG0xgPnS866nUcyM/Sz0dwyqHXKBnULopPrh+I1elBcWQOa06VOWyIliKj6uaQTAZyNeCVoqWUx2c1XSD4K9hdHnYW1HDVwLTjdg6FXEb7OCNrsyu5ZeRxO41AIBAIBGckrRY9Hn30UR577DE++OADdDrd8RiT4DQhJuQkrwifSsjl/saMwYwaBYJThGjjKRyZFWhOCQSnIDsLa3B7fcfNz6OOzvGhzN9WiNvjRakQvh4CgUAgELSUVosel19+OV988QWxsbGkp6ejUqka7d+0aVObDU5walJtdZJfZWPepnxq7W4m9EykU3zIyQ+Lbyl2E5jyYeuXUnnMLhMhoWfDA5bTIpWX3f4dVB2C1MEQ3UEqrRmRJsLqBQK7CUwFsO1LMJdB5wmQ2OuvixSWCqjOga1fgMsGfWaCLhwOLsSXvwFXQh8qk0axuEhN/3YxJIbp0GtE8THBqcGWvGrUCjlpx7lce6eEEL7a4GF3US3dk4WHmkAgEAgELaXV3xpnzpzJxo0bmT59OnFxccgCueEJzliqLU7eWXaQt5Zm12/7ZmM+vVLCeXt6X+LDTnHhw14Dmz+FBf9s2Lb1c4juCFfPA30U7P8dvr0WfEdK8277SnqYu/gNmH83XPiCqMIiOHux18CWz+C3hxq2bf1cEgav/h7CklvXn6UcFj8FG+ZIr2M6Qsfz4eMJ4LIiA9TbvyZebWDUpG+Y/mket47K4vyucejUQvgQnHw251aREWM47tEXmTFG1Ao5aw9VCNFDIBAIBIJW0OpvjD///DMLFixg2LBhx2M8glOc3CprI8Gjji151Xy/uYDZ57RDIT+FhbCawsaCRx3le2HlqzD0DvhuVoPgcfRxq1+XxJGDi6DXtBMzXoHgVKO2qLHgUUf5Plj5Cox7QvKxaSnl+xoED4DBt8MvD4CrSVlsp4X4X2/gn8O/5IZvttIndQRpUUL0EJx8NuVW0Svl+JuLqhRy2scZWZNdwazhwltNIBAIBIKW0upliZSUFEJDQ4/HWASnAV+uyw2676NVhyk3O07gaP4Cu38Kvm/zJ2CtAK878P6DiyBtCKx5U1qdFgjORnbPD75v8ydgKWt5X24nrH238TZ9hJReFoiaAtobbHi8PtYfrmz5eQSC40RZrYPCajvtj2PllqPpFB/CukOVeL2+YzcWCAQCgUAA/AXR44UXXuCBBx7g8OHDx2E4gr+C2+M9Iefxen1U21yNtsllMDQrikm9k+gYZ8TXNEIiGJ4gwsLxxlYVfJ/HAWoD9JgKKQP89/t84POCwww+z/Ebo0BwAqi/b3jd/pFNwdpCwxxS6aHThVIp5ugO0jaX7Zh9NcLnAUdN423e5ueW3Cfdg2psJ+keIhAcxZa8aoDjWq72aDonhFJjd7O3pPaEnE8gEAgEgjOBVscGT58+HavVSmZmJnq93s/ItLJSrL6dCBwuDwXVNuZtKmBvSS390iIY3z2BpHDdcUsvkctlXNwriV+2FwMwokMM1w1NZ9n+cnIrrfRMjcDu8gZ3lvf5oDoX9v4Ch5ZBdHvodRWEpYD6+BrA1dPpAljzhv/2DudJYfWbPwNHLWScA+c8AIufhMLNUpuoTKgtlh70tMc/lFkgaGvcHi+F1XaW7itlWKyD2KqNGA7Mx2eMQ9Z3JoSnSQaiSCJnQbWNBTuLWXuokvaxRi7vm0xax/HI8UHqINj7K9iroe9Myctj+zegCWn5gFQ66H4ZHFzYsE2uBKUW3Hb/9kot1bJwwMTAdpF/+XMQCNqKLXlVhOtVRBlOTBnl9rEhKOUy1mZX0DlBRN0KBAKBQNASWi16vPzyy8dhGILW4PZ4WZNdyfUfrcd9JMT1j10lvLpwP1/OHnxcDc56JofRIc6IQa1kUu8kZn28AZenYQzvL8/myxsG0SMl3P/g0l3w4Xip8gPAXmDVazDlY2g/rnU+AH+VqPaQMhDy1jZsSx0MXSbBJxc3rDLv/UV6eJv8Hvz2oFRZYsQ/YN170jblifmCKxC0JbuLarjmg3V8e2UyGT9NlURIQAaw4QMY8yj0vx60oewprmXqO6updUgRFX/sKuGdZdlsvbsrBoUK2VfTGzre8zNEpMNV39SLJi0m4xyIyJAqJYFUVWnI7bDsv35NTUMe5J2NtYztHEvCqW6aLDgr2JJXTWaM8YSZuquVcjJjjaw9VMnMoRkn5JwCgUAgEJzuyHwtzkc4c6mpqSEsLAyTyXRa+JUUVFk5/+Xl9Q8jR9Mu2sCXNw4iNuT4PRAUVtvIr7Jyw8cbMTVJdwFIi9LzzY2DG5ewtZTDp5OhaKt/hyod3LoOwlOP25gbUVMkeQ+se1dapZ72DXx9tRTh0ZSYjjD8PmmM1fnQYRxEtgNRtahNON3m3ulMaa2dKW+vZni6kf/jPbQ7vwrc8NZ1lOvSueq9tX4h9CqFjCUz4kn6fFTgY/vMhPHPgqqV95/qPMkrZ8tn4HbAxNdBoYIlT0PFAYjKonzAP/jTmolZHsLFvRKJOY73uLMBMff+Pl6vj56P/c747glM6p10ws771fpclu0vZ+O/xooKegKBQCAQtIC/bH1fWlpKaWkpXm9jP4kePXr87UEJmqfIZA8oeABkl1uosriOq+iRGK6jpMYeUPAAyKmwUmlxNhY9rJWBBQ+QfADK95840SM0AYbdC72nSyk3VYcDCx4AZXshtjPooyFrjORjIBCchlRanByusPLi+dFo/zcveMO9v2LrckNAz4CeyeGEHPo1+LHbvoAR97e+bG14Cox9TIrwwAfacMlfJ20oeJw4UOCVhTMKiDFqkJ/KFaIEZw2HKizUOtwnzM+jjk7xofxvSyEHy8xkxbYinUwgEAgEgrOUVoseGzduZMaMGezevdvPtFImk+HxCIPH443T3bxxqdt7/I1NnUcZGyaGaRnWPgaZDFYfrCC30lqfdlNPsIoodbhsx2GUzaBQQGii9Hv53ubb+nySUOJxg9MCSh3I5dKYZYqWp7p4veC2gUIDClFqU3Bi8RxJQ1PKfOAJLFgC4DBTN33lcrhhaDo947Xk1XpYl2NC4bIEP9btkFLEPO7WXeMuqzSX6uZkHcYYADRAbMt7C47HLRkW181hgeBvsPWIiWm7aMMJPW+HuBDkMlh3qEqIHgKBQCAQtIBWP3lde+21dOjQgTlz5hAXFydCK08CyRF6lHKZv7AAROhVROiPv99EQpgOg0bB/13QGblcxu87S/D6fMw+px1GjdLf1E0XLj3Q1BT6dyaTSdEUJ4uIDJArAleNMERLhor5G2Hzx1JUSNoQ6DAeNn4kleccOBtiOkltA+FxgykXtn4FeWsgMkvyTYhIk1azBYITQIRBTbhexZoCN13ThqHIWRG4Ycfz0SjlzBqayj39tci3f4125zpc4e24ZuwsvNbxsCGAGTBAu5Fw4A8o2g4DbpB8PjTNrIKbCiBnhTQ3VDoYMBtiu9SLHW2Go1aau+veh+rDkD4cul0qGbcK8UPwF9maV01iuBaD5sSK2Dq1gnbRBtYfrmTawBMUISkQCAQCwWlMqz09QkJC2Lx5M1lZWcdrTCec0y232epwM2fFIV74Y5/fvlev6MWFPRKPWwWX+jE43ewsrOHF3/exOrui0b5eKWG8Pb0v8WG6ho0+H+xbAF9e4V/ScsgdcM79oD1Jn73DLBmqLn3Gf9/E16TKMt9e33i72giXvg+/3AemfOh9DYx9JLDwUbAZ5l4grWbXIZPBZR9CxwtOjIHrKcrpNvdOZ9weL/O3FfH8gr0suSoM5dzzweNs3KjdKLj4DbwhibjzN6H+5CK/69Z9/UIUi/6DLHtJ42OVWpj6CXx/o5TOBnDJW9D1ksBpYaYCyTy4fH/j7T2ugHFPtJ3w4bLBzu/hfzc33q4JgWt/hfjubXOe0wwx9/4+E19fQYhWxW2jTvz3oU/W5LA5t4rVD4054ecWCAQCgeB0o9VLXGPGjGHr1iDeDIITgl6jZPqgNN67ph9dE0MJ1Srpnx7B1zcOYlSn2OMueADo1UrKah1+ggfAljwTy/aVN94ok0H6MLj+D6lagzZMWtG9fC4MvfPkCR4grUQPmA2XfwTxPaSxpQ6WXsd0gu9v9j/GaYbFT0H/WdLruiiQpphL4PvZjR8cQRJ+/ncz1Ja0+dsRCAKhVMgZ3TmWT2d0g70L4MovjpRfDpeinUb9E3pNA0ctLlMJ6h9vCnjdKj+djO+iV2D8c1IkhzYcOk+EKz6H5S80CB4AP90B5jL/wXjcsHGuv+ABsO1Lyby0rTCXSONoiqMWfrhNMlkWCFqJ0+1ld1ENWTEnJ1qvc3wIRSY7+VXWYzcWCAQCgeAsp9Uxme+//z4zZsxgx44ddOvWDZVK1Wj/xIkT22xwguBEGNSc2yWOvmnhONxedCoF4ScgraUOs8PNJ2tygu7/eM1hzu0SR8TRaS4aIyT3hymfSA9TCnXwlJATjSEKSndLK8y1xVC2R4r8GHiT/2p4HcXbYOQ/Gl5v/hSS+zVuY62Ccv+IHEBaga46BBEiPFlwYgjVqgjVOWDVC7BOCz2mwrmPSSLejnlQsBHGPIK8/fnBr1t7Na7yQ2gGzIYul4CjBla/AV9fI/VzNB6XVKo6Iq3xdmuZJBQGY+OHUmnptkg9KdkV3MOkaAvYqk6d+5DgtGFfSS0uj492MSfWxLSODvGSl8eGw1UkRwiDbYFAIBAImqPVoseqVatYsWIFv/7q7+AvjExPPJGGE5Ma4XR7sbk86FQK1Eo5Xq8PVzOGqk63F2+wzClduPRzKuCyS54dagPUFEhix4oXJeEjJD644FHH0aaxTVfFAXzHmA/H6l8gaHN8krGw0wwb5vjvdtkaX9eBevA4pOitkDhJNNj4YfDGbkeADnzNX/suuzTOANTaXchlYNCoAu7345hzWPzNErSerfnVyGVSifaTQahWRXKEjnWHK7nkBJbLFQgEAoHgdKTVy2h33HEHV199NUVFRXi93kY/rRU8nn76afr3709ISAixsbFccskl7N3buJKGz+fj0UcfJTExEZ1Ox8iRI9m5c2ejNg6Hg9tvv53o6GgMBgMTJ04kPz+/tW9NEACby8O+kloe/XEHMz9Yx6M/7mBfcS1qpZxL+wQvSzmuazwHSs2U1thP4Ghbgd0EBZvgh1vgs8tg4eMw6GbJULXTBKlNbbEUvh/MrDciA6xHhcb3nObfRhfhX5GiDrkSotr/rbchELQaTRhkjAy+v9tkFEpVM9etAkX0UdetLlyaC4GQyaSUsaPweH2Ue424OjYTFdh7umQufBRFJhtfrsvlurnrueHjjfy+s5jS2hbcX+K7B5/Dke2kOSoQtJJteSZSIvVolIpjNz5OdIwLYd2hymM3FAgEAoHgLKfVokdFRQV33303cXFxf/vkS5cu5dZbb2XNmjX88ccfuN1uxo0bh8XSUBLxueee48UXX+T1119n/fr1xMfHc+6551JbW1vf5q677uL777/nyy+/ZMWKFZjNZi666CIRdfI38Xh9rD5YwfkvL+PzdXlszqvm83V5nP/KMlYdrGBEx5iApfqSwnX0Tgln6rtruPfrrS17MDmROC2w/Vt4bxTs+A7y18OqV+D9MeDxQOcJYDhiorjnZ8nvoykyueSBsO5d6XXakMAVaEIS4KJXAj90jXxIhNULTjy6MDj/KalaSlNGPwy5a5B/fwOM/lfA69Yx7EG8+qNMRkPi4cIXpDnRlMG3+/n17CupZewrqynoMhv0kf7HJA+AhMbmooXVNq56by0PztvO+sNVrDpYwexPNnL/N9uOfX8xxMDQu/23yxUw4WUpWkUgaCXbCqpPeKnapnSMD+FAqZkqi4gYFAgEAoGgOVpdvWXGjBkMHz6cWbNmtflgysrKiI2NZenSpZxzzjn4fD4SExO56667+Mc/JO8Eh8NBXFwczz77LDfeeCMmk4mYmBg++eQTpk6dCkBhYSEpKSn88ssvnHfeecc8r3CxD0xhtY0LX11OldU/Hz5Cr+LXO4fj88G8zQV8vSEPj9fHuC5xDG8fw4PztlFSI4W1fzCjH6M7n0IPFlWH4bW+Uoh/U+K6Sj4eoUmS4LH/dxh6B+ijYM2bUsndxD5Sm21fQd5a6feOF0BoQuDzOS2SYeOSZ6B4K4Qlw4h/SP0Eeug7ixBz7yThcUt+MitfgYOLpGiNwXdASj94c5DkgdF1EvSYAhs+gJKdEJaMZ9h9uOL7oA07SvQwl8HSZ6H9uVKaS9FWSezrO1MyCe06GSLTAai0OLlu7jq25JloF61nzsWxxO75GMOBn0GlxdPvBhRdJjaaSx6vj7eWHOD53wN7jHx0XX9GdIht/v1aKiB/HSz7L9QWQVJfaQ5GZQauLHMWIObeX8fu8tDtkQVcMzidc7ucvL9tJTV27vpqC3Nm9GPMqfQ3ViAQCASCU4xWe3p06NCBhx56iBUrVtC9e3c/I9M77gjgkt9CTCYTAJGR0oPgoUOHKC4uZty4cfVtNBoNI0aMYNWqVdx4441s3LgRl8vVqE1iYiLdunVj1apVAUUPh8OBw9GQZ15TU/OXx3wmU2FxBhQ8AKqsLspqHXRPDufK/qmE61RYXR5W7C/nw1WHG1Wl/WRNLkOyotGqTl4YcCOKdwQWPEB6uNNFwOeXQ4fxMOgmcFohbwNcNldKSdGEgEIF0e1Bpjh2aU21ARJ7waXvSeVxVdqzNqRezL1TBIVSun7H/xccJum6NkRLZaXrTD93fg8HF0vCR+ZosJSjiMpEEdbkerdVwvr3JBGwxxSpOpO1ApY9D9U5ENOxXvQw2ZxsyZPu89nlVs79MIfxXSZzbv+p2N0+4sJTGBHaWMCosDj4akNe0Lfy2ZpcBreLRq1sJnDREAUdx0vmqG6HNIc1J8eA8mQh5l7bsbuoBrfXR7uTVLmljtgQDZF6FRtyqoToIRAIBAJBM/yl6i1Go5GlS5eydOnSRvtkMtlfFj18Ph/33HMPw4YNo1u3bgAUFxcD+KXSxMXFkZOTU99GrVYTERHh16bu+KY8/fTTPPbYY39pnKcjTrcHi8OD0+1Fp1YQqmssVNXaXbg8XkK0KlSKhgeHYwUB1e32+Ly8syyb3MrApfO8Ph++IKaEbY7HLfl1KFRSWL29FjwOqQyt4sj79jVv0ojPJ5kb7pkv/YAkXJz7SONojpD41o1NEyL9nMWcbXOvEfYaSVDQhjZci02xVgFe0EUGTC3xen2YbJKRZ1hbVGtS66SfOprODXt1QwoXQPfL/fuouxE4amD9+/77jzIKbXpLcXt9/LSjnJ92SK9fuNw/YsPn8z+uaR9H36tcHg81NjdqpZwQbZPP+SyOrDqr514bs6PAhFIuIzXy5EYJyWQy2seFsF74eggEAoFA0CytFj0OHTp0PMbBbbfdxrZt21ixYoXfPlmTL/8+n89vW1Oaa/PQQw9xzz331L+uqakhJSXlL4z61Mbt8VJsslNmdvD7zhKW7itDq1Jw7dB0BrWLQi6DrXnVvLs8G5PNxaiOsVwxIJWUCB0ymYwog5oQjZJah39URKhWiVIho6zWQYRezaTeSbyycH/AcVw5IAWdqtWXWuvw+aRV5U2fSKvVQ24FdQisfQdsFZA5FvpdC+FpEN9N8h8IJH5Et4faQv/tXSeD9ux9YGorzpa51whLuVQKdtWrYKuW0kD6zGhskltTBNlLpIgJrxt6XAFdJkqpUEcoqLYxf2shP24tRKOUM2NwOoMzo4gN1bbdWGM6SV4XgSqaRGRQ5TOisLsIPVpM0EVAVBZUHPA/RiaX5tsRwnQqOieEsLuo1r8t0DvVPwIq0qDm4l5JvLE4QP/Alf1T0KgUeL0+8qusfLY2l6X7yogwqLnxnHZ0Tw4j6gRVuTqVOSvn3nFiW76J1Eh9o0WCk0WHuBC+Wp+H0+1tPtpJIBAIBIKzmOP8JNoybr/9dn788UeWLVtGcnLDl/z4eGklvbi4mISEhhX20tLS+uiP+Ph4nE4nVVVVjaI9SktLGTJkSMDzaTQaNJoz/0vw/lIz5bUO7v1mK6W1DWHNm3KrGNUxlumDUrn+ow3123cX1fLp2hy+v2UomTFG4kK1PDm5O3d8sdmv7/vO68jj83cTZVTz2MSuTOmXwrcb8ymotjVq1ysljN4pJyCVo/IgvD9WKp857G44tBy2fNawv2Sn5Ddw/R/Sg+So/4NFjzfuQ6GCsf+BP/7deLs+Cobf23hFXPCXOFvmXj3WSlj4H9j0UcO2kh2ST8asPyG6g1Ql6JuZkLemoU3RVinCYuZ8CEumoMrKlHfWNJpfm3K3MDQzipeu6EVsSBsJH8ZYGP0I/Plw4+1yJaWjn2faZ9nMGOzhsr7J6NRH/nyExOG68DVUn01sSI05gmP4Q6CJou5/PMqo4enJ3Zn6zhocTUpe3zoyixij/7WhUsi5ckAK8zblU2RqbFraJzWcHinhAGSXm5n0xqpGIu3qgxVcPSiNe8d1ILwtImNOY866uXcc2VZgIv0km5jW0SHOiNPjZVdRDb2OzAWBQCAQCASNadGywDPPPIPVGjh1oSlr167l559/blFbn8/Hbbfdxrx581i0aBEZGY3LHmZkZBAfH88ff/xRv83pdLJ06dJ6QaNv376oVKpGbYqKitixY0dQ0eNsoMriZP62Qn7aVtRI8Khj8d5SauxuogyNHwRqbG6e/XUPtXYXSoWc0Z1i+eHWoYztHEu7aANjOsfy7tV92Z5vYnV2BfO3FZFbaSUpQsfXNw7mH+d1pEOcka6JoTw9uTtvT+9HXFgbrkQHwmGGP/8jCR4KtZS3f7TgUd+uBhb8E3we6Hc9zPgJMkZIq9Q9r4IbV0gGhyMfhKQ+0gPp8PvghsUQGaQkp0DQHDX5jQWPOuwm+P1hKeUlf31jwaOOqkOw9Uvcbjefrc31ExQBVh6sYFdhG3ozqA3Q5xo818zH024URGXh6DKFgiv+4F8bdBwoNfP4/N2UmRuqRZTVOnh0k4aCqX9g73oFRLfHmz6C4knf8KlnLMX2xtp6l8RQfrljOFf0TyEzxsCgdpF8ct0AZg3PIEQXOO0nOULPtzcP4b5xHWgfK91fnrm0O29N70tcqJYam4v/zN8dMCrtkzU5lJyqpbMFpx12l4cDJWYyThHRIz3KgEohY2NO1ckeikAgEAgEpywtivTYtWsXqampXH755UycOJF+/foREyOZ2bndbnbt2sWKFSv49NNPKSoq4uOPP27RyW+99VY+//xzfvjhB0JCQuo9OMLCwtDppBSLu+66i6eeeor27dvTvn17nnrqKfR6PdOmTatve/3113PvvfcSFRVFZGQk9913H927d2fs2LF/5TM5I6ixu0iO0PPZ2tygbRbvKWVARiS/7mjsffLn7hJqbC5CtCqMGiVxoRoSw3X0Somg2GTjvm+2UmNveLj435ZCeqdGkBShY/aITC7vn4JcBpEnKqTcXg17j3hvxHWDvHXB22YvklIMItIkw8WEXuC2S6kwdZEc3S+HzDFSiL8uPLj/gkBwLPYuCL5v/wLJBHTjh8HbbPkMV7dpfL+5IGiTL9blMiTzGEaerUEfwSFjb36M+BeJKTJ2lHv49pMS7C4pMsPp8XKo3FzvZ2CyOflsQzHfbZUzufv19Og6m0KLjy/mV1NWW0RyQiJpUQ0PiGqFgsxYI49O7IrZ4UatkPv5DAUiKVzHzSOzuGJAqt/9xWRzsXx/WdBjF+8po2O8qFAi+PvsLqrB4/OdMqKHUiEnM8bIxpxKrh8mxHmBQCAQCALRItHj448/Ztu2bbzxxhtcddVVmEwmFAoFGo2mPgKkd+/ezJ49mxkzZrQ4hPatt94CYOTIkY22f/jhh8ycOROABx54AJvNxi233EJVVRUDBw7k999/JySkwRDypZdeQqlUMmXKFGw2G2PGjGHu3LkoFKdItZC/ic/no9LixOeTSsUq2iiPWCGHfmmRLN5bWv9A00CDH4oP+Gp9nl84eh11o3G6vZhsLpRyGRGGNgglt9eC2woqw/GrtKANBUIl3wWXRfIz8DhBbZRWvetwOySxRKE6q80QBScYmSygoWn3pDCmdNGhUUCZxxCoSWMsFeBzSwapLRDxPF4fr64sCbhvUEYEHYwOMJeCPpq6e4Xd5eXzTWV87vceGr+stbuwu7zo1QqiA6SzAFRbnbi9PsLkDlQeG6j1oAlBIZcFPaY5jvX5eL0+Kq1S9EqkXo1cfqwPVHC2sqPAhOIUMDE9mqxYIxtEpIdAIBAIBEFpsadHjx49eOedd3j77bfZtm0bhw8fxmazER0dTa9evYiOjm71yY9VHQQkE9NHH32URx99NGgbrVbLa6+9xmuvvdbqMZzqFJtsLNhZwmdrc3B7fFzcK5HL+qaQFNG8v0SYTkVhlZXzu8bz5frA5R4n9U5m2b5S3pnel3eXZ7PyQAUA47rEE65vuDTC9Som9Ezk2435AfuZ2CuRwxUW5izPZum+csL1Kmaf046BGVHEhPyFaA+bCcp2w9LnJK+O2C4w4gGIah9Y/NBFQOeJUonNku0w8h/B+84a27hcbE0R7PtNSjcIT4WNcyVD1PgecM79ENkOLKWw5h1pZV4bBkNuk9JijP6VJgSCRnQ8HxY/EXCXu/14qrxGovpej/zAwsDH974aVUgMk/sk88biA+jVCj68LIX2VSuI3DkX3HbcXSajrJ0hXb9NqS2C/X/CunfAaZHmSb/rpEinZojQq0iN1DeqyBShV/HxZcmkFP1K+Lw7pY29riK966X0T4tgfYCHLoVcRqd4SaCusbnYV1LLq4sOcKjcTKf4EG4f3Z7MGCMGjXS/qTA72JhTRUFJKRcl1iLf8DKU78UX0xHZiH9AdEfQ+ldACterGNkhhsV7A0d7jOoYfK4WVtv4YUth/f1tav9kJvRIJCFcePgI/NlRUENqpO6UMDGtIyvWyPxtRZTU2IlrS2NjgUAgEAjOEGS+ligPZzg1NTWEhYVhMpkIDT11QqCLTTau+2iDX85+YpiWb24aTFJE8ytNe4prKKt1cP832yhuktM+skMM53ePJ8qg5pbPNvHqlb156pfd1NjcfH/LENrFNBYXciosXPrWKsqPyuUHuLhnIneObc+E11ZgcTau+DChZwKPTehKZGtWZl122PYV/NSk9LFMBlM+gY4XSNEYTak4CHPGSsaRQ+6QUl42NUmz0oZJRqYxHaXXtUXw1QypDG1sZ1jydJNzyuGKL2DlK5C7qvG+jhfBxJfBENPy9ybw41Sde22GtVIyzN3wQePtughyJ/3ARZ8Xs+jGTkT9djOynJWN20RlwjU/HjEytTH13dU8MSaaYZvuQVm4vnFbY5x0bR8tZtSWwHez4PCyxm31UTBr4TF9atYdquCq99fi8kh/In64Op2ei2dAeZMqTVGZWK/4nt6v7PKLBvv3RZ25sn8qCrmMeZsLeGje9kb7ZTJ466q+nNsljlq7i+d/30deWTUvdc8h8rdb/Ac1+X3oeknAaJUDpWYmv7myUeodwLVD0rlzbPuARqaF1TamvbeGwxWNPasyYwx8ev3AM1r4OOPn3nFi/CvLiA/VMvuczJM9lHoqLU5u/XwTb0/vy/ndWllKXSAQCASCswAhenDqfvn7aWshtweonAJw19j23D4qq9lUF4/XR3mtg1Kznd92lLB8fxl6tYKLeiQiAx79aSc3j8xk1YEKau1u/nVRZ9KjDCQfKVnblPxKKz9uLeTXHcUYtUpuGJZBl8RQ/jlvO4uCrLD+dNtQuieHt/xNV+fCGwPA5W/aiD5SMhsNSwp8bFWOJJjs+RkG3ii1X/eelLrS/lzoPV0qWVv33vYtgM+nwJVfwtdX+1WeACAkAUb/C3641X/frIWQ3K/l703gx6k699oUSznugi3IV7+O3F5FbdoYSttdyo3zyzhQaiEpXMvC2Z3QFqyWStZ6XNBzGnQc3+haLzHZCcn9E/13VwU+z+DbYMwjoDzycH9wCXxyceC2fWbC+GdBFXxV2On2kFtpY+7KQ5gdbp5IXo/xz/sDtvWd/yyFHa7mvRWH2JBTRVKYjhtHZJIVayBUpya/0srYl5YGSKOTIkh+vmM4ZoebcS8t4/urUuk9fzw4ApS11YTCLasblfKtH4PPR16VjW825LF4bylRBjU3nJNJl4SQgP5CPp+PD1cd5j8/7Qr4np6a1J1pAwNEz5whnBVzr41xuD10fXgBVw9OY1yXU0tcuO2LTVzWN5mHxnc+2UMRCAQCgeCU45QoWSvwx+b08FWQtBSA/20u4KqBac2mjyjkMrRqBY/8sAulQsbQzGicHi/vLsuuD1v/c1cpY7vE8urCAySG6UhpJk85OVLPjSMymTYgFaVCjlGrpKDKyuJ9wQ0E/9hV0jrRo7YosOAB0qq5tTy46BGRBsPuhf6zpCouGiOkDQePQ3pYUhx1uXtcUlUNtUGq+hJI8KgbjybIA8HO/wnRQ3BsDNFs1/bl94h/Ea6GNYUuliw9TJ3cXFBt56DdSNful0ninNcLev8yz3EGOez0c8xoYMe3kvARmgA+H2z5NHjbXd9LaWCqxKBN1EoFWbFGHp7QFY+1Ct1X9wZtK9v6OUk9LuefF3TG4vSgUcrRqxvmW0mtPaDgAVBldWFxuFl1sByAWHlNYMEDpApMlrKAoodMJvks3DmmPdcNzUCtlNenzQSi2uriuyApewDfbMjjgu7xZ32pW0EDe4trcXt9tDtFTEyPJjPGyOac6pM9DIFAIBAITkmE6HGqIvOhbMZMTyGXN/UHDIhcBkqFjHWHKll3qDJAPzI83oa2x0IhlxF+tEmpTIZcJsNzVMDQ4HZRjO4ci1IuI8qgpsrqJFynotzswOr0oFNJBoYBzQJlzedJu30g8/pQBBusQtHYaFRjAAJ9QZWBXAU+b+B0mUZNm5xLoYYuF0O3yVJ0iUItVXmxVwNyKeVFfurkewtOPh6vj7dWBTYGBZDXXWPasOCdyOTSNRu0E2XDtSqTSddl0LZNrnmPS4qIkslAHyPNoyOolXJQKaX+m+nP4vSCuwad2wIKDahj8Hi8VFicGDVKzu0Sx6I9pXi8PkK1Si7ulURWrBGTzYVGKUd1ZM4omxpQa0KkikqxnaUSv8rmU06UCnkjI2Wfz0e52YnX5yNMp0KrkvqXwTHusbKG/xeBAMnPQy6j2cWBk0VmjJH/bS7A09zfR4FAIBAIzlKE6HGKolMpuXpwGkuCRFFMG5hKlPHYK5AhWhVXD0oLKHgAXNg9gR+3FtI7NfwvrWiG61SM7xbP/G1FhOqUvHB5TzblVPPx6sM43V4u6BZP54RQFu0p5b1l2ZSbHfRJjeDGEZkkheuID2sSXh+SID342U3+JwtNZGOZgvV7DnBp32QSwv5Gvr1CCf2vh13/k6I9VHpwWf3bRbZrPJbYLnDBfyVPge3fwO4fpe3dL4d2o2DRE9DpQugxBUKDr6ILzi4SwrQY1Ao/3xuQSrFGtqTakUIpmZDu+l/g/b2vkQSLOvpcA1s+C9y219UNfjTVeZLnyLavJGGl93Tp5+hoCl2YFEGVtzZgd74eV6Jzm/Ctn4NixzcQmoR72L3sVXflxnmSCfO5XeL4cGZ/ftxSyIU9Evh0bQ6/7igiNkRLRrSeAe0ieXtSKpEGl+Q7Yq2Q5tTgW6Wyvrt/ksyDw5Kk/cZj++mU1Nj5dXsRc1cdxuL0cG7nOGaf047USD3hBjXTB6Wx9dttAY+9Zkh6i0rpCs4edhaaSI7Qo1GeepXhsmIM2FweDpaZ6RDnb/YrEAgEAsHZjFiOPoXpnhzGiPb+X+w7xYdwQff4gL4bgRiQEcmAdP8yq92TwkiK0JFbaeXpSd3/UplZg0bJfed1JNqo5vGLu/H8gn28tfQgeZU2SmocGLQqXl24n3u/3sqe4lrKzU5+31XCFe+uZlehibJaR+MOjfEw6W3/iA+FirJzX+PRJRU8//s+rp6zjiJTkDSYlhLTCTpNhPVz4Nz/+Ed0KDVw7uMNho8qHZz3JDjNMG82rHkTTPnSz4qX4MfbYPg98Ocj8NkUqCn8e+MTnDHEhGp4YUpPv0tMrZDz0tSeLa+4ENsZOk3w3x6VBb2vahShQVQWdJ/i3zYiHQbcIAl31XnwwXmw4kWoKQBTnmTo+9EEMBU0Pi5jOKQM8u8vqQ8yfSTyd4ajyBoNXjcUbUH5zdWk7Z3DhI5GimvsfLImh/u/3cqs4Rnc8MkGFu4updzsZFdRDbd/sQWftYqx+a+j+u0+OO8pKV2t//Xw5ZWST4+lDEp2wvc3wR8PS+luzVBaY+e2zzfx6E+7OFxhpazWwefrcpnw2gpyKi0AnNMhhl4p4X7H9k+PYGCGKE0taMz2AhNpUadelAdARrQRGbAlr/pkD0UgEAgEglOOVhuZWiwWnnnmGRYuXEhpaSleb+M87ezs7DYd4IngVDZ0K62xs63AxMerDuPy+pjaL4WB7SJbHeVQUmNnS141n6zOwevzMbFnIjEhGvYW13JRz0SSw3WB001aSLHJxvL95dx/1KqpWiHnzel9mPXRhoDHdE0M5dnJPeiW3CSk32mF6hx8a99FVrYba0wPKjpN55FltSzaX13f7KWpPZnU2z+3v1WYS6FoGxRvg8Teki9CxUGI6ypVilnxImSdK6XMmMvAbZfEkaXPBu7vnAcgZwXkrJKqzXSZ+PfGd4ZzKs+9tsbqdJNXaeWjVYfZX2qmZ0o40wakkhyhQ92aleO6a3bt29L12GuaVEI5kNeNuVQSCta8BS6LJIJkjZGiOLweWPEyLPpP4PNc9DL0u7bxtpoi6dre/Angk6KaNKHw872SGJg5RhJmVr8utZfJOHzlMkZ9mFfvYXLD8HbsLa5h2f7y+m7VCjkLr44h5csx0oaO42HkQ/Dj7VC0NfD4bl4NcV2Cfkwr9pczfU7gyJRLeify1KTu6NVKSmrsbDhcyefrcpEhY/qgVHqnRpzxpT/PprnXFrg8Xro+vIArBqQwvlvCyR5OQB74divndIjhyUndT/ZQBAKBQCA4pWh1esusWbNYunQpV199NQkJCS2ONhD8NWJDtYwN1TI0Mxqvz9esMV9zxIVqOa9rPMPbR4NPynt3uD2M7BjbJvm/4Xo1P28rarQtI9rAjoIAaSpH2FlYQ5XNicfrRXG0B4ZaD7GdqRn1JN+s2c+GQjt/fpiP29tYn/t2Yz7ndYlH/xc/E0AKl0/uB4WbYP/voIuEBANUHYbPL5ceDB1maRU9uS8cXgEH/gze377foMM46cFwy6fQ4TwpYkRw1qNXK+kYH8pjF3fF7vKiUylQNlN9KSjGWGg/FtKGSJ40GmPzbY2xkDpQupY1R4W926okkS8YW7+ArpOl1JY6QhOgJl8qpwuS8FJxsGF/9iIpraYOnw9d5S7iQmLqy2Yv2lPC5D7JjUSPzFgDhsKjykLv/VWqYBNM8AA4vLxZ0eP7zcFNShfsKOEf53dCr1YSF6rlwh6JjOoYCzIaGbAKBHUcLDPj9HhJjzr1TEzryIg2sD0/+N9cgUAgEAjOVlr97e7XX3/l559/ZujQocdjPIIg6NRtk0N89Bd6tdL/gavW7sLscKOSy3AfCeKJNKiRyaDC4gSfj3C9ut4MsA65TIZG1bQ/H8ZmBAmZTDISrLG5cLi9hOlU6I4anwsFeyu9XNxew1Vdk1ie5+TzzZWYHW4AtCpFG/mFyiQho6lfQcZIGHonGKKlB8baYqkqhqIZEUOpAY/7yO+6YxqzCs4+VAoFqqZmncGwlIPbCSoN6KOwudyYrC5kyIgyagOKJnVzWCGXEWPUSMK06khIvrlUSj9R6aVrszlBTqkFuZxqqxOby4NKISfaqJGqK61/P/AxchXgg74zIbk/uO2oQpLweO31TTRKBUq5jPvOiadXnJJym5ffDnvxKppEVshk0k+wYERV82kGzd0zNarGRtBOt5cauzRvlXJZ6yJvBGcFOwpqAE7Z9BaQUlw+X5eD0+0N+PddIBAIBIKzlVaLHhEREURGilznMw27y8OBUjPvL89mcp9kfttRxK87ignXqXlhak9+31nMNxvycXm8nN8tnltGZpEWpa+P9FEr5VwzOJ0FO6UKFckROv7vwi7o1Iqgzy3DsqIxqJVMfmsVFWYn53eN55ZRDf1GeSt5InYxmtXvgb2GQemjmHHVvdy70MyanBquGZyGVtUGq7Jel7Q6fbToceWX0sPdnw9DZTZEtYcRD0jh/MZYyF0duK9ul0qmi3CkdK4wQhT8BaxVkLcGFj8pRVJEd8A96l+sNKdyz4+HUSvlTO2fylUDU0kMl1Ld7EdMDJ9fsJe1hyqJNKiZfU47zu8aT6y8VopkWvGCJN4l9oaxj8Hof8OnkwMOwdTjOkqqfHywcg8/bi0kMVzHXWOzGN9pAoolTwced9dJEJ4umZD+9iCoDYT1uY73Jk1g+td5mB1u7hudTA91MRGrnkKxZSMY4xje+zZ0mcNhyVE3i9w1kpHpwUX+55HJpEiXZri8bwqfrskNuG9qv5R6I+i8SivvLc/mx62SB8+k3klcNzTjlKzQITh57Cw0kRCmPaUjgdrFGHB5fOwrqaVbUjOVoAQCgUAgOMto9VLA448/zsMPP4zVGqDSheC0ZUeBiYvfWMlFPRN58LttfL4ujyqri/vO68i9X2/l7aXZVFic1NjdfL0hn4vfWEleZeNroGN8CBN6JCCTwROXdOPeb7by+dpc7h7bwe98UQY1d4xpz/rDlRwqt0r9bsxn4hsryK20Qm0Jsm9nolnymGQI6jSj2vcTSV+fx4ujdUzunUiXhDb4Umcphx9ulX5PHyb9O/Y/kg/Ct9dC8XZwWqBoC3w5TRJGEnpD+nD/vtKGSZVnyvZK1VxiOv398QnOPlw2qerKF1dI15/LKhmDfn4Z/WoXMSgthHKzkzcWH2hk6Lu7qIaJr69k8d4yrE4P+VU2Hv5hJ+v3ZONd9Dj8cIskoDgtUprWnLHg80DPK/2G4MwYyzpXOya+sZLx3eJRymUcKDVz2+dbWF2qwTv8Pv9xh6XAsLslEWX3T+CohdpiFEufovvym3l9QgITeyZwju4w0V+OR5GzTBpLZTZRC+9Bk7cC19gnG/rb/CkMvk2KtGrK+c+AMa7ZjzElUsfVg9L8tmdEG5gxJB2VQkF+lZXJb63i49U5VFtdVFtdfLjyMJe/vZqC6r9plCw4o9hxCpuY1pEWpUcug6351Sd7KAKBQCAQnFK0yMi0d+/ejbw7Dhw4gM/nIz09HZWq8Ur2pk2b2n6Ux5mz3dCtwuzgqvfXolcrGd4+mlcW7gegY1wIl/VL5smfdwc87tqh6Tw0vlOjUPBys4Nik52ftxXy1lLJ1Hb6oDRGdIjmtx3FVJidDM6MYlSnWEpMNmZ8uJ4mVh3cOKIdD3aqQPbRhQHP680ci/3i99GHRvz9N1+wGd4bKYXyn/+MVBY0sa+0ze3wb682wuyl0gNdxX6pfKjPB72vlvbv/B/0vQaiO7aopObZztk+9wJSlQNv9A9y/RnYNnEBEz9tiGB4fVpvhmRGcc0H6+pD8OtQyGUsmpFA2hcjA58rIh2u+QFH0S40Wz/Gi4LSjtPY4kzmjp8KcXq8jOoYS3q0ng9XHgZALoPN9/cnzJqLb/17yCzlODpejKrdMGQrXkK2+eOAp7JP/QYSeqD9eLwUPRWA6uvXYDHXELP7E9TWImydLkObMQDZgYWwfwGEJkkVXSIyQHvs66XS4iC7zMLHq3OosbuY1CuJARmRJITr8Hh9vLXkAM//vi/gsf+6sDPXDc34WwbPpzJi7rUcr9dH90cXMKFnIhf3CmAYfArxj++2MTQrmqcnCzNTgUAgEAjqaFGc5iWXXHKchyE4mdTa3ewprmXW8AyW7y+r3947NZzl+8qCHrdgRzE3jcgkLrRB9Ig2apABi/c2HPfpmhy+25jPiI4xxIZqOVBmxuX2Ehum9RM8AIqqbfj2/ESwRw159kL0PgvQBqJH9hLpX7cd5t8FoSkw6c3AD5wgVaiwVkjGkHFdoPPEI94IaumY9ueKlBbB38Nc2sz1ZyFSVtto0/ebC+iWGOYneAAkhGnRlW8Pfq6qw+B28EpOBhWK+/HhY+lv1ZTUNJiALt9fxuQ+vepFD68PNpb6GN2pH7LEXuBxo1Jo8NUUIN/xTdBTaffMg6RuQQUPANehlVy/qT2Z0TOI0Mop2uHlqQ5JxA2cLaWgyVWNy/Ieg0iDhkiDhl4p4Xh8PjRHCbQmm5NfthcHPfanrYVc1jeZcH3rS3kLzixyKq1YnJ5T2sS0jvQofbMG4gKBQCAQnI20SPR45JFHjvc4BK3E4/VRXuvA4/OhVyuwuTz4fJKhqA8fKrmc6JCWVw2Ry8Dl8TUyKHV5vM2aAerUStweL8UmGw63F6VCRoROjdPjQdfE6NTm8vDbDukBY3BmFLFGDU63N1C3UiUXzVErjykDpBB8XSSU74Nt30jRFW4nWMul3zUh0sqvz3fEcNQrRW8YogIP3lIuPVhmjoKQOMkUsbZICqmXH0O0qBM1mppAiiotgiPU2CQzUZkMoo3qlhuXgiSggVQ2ufd0CEkEU55UJrZsLz5Z49u2UaNEIZehkMvwNFERPV4fat0xVvHlSnolyIhIdJMYaeRfI6LxeH1sL/fw79/yqLI4wQf3nhPP4GQ1NpePmBBV/bHIlchriyVvHJVOSs8JhDZM8uJoBp/aSI3Nzc87SgHIjDFQH2ih+uslZJUKud8fO4VcjlalQKuSM6FnIsOzovEBy/aVMX9bEXq1sk0qWwlOf3YWSiJCRvSpL3pkRBtYvS4Xl8eL6q9UhxIIBAKB4Ayk1Y5c7dq1Y/369URFNX6YrK6upk+fPmRnB1/FE7QNpTV2vtmYz9fr8/j3hC6s2F/OvE35WJwehmRGcd3QDH7YUsC4rvEMzowiIshKpdPt4WCZhbXZlYzoEMvC3SXcMjKLVQcrAFiyt4x/XdSl3py0KTeNaEdZrYO5q3L4bWcRXi+c1y2OawalM31QKpvzqgMeN6FHAlanhwU7A6+y9k4JQ5Y5STJdnPAKmApg2X8lUSK+J5z7qPRgtfBR2PQJuCyQdS6c95RUhWXlyw1txz0ueXBoj5TqtNdA0Wb4/WGp7GdYKqx+XUpViciAAbNBFw6GGLAEiHIJTwW9MPIVBMbp8ZBdauGZX/ew/EA5BrWCaQNTmTEknYQwXcs60UdL17JcAWvfgapDEJUFA28Gp4X9lsbi2tWD0og0qDm/azw/b28oGx2uV/HkpO44dFWgUIPH6Xcqb+pQ3F4ZYw+/hLzzBbB9IWz7GpxmhqaPZMGVj7LNEkF7ZREXlP4HxfploA3D2/c6CJkFcjXs+1Wan7GdoPsUqYxtIHpNkyoaZYyAQ0v99ytUVId3pdDUkLozY0i6VDHmOBCmU3H76EwUcjlfb8jj/m+3AXBe13jeu6YfciBEK6K2BFLlliijmlDdqX89pEdLZqb7S8x0SRRpSwKBQCAQwF8wMj18+DAej8dvu8PhID8/P8ARgrak3Ozg3q+38t8Fe7l9TBb//W0vc1cdpsbuxuP1sXx/OTd+spGLeyXx4LxtfLM+D4fL//8L4GCZhYmvr+CF3/dy7dB0XB4vHq+PMZ1jAalEbWmNnQk9EvyO7ZcWQbekMGZ/spH/bSnA7vLi9Hj5aWsRN3yygU4JoZzTwd+AcHSnWAwaJed1jScp3P8hsG9qOOd1S0AWlgRXfAH7/4QlT0tmpj5fY0PRnFXgqJHKycZ2hkWPw68PNG770QQ4tFh67fNB9mL4aCKEJkir1D/dIUWP+HxS2P1vD0LBJpj8vn/khkoPk9+VfBAEggBkl1qY+PpKluwrw+P1UWN38/bSbK6fu56SGvuxOwCc6jC89hr49R/SNenzQfl++PkevF4PKwsaojmmDUihXbQBg0bJP8Z3IjGsIRrisYldefLn3WwtdeO78EX/KAtDNLLzn0K98F/I0wbCoidg3XtgN4HXgyJ7IdoPx9DfWE7EvKkoDi+RIqhsVchXvABfXAmlO+HH26A6B/b/IUVOxXb2e0+WAXdg1iZIguKFL0qi4tHIZFSe+yqvrm1I0emfFsF5XeIb+Um1NRnRRu76agvztxXhcHtxuL38uLWQe77eIqq3COrZWWg6LVJbANKjDMiAHYUixUUgEAgEgjpaHOnx448/1v++YMECwsIaKmd4PB4WLlxIRkZG245O4EdBlY3lB8qJD9Xi9vrYW1Lr18bp8TJ31WEu75vCi3/uY3z3BL8v8DU2F0/9shuXx4fL4+ahedt5bGI38qutXNonmWkDUvl5exG7i2q48ZxMrhyQyneb8nG6vZzXNZ64MA0/byuitNbfe6Da6uLrDflMH5jG9IFp/LK9CJlMxvnd4okyqMmtsuLDxwPndWRK/xS+Wp+Hzenm0r7JdIoPJS5UC2ilahC7/hf4g1jyNIx8EH68XVoRTx0MK14K3PbXf0BSvyO/PyD92+sqmDc7cPv5d8LtW+DG5bBjHpTthrju0PUSKdJDIAhArd3Fs7/twenxT9vaVVTLnqKaI9f2MbCUIV/xYsBd8mXPMvPK8Zg9KVzRP4W0KAORBimSKzVSz7c3D2HdoQo25Vbh9vrIqbDQKSQE2faFMO1rKRKqthgSekFsJ2QL/gl9rwWPHUp2+J/Q45TExB5TpYiooynaArWFUglnc6kkiHx/E4x/Flw2fNlLcKnCKG0/lfl5atoX+xgTAURn4Zm1COuehYTkLcZqTMXV7Qry3RGoc8uY0EPHhJ6J9EwJb9nn9Rdxe7x8v7mASot/BEy52ckv24uYPSJTpLic5fh8PnYUmBjdKfZkD6VFaFUKEsK17Cr09/gRCAQCgeBspcWiR52ZqUwmY8aMGY32qVQq0tPTeeGFF9p0cAJ/VmdLqSfdk8NYd6gyaLuVB8q5tE8Sc1Z4qbQ4/UQPs8PNygPl9a8Lqm3c9OlGOsQZ6RgXwoCMSLJijGwvMLHsQBnhOjUhWiWlR3xEluwtY/WRNJhg548yqPl49WGGZUVz68hM3l2ezaqDFeRX2fh81kCGZEUTadTQJzUCn8/nv6Ib6CGsjpoCySMAICQBKg4007ZQSmvhiN8HSA9oriBllz0uqf/UgTDqISmSRN4KTwbBWYnZ7mb5/vKg+3/ZUcyIji14cDKXBkxFAcBlI15l4dlL+wbcnRiu45LeyUzsmcgdX24hPkyLtnw77PxeKiObPlxKz9o5DxYeMTgd/TBs/ijocGSHlkDvqwLvzFkF8d3hwELpta1KEhPD0yib/A3/XFzDqq8qsDo9TO6tZExnqcxsmTyW27d2JD60B9VFblavyEWvLmBwZhRKuZyFu0sY3fn4PmSabC5+3xXcyPTXncVcOTBVGJme5RTX2Kmyuk6bSA+AtEhDvQ+JQCAQCASCVogeXq+0epmRkcH69euJjvZPXRAcf8KP5BTbXR4SwoKvghq1Smwu6f9MpZTh9foorXXg8njRqOSoFTIMaiW1Dnej4/aVmNlXYiYmRMPuolpWZ1fQMyWcGpuLbfkmNuVW0z89EnySgWIwQjRK7C4P5WYn/9tSyAXdE/h6Q0P6U1OD1IAh7McqSVknRLhskpFpcyhUwFEmj/JjXPpHp7YcLXi4rFL1Fp9PMlvVhTffj+CsQSaTYdQqqba6Au6P1LfQD0DRvIeFT37sh3C5XE6EXo3D5cWnOSIOet1Sepd/j83PH01ocHNSTUjgfdU5mC0W/txdWr8pwqCuNxqOcju4vX8Iz62qrq86U2N31/sHTe2fgqIFaS0elwt3bTF43fiUOrTh8cc8pg6VQt7sPcyoUaI8ygiyyuLE4nQjl8mIMqobVYIRnLnUXZ+ng4lpHelRen7cWojX6ztjSy4LBAKBQNAaWu3pcejQISF4nEQGtYtCLoN1hyoZ3j4maLtJvZP4dUcRKRE6wrQqPlp9mAmvrWD4c4uZ/OYqVh2s4KWpvYIePyQrmg05lSjkMjrEhfDVhjwu7pUEwG87imkXY2T6oLSgx1/cO6m+WktimJZKa8PKdYReRXxLwtajO0nVIAKRPgzyN0q/Wyuk1etgbdOGSft1UZA2VNpWWwSR7QK3D4mXQvabUpUL8++DV3vDy93h62ugeAe4Az/kCs4uooxqrhrYzJw4Mn+Ohc8QLUUvBSIiHZ8+SEWiJkztn0KZ2UGlMVOqZBToXBkjJC+OdqOCj6fXVbD7x8A7M0ZC3jr/7XHd2FDa+GHrjkHhsGEOvDsC1Ws9OWf5ND7ud5gnzkv0O/yqganH9PJwVhfiXvIcmneHonm9F9rPLsa59w/s5qpmj6sjVKdi1vAg9wBg1rAMjBolDpeHbfnVXP/xeoY9u5hRzy/hifm7KKwOIgQJzih2FpoI1Srr08hOB9KiDFicHnIrg0QzCgQCgUBwltFq0ePVV18N+PPaa6/x3nvvsXjx4oBGp4K2ITZUwwuX98Tp8bJ8fxk3j8j0a9M1MZQhmdFszKnig5n9eXdZNo/9tIsys+S/kV9l444vt3C4wsI1g/09Km4bncXiPaW4vT5enNKTgmorh8otqBRyRneKZWdhDVmxRvKrbFwUwOR0bOdY1Ao52eUWDGoFj13clfeWHQJAo5Tz7tX9iG2J6OEwwwXP+0dlhMTD+c/Cxg8btq15Cya86t/WGAcTXwFdBOgjpDbGWFj1ulQhQ9MkmkSlh6mf+j90mvJh7gWw9TMp/QWkChTvj4YqUbFIIEUOTB+USrcAFRMePL8TiQGMewOhDk/EdfknoG6ysqwJwTV5LpoIf5EgEMkROu4c055nVpgov+Bd/xStkARs5z6LvTJf8voYfq9fH574Hnj734DXVOS3j/Oekso9N+1XF0HB6Nd4cWVD+t2cKzsRsuF1+PleMB+pBlWdS+SC25joXchFXRuEnFtHZpF6DBNRu6kU+Y+3o1n5nGS8ClC2B/UXlyE/vKLZY4+mT2oEF3b3jw65pFcSPZLDAdhfambym6vYlFMNgMPt5ZM1uVzzwTqKW2hOKzh92VEgmZgeT0Pdtib9SFTKriLh6yEQCAQCAYDM5/P5jt2sgYyMDMrKyrBarURESF4M1dXV6PV6jEYjpaWltGvXjsWLF5OSknK8xt2m1NTUEBYWhslkIjT01C/xZnW4KTLZ+WV7EQnhWrJijPy+q4Rqq5ORHWPRqhSU1NgZ1C4Kj9fH6BeW4A3wv2xQK/j5juHkVlr5bUcx4XoV47rGs7+klpIaO+O7JZAQrsXh8lJQbeOnrYX0T48gKULHS3/s48/dpdw1tgPt44ys2F+Ox+tjWPtoEkK1LNtfRmyIlsGZURwsM/Pn7lI6J4QyumMMCWE6VMpj6G3WSvjscilCY8ANkLvmSBnaHhCahFcXiV0ZhmLnN6hdNci6ToKYjuC0wp75ksdHxghIGQDhTa7D6jyp+kv5AegwDoq2QuEmqcRt+7EQmgyKJuLJli/gfzcFHmu3yyQxRXP6hD+fKpxuc68llNTY2VdSy6/bi4k0qJjQM4mEMG2ryl263S681Xl49y9EWbIVd0If5JmjUUakIFe0PK3CZHNSWG1nT14pw+OchOX8jqIqG2fqcErDe7KuSs/5aQoMllypNG5YCr79f4CtCmeHi3BHd2ZZoYwsvZUkRza67F+RGaKRdbsUQpNApZUEwX0L8JXsojauP57Uoayv1PPH7lIMGiVDs6LpZawiZu4QyR+nKSo9BVct4ePdXi7ulURimPaYPhrOgm2o3xseeGdYCvYZC9BGtiyypsLsILfSyg9bCpHLYGKvRFIi9UQZNJhsLm7+dGN9Ge+mfDCz/2ljcNmUM3HuHQ8GPb2Q/mkRTGsmiutU5JbPNjJ9UBr3jut4sociEAgEAsFJp9WixxdffMG7777L+++/T2amFGVw4MABbrzxRmbPns3QoUO54ooriI+P59tvvz0ug25rzuQvf6sOlDPt/bVB9/98+zC6JoUF3R+IYpONcS8to8Yu+YEY1Ar6pUciAzblVjE4M5o3r+qNQt7qQKIGqnOlFBKQSm0m9QN9FFTsh4qDWPrdws0ll2DUKHlyUnfJL+B44XbCN9fA3l8D79dHwk0rIbRlK/CCBs7kuXe2UWlxcPWcdewsrCEmREP3pDDsLg9b86r58zI1CfMmBT949hJI7N3ic9nWfIDut7uD7nfctBZNfKdWjD4wRdU2hj67KKBoDHBF/xSeubTH3z7PyUDMvWNTYXbQ94k/uWN0ewZntiyt7FThmV93Ex2iYc6M/id7KAKBQCAQnHRabGRax7/+9S++++67esEDICsri+eff55LL72U7OxsnnvuOS699NI2HahAwmR1YrK7kCEjXK8iROu/emx3uik1O7C7vEQZNUzpl8KPWwvokRTGraOyiDCokctk5FRY0GsCrxp7vV7yqmw43F6MaiU+wOXxIpeDUiYnVKeqFz0sTg9L95XVHxtjlPr/W8gUUni/0yKZH+avb7TbZ4jlzsERhMrsGBzFoE30j85oCdYqsFdLwooussE8tbYEXBaQqySxxdiMQaI2XFR3ERxXSmvsWF0eVHI5MSEa1B6LFA3l80pVjPSRfsdUW53U2FyEyJ2E+mqQ+TzY5AYqCUGtkBMXqqW81oHN5SZMZsXoNUumh7qI+spIPp+PkhoHdrcHGaBSyPD6INKgRiaDCrMTj9eHQaPE4/HVl7Auq3WwMaeKa/tG8dDgRKJC/EtbNyKI50gwZIZmHkBl8iPGxX8fmUzy/ghmThsbcvxK6gpOPjsLTz8T0zrSogysOxy8wptAIBAIBGcTrX5KLCoqwu12+213u90UFx8xrkxMpLa29u+PTlCP2+Nlf6mZR3/cydpDlchkMKJ9DP++qAvtYhryjQurbOwqMvH87/vYU1yLSiHjvK7xfHfzELbnm/jHd9sprrETolEydUAKHeNDsDnd6NQNl0KxycbivWW88ud+bhmVSYXZySdrcqi0OIk0qJk5JJ3/XtaDK98LHEFy5YBjmxAeE0MM9L0WVr/uv08mQ99pLH2/uFgqR6sNh8G3QJ+ZksdAS/C4oWwP/HI/5K6Snm6yxsK4J8FSAfPvgPL9UhWXHlfAoJthx3fgCJAjPfg2yTtEIGhjamwu1mRX8MTPu8mttJIZY+CbKfFErHwc2d5fJNEjqZ/kfRPXFZRqXG4v+0preWXhfh4apMOw4VkU++eD14MhoRfm4U+wuDqO2Igw5m3M5Z/95RhWPow8d4U0DzLHwHlPYTJk8MfuUp5fsLf+njGlfwo9U8JYtq+MaQPSuOWzTRTX2OkQZ+TFy3tK94YFe8mMMTB3QgQJax5HufFPmPiG5JNTG8AbJKaTJCy2AnlCD0kocft7arjbjwd925htRxs1zByczssL9wfcf1HPIIazgjOC7QUm9GoFsaHNV1Q6FUmNlCq4VFudouyyQCAQCM56Wp1/MGrUKG688UY2b95cv23z5s3cfPPNjB49GoDt27eTkZHRdqMUkFdlZfKbq1h7SFq58flgyb4yJr+1ivwqqYpAWa2dnUUmbvhkI3uKJdHJ5fFRa3ezaE8pD87bXm+8V+tw8/7yQzz329764wEcLg8/bSvioXnbGd4+mn3F0sNTpUWqvlJpcfLiH/v4Y1cJ9wfIFf7HeR1JjmyZYWOzKNUw8EbJw+NoZDIY9wTyqsMQfiTH2l4Ni5+CPx8Fm6ll/VcdhjljJcEDpA90/x/wwXngMkuCB4DbAZs+gnmzYeon/v1kjoVOF7b+/QkELWDlgXJmf7KxvgrDqxdEE/nVRGR75kuCB0DBBvjgXKg8CEDukXvFLb21ZMyfgnrvDw1eGkVbiPv2Yi5OsXL/t9t4YJCWxO8moMg9Yv7p80mmpnPGQXUO93+7tdE9Y86KQ/y8rYjYEC33frOVB8dLKST7SsxMeGMl53eNZ2BGJO9MiCHl+4tRZv8h9WnKg0ve8i+Nq4/EN+FVnMpjlJxugiw0Hudln/gbF0dk4Bv3FBpD61L2gqFUyLlyYCr90iL89j1xSbdmy4YLTn/qTEz/duTiSSAtSjID3l0kFqAEAoFAIGh1pMecOXO4+uqr6du3LyqVFELsdrsZM2YMc+bMAcBoNPLCCy+07UjPYhxuDx8sP4TN5W8CaLK5+GFLATePzKK01sFbS7Jp6tJyWd9k/jlve8C+f99Vwowh6cSEaAjXqyk02XntyKrmed3iuemTjQGP+3h1Dt/dPIS0aD3b802E6VSM6xpHbEjrDBubxWGG/rMks8S8daALh5SBsPVLWPESjHsCclc3tN/6uVSBQneMBx6XHVa9Bq4AJSdtVZCzWiptm7OyYXvxNinVZvo8qWqLywadJ0qr1MbgpYMFgr9KSY2dJ37eXf+6a2IoCWWrGqqfHI3HBUuexTXhNd5Zmk1yhJ7kms2SyWhTvB7US57g8fOfIHrne9J13RR7NfLtXzE8cwzLDjQOkV+ws4QPZ6by9tKD2FweEsO0FJrs+Hzwj++28dZVvYjY/JY0l+pI7C2JkpPfhYqDkugY3QFCE5H9fA9c/iloWy6Uq9Q6XBnDcd28Ds/+RchMufjShyKL64omsm0NtONCtbw1vS+Hyy0s3ltKuF7FmM5xxIVqMWr+Qkqd4LRhR4GJ7q30vDpVSAjToVLI2F1Uc9r5kQgEAoFA0Na0+htbfHw8f/zxB3v27GHfvn34fD46depEx44Nq/6jRo1q00Ge7dTYXCw/UB50/+K9ZcwYko7V4WFTbpXffrlMRq3DPyWpjsPlFjJjDPXnqvPqcLq9uIM4+Lm9PvKrbPxz3nZuG53F7HP8S+f+bYq3w093SKHvsV2kh7PlL1Cv6qgClLWszoHorOb7tZsk4SIYeWsgvntj0QMgZ5W0qu7zgkINSj1kBKkgIRD8TcwONwXVDcJczwQdkXl/BD8gZzkem4mVByronRrWbFtF3ip6Dwf92mVB24TkLaZ/wvksO+C/r8hkI1SnYnNuNe3jQig0SdEgG3KqUHusKA/+ftTJVFJaWNEW+OJKiG4vpbocWChVjAF81kqgddGBKo0OYjJRxRyHe08TYkI0xIRo6J/h750iODMxWV3kVdmY2KtlVYBONRRyGSkRevYUi7K1AoFAIBD85WWqTp060anT33fHFxwbtUJOTIiWHsnhjOsahwwZdpeHbzbmsSa7khijGpVCjkIO4frGpnu9UsJJimg+BDtMp5LSRgCtqiHjSaVoPvspLlTD05N70C5GT36lFYfHi0YpGS1qlP7GnuVmBzU2Fwq5jAi9+tgRIYYjefnWCji8vPE+mUwKbe89HdqPk4QQR630MHUsFCrJ+PHIA5cf+iipr0DbbdWSFwhA+3OPfS6BIBBeL5iLJCFPoQVjDNUuBSabC4/XR7hOjU6lQCmX1QuPVXYfTmM8QbPz9VHI5EoiDWrMdg+O2ASCJprpo3CjlOZY+b6ATTy6GCqD+I+GaFXYXR4iDSoOlJobtmuUkiBoOMrjxutuLFCW729IHzuCTCXSRASnFjsKpVTJ09HEtI6USD27CoXoIRAIBAJBq0UPj8fD3LlzWbhwIaWlpXi93kb7Fy1a1GaDE0iE6dU8cUlX5qw4zL1fb8Xh9hJlUHP9sAyGt49hQEYkWpWC+DAtl/dN4b3l2QDcMLwdSRE6Vh+sYGBGZL0fyNFEGtQkhGuJMUpGbeF6NYPaRbImu5LCahvtog1kl/uHv7eLNnCg1EyV1cmhcgsfrDxEpcWJRinnigEp3DIyi7hQ6UHG7vKwLb+a//t+B/uPPCANbx/NYxO70i7GGPyNR3cATWhg89CssZLA4bbDd9dL4f0hCTDmEQhNkCpQBEMfCcPuhq+mB97fdTLMb1IOU66UjCLrBA+ZDDpeFPwcAkEwrJWw9xdY+BiYS0GhxtvzSiq63sZFc7OxuTx0SQjlyUnduHNse174XRIlluyroPTKq0je+nHgfofciSY8nptG+Lj7qy2UD5lCyoa3A7ftfQ0RlkNY+92KPmdVwCbVvWbzwzz/yLEogxqP14fD7WVgRhTvLsuu3zdjSDp6g1EyFt79g7TR55OEy7AUydujKfHd8eraxnhUIGgrtheY0KrkJISevoJcaqSeNdkVuD1elMdYxBAIBAKB4Eym1X8F77zzTu688048Hg/dunWjZ8+ejX4EbU+lxcETP+/m6w15ONySyFRhcfLcgr3IZdAhVhIOIgwaLuubxMCMSLonhZEcoePRH3fy9tJsbh2VRXJE43Vfo0bJS1N6khapRyGXIj3iQrU8eUl3UiJ1vLssm4cndCHK0HhtOcqg5uEJXdhwuAqby8Pzv++tNzp1uL18tCqHB7/bXr/tQKmZK99bWy94ACzfX86Ud1ZTUGUN/sZDEuCqb/zTWKKyJHFjwf/B9m8lwQOkyhD/uwn2/4mfsUlTUgZB72v8t5/zAJjLJHPUOhQqmPAyrHtXei1XwOQ5krgiELQGrwd2/Qg/3CoJHgAeJ/JNH5G2+E6eHCeVRt5VVMOUd1Zzbuc4eiZLngI2l4cv9ssxnfOYf79dJ9dHHg1sF8nFvRKZu9NN1ejn6qO46skaAxGpGL+5DIUhEnuva/26sw+6C19MZ/TqxhFbRo2SZy7tzttLDvLIhK58tymfugy4AemRXD04TYoQi+4A5/yj4cAVL8GFL/iLkSHxuCZ9gDZcVD8SnFpsrzMxlZ9+JqZ1pEbqcbi9HK5o5u+sQCAQCARnATKf71hPh42Jjo7m448/5oILLjheYzrh1NTUEBYWhslkIjQ09GQPx4+9xTWc9/LygPtCtUp+vWs4SeGSMGBzeSitsWN1erhu7nqKjuTax4Zo+OcFnXF7vewrMdMlIYRuiWFEGNREGf3L8eVWWNhfaianwsKAjCgKq23sK6mlfVwIchk89cse7j23Aw/N2x7UL2TBXeeQGK7lji82s3hvWcA2T0/qzpUDU4O/eY8bagogfwNUH4aoDuDzgCEW5o4PfExoIsxaKP3bHNZKqC2GgwtBroKs0VJYvscOlYckD4+QOMk81e2A/b+DLlLy8TDGgzqAp4igVZzqc6/NMRXAO8OlyIcA5E9ZwJjPq+rFzYt7JfJ/F3TiULmVjTlVJITrGJmmJdxbhSx7Mbis0G60dK0bGswKqyxOimvsVFZWMjjKinzvfKltUl/JSHTxE5JYqI+m5po/wGlGn7sUmxuqEoZh18Rw4zf7uXNsB9xeL7mVNjKiDaRG6thfaqZvagQalZzVByuoMDsZnBlFSqSe6KPvJTYTmEvwHlyEz+3EmTkOpUqNp2iHFDGV0AN5XFfUbWw8KmgZZ93cayXnPLeYromhXDM4/WQP5S9TY3Nx46cbeX1aby7qcYy/hwKBQCAQnMG0Or1FrVaTlXUMo0hBm3J0znxTauxuau0NooNOpSAtykB2mble8AAorXVw11dbiDaqSQrXkRVjJCsueJnI1CgDqVENuczdksIY1zWe3Eor5zy3GAC5vHmD1EPlFkJ1SjYGMFetY/HeUi7vn4xSHiToSKGEiDTpx1YNX06TqkJ0nhC0T2oKA1ekaIo+UvqJ69JkRxgY4yB1UOPNcV2P3adA0ByO2qCCB4C6Yjdxoen1JWo3HK4CmYyB7aIY2O7oCgxRzRr2RhjURBjUkBCKY8NnaHZ+D0oNrH1HEj/qsJajcNuY/F01D5w/hTcWH2Db70W8ckU8hyqsje4ZC3Z40ajk2JweosdrGNs5jtTIZvwOdGGgC0Me00F6eWSzKrodMPEYH5RAcPIwWV3kVlq5qMfpHc0XqlMRqVext7iWi3ocu71AIBAIBGcqrRY97r33Xl555RVef/11ZKdh7frTkegAkRhdEkKZPiiNcL0KOTKqrU7C9Q1pKGqlHI1SXr9iXEe52Um52clNI6T/+hqbi1q7qz5H3+XxYtQoMaiVVNuktJEwvap+DAoZXNQ9nvO7J5AQpuWt6X34Y1cJP24p9Kv0EqZT4nR5iTFqqLEFFkeSI3SBBQ+XDcyl+FxWXAo9FYSjkXmJUBuRJfaWwueDoVCD4qjPzOOSIjqcFlDppBKzgSq/CATHG6VWSo/KGAm9rpSuVbkCDiyCrZ/jNcRRa28wIo42qlHKZRRW2zA73GiVciKNaoyaxibAbqcdb00xPqcFVHpqlZFYvEqiDWo0CV1h9P9JqTW2Klj/vlQZCUCuwKNQs7ekFo/Xx5a8arw+6f5RR90942jC26os9d/E7HBRaXZid0v3rbgQDYom3gXVVidVVhdur5dQrareaygQpbV2qq0uXG4vRq2SlAgd8mCCrOCMpc7EtFnPqdOE5Eg9u4uEmalAIBAIzm5aLXqsWLGCxYsX8+uvv9K1a1dUqsZffufNm9dmgxNIpEbqiTSo6z0yJvZMZHj7aF5ZuJ/8KqmkZb/0CJ6Z3IOsI/4e0UYNl/ZJ4vN1/saBWpWcbslh5FVaya+y4nB7WbqvjK/W52F1elAr5FzaN4mhWdHc89VWUqP0PHtpD7onhaJUyEmM0PPAt9vq207slcgrV/Tm7q+24PR4j5xfTXGNgyd/2c30QWk89tOugO9tav8Aoe21xbDsedj8MTK3A7VKT2jvG9iadAX9x/4H1coXJPFCbQgc0dFjKhhipN/NZbBhDqx6DZxmyZ+j5zQY+eCx018EgrbGEC35wZTvgZ/vk7xj5AroeCFM+YQSZyJV1sP1zW8emcmyfeU8+tNOqq0u5DIY1yWef1/UhaQjHj2O6iJkq99Avel9SSxUapH1vAbFwLuwF1eiX/gA5K2VOgxPla79g4tg+7e4O17EqkIZyRE6KizOen+OTTlVDMuKYsUB/6iUmBAN1TYXh8stpJ/EyhYF1Tae/HkXv+0oxuuTqlDdc24HJvRMJPKID9HBMjP/nLeNtYekaLPEMC2PX9KNQe2iMGga//nbX1LLwz/sZHW29J7jQqWUwKGZUUSHnL5mloLWs73AhE6lICHs9P9/T43Uszm3+mQPQyAQCASCk0qrl7DCw8OZNGkSI0aMIDo6mrCwsEY/grYnLlTLR9cNIFSnJMaoYXz3eO7/dlu94AFSGPyUd1aTf8QYVKOUc/XgdHqlhDfqS6dS8PLUXsiBRXtK2ZBTxe87S/hw5WGsTg8ATo+XL9bl8cOWAmYNz5CMSN9dQ06FlQ9WHOLdZdmN2n67MZ95m/OZNTwDkB4+Xr2yN28uPsCOgho0SoVfmLBCLuO5y3qQFNEk4sJWDb89COvfk3w0AFxWDOteYUDeByj2/w7bvoYVL8LFb0rCx9Ek9YNRD4FaBy47rHkTljwtCR4gRX1s+gh+ukvy9BAITiRKnWS4u+SZBrNcrwd2/4hvydPUHBVQMbV/CjFGDXd9taW+DLXXB7/tLOb6j9ZTWmvHYalBtuRp1GtfkwQPALcd9cZ3CbPlEvX1xcjqBA+A6lzJRLXLJXgzRmIf9R+eX5rHE5d0480lBwDJ9zTSqObJSd3JjGk8v8L1Kp69tAdP/7Kbqe+uprDaxsmgrNbODR9v4JftxfVCjcnm4pEfd/Lr9iK8Xh8FVTamvrO6XvAAKDTZuf6jDX4r34crLEyfs7Ze8AAoqXFw55db2F4gVsnPNrblV5MerUd+BkSzpkToKTgSKSYQCAQCwdlKqyM9Pvzww+MxDkEzyOUyuiaE8usd51BpcfD4/MBRE5UWJysOlHNF/1RKauzc/OlGZg7NYPY57dhfUkuUUUNsiIa3l2Zzz9j2JEfo8Pq0vLpwf8D+/thVyrQBacBBnB4v7y3PxuMJ7Hu7cHcpt47KIjVKT6hWBT5ffbWWf/1vO7eMymL+7cPYmleNQaOkd2o4MSEa9Ooml6ClHHZ+H/Acys1z4fKPpBf5G2D16zD5XagtkY5rNxIiM8AYK7Uxl0iiRyD2LwBLmeTpIRCcKGqLYNlzAXfJCjbSa4yFB8/vRIf4EHQqOTd/uilg2z3FteRWWumhr0K97VP/BvHdkeeuDuwf4vPhW/M2rovfZme1ng9nDsSHj1tHZaGQyeiXHkm0UU2IVsVH1w1gW76JfSW1JITpMGgUPPPrbg6WSRFWW/KqSQzX+Z/jOJNfZWNXYWAx4oU/9jGuSxyrD5b7peXU8fSve5gzo199SuCmnCpKahwB2z7/+146xYeQcBLep+DksDXP5LdgcLqSEiktLOwtrqVvWjOl3AUCgUAgOINptegB4Ha7WbJkCQcPHmTatGmEhIRQWFhIaGgoRuPpnwN7KiKXy0iK0CGX0+zK4/J95Uztl4LN5eFwhZVHf9yJXq0gKVxHrd1NcY1kbrq/zIxaKSdCr8YVRMgAqLa56r1BNuVWM6Fn8JSQvEorry88QH61jTem9anf7vXB64sOkBqp46pBac2/UUtp8H0eV0N5WoD89fDlVVJpW20YdL6oQfAAsJvAbffvpw5THsR0bH48AkFb4jRLvhrBdhftZMWBzvz39728dmVvSs2BH8QBdhXW0COpSooUaUp0ByjaEvRYWdFmgEbmqIFMSeUyGQ9+t424UC1VVn9vj9UHy7mg+4k3e9xTVBt0X6XFidvnY9n+wBWjAHYUmLA5PRwpesW6Q8GjvnYW1uDweIPuF5xZVJgdFFTbmNwn6WQPpU1ICtchl8Ge4hoheggEAoHgrKXVokdOTg7nn38+ubm5OBwOzj33XEJCQnjuueew2+28/fbbLe5r2bJl/Pe//2Xjxo0UFRXx/fffc8kll9TvLykp4R//+Ae///471dXVnHPOObz22mu0b9++vo3D4eC+++7jiy++wGazMWbMGN58802Sk5Nb+9ZOKi6Ph5IaB7V2N1qVgkiDmrAmZoG1dhcOl4f4MC2HygNXJ0mLkkJZFTIZmTEGDpZZsDo9HCgzM6FHIhf2SCA2REOYVoXN5UEmgzev6oNKIae81kGYXjqnSiFnc24V4TplvU9HepSejCg9713TF5VCToxRg8XpxmRzszm3ilCtkgKTFO6uUflnTjVnIFiP5hilE5X+pq7UFoG5WDKJPJpjlZTVHVUNw+uV+rFXSeVr9VGS/4JA0JaotCBXgjdwqLlTn8COwhrkMog1anjvmr5EauVk6mpxWaoxuZT8cdjNW2vLpQiLYIa81gqI6dTwWm2EvjMhfagkHCo0kolqEGrtLirMTswONy9M6YVS5mFAlBOF0wQKDVU+IysLfWTEGNmWXy0ZhIaoKbe4MDvc6FQKoo5EiwR8n24PpbUN97sog5rQIOaoVock1tbYXOjUCsJ1KhLCg99L1Ao5CpmMjOjgAnxcqBaFoiF1oW41PBAxIZrW54EKTlu2FRwxMW3m+jmdUCvlJIbr2FscXCgUCAQCgeBMp9Wix5133km/fv3YunUrUVEND42TJk1i1qxZrerLYrHQs2dPrr32Wi699NJG+3w+H5dccgkqlYoffviB0NBQXnzxRcaOHcuuXbswGKRVybvuuouffvqJL7/8kqioKO69914uuugiNm7ciEKhaO3bOylUWhx8vSGf1xbux3LEK2N4+2iemtS9/st4YbWNR3/cidXp5obhGfzz+x1+/chl0D89khH/XUK7aAOPX9yNZ3/bw7YCE09P7s6eolp+3V7EsPbRuD0+CqptzFlxCKvTw/3ndcTj9fHkL7vrc38Ht4viwh4JRBs02FweHhzfiWd/28PC3aV4fZJvyBUDUukUH8KW3GqprVFDRrSBbfnVjcYWplORFRu8RG49hhhplbp8n/++lEFQcTDwcZljQd9EpNBHQ7vRkL3Iv314KoTES7/ba+DgQvjlfinlBSChJ0x6G2I6SyYHAkFbYIjB1+0yZNu+9N+nj6RQkYTVkc+rV/bms3W5DEmAga4/CV37EjjNxAAZ6SM5/6qn0cSE4JXJILEPFDZJgzm8HIbeDevelaKgJr8Ha9+CNW+AzwdqA7KBd2DtPRN9ZHyjQwurbTz2005+31WCzwcLbupBevkyND/9q94HJyGpDxMnvMU/lpbz/eZCwvUq7h7bgWKTnbeWHkQmg7GdY3lsYje/9JcKs4Mv1+fyxuKD9d5AIzrE8OSkbiQ38fjJr7Tyw9ZC3lpysNF96clJ3QjVKQNWhZrUJ4nSGjv90iOQy8AbIJDtllGZxB5lTjquSxwv/bHPrwIVwHVD00kME6ktZwvb8kwYNAriQgMI7KcpKRGigotAIBAIzm5kPp8veG5DAKKjo1m5ciUdO3YkJCSErVu30q5dOw4fPkyXLl2wWq1/bSAyWaNIj3379tGxY0d27NhB165dAfB4PMTGxvLss88ya9YsTCYTMTExfPLJJ0ydOhWAwsJCUlJS+OWXXzjvvPNadO6amhrCwsIwmUyEhh4j0qCN8Xh9fLY2h4d/2Om3r120gS9mD0IhlzHro/VsyTPx0tReeLxeVh6o4PvNBfVtNUo5j0zowvL95fy6oxgAvVrBhzP789naXML1Kn7YUsjzl/fk87W5dE4I4c0lkoDQPz2C0Z1iefa3vX5jSIvSc8fo9kQZ1by2aD8bc6r92swckk652cG2fBMPje9ElFHNDR9vxHSk5G2EXsXH1w+ka0IocnkLBITyA/DpJMl0sY7oDnDpHCk94NvrpKiMOuK7w0WvQFxXaSX9aKrz4IsroOQokSg0Ea7+X0Nqy6Hl8NFF/uPQRcCNyySBRHBcOJlz72Tg8nipLs4h4pcbURYcZTCqj6L44i+Z8YuNkR1jyS63YLE7ebPjFsIXP+TfUVQW3hnzqVFEorXko/1qClQcaNgfkY7zqv/hLdyK1lkJ6+c0ngNHcIz4N77Bt6HVSvOm3Oyov9cATO2XzH+6FKH5eqr/GAwx2K5dSOfnG/p9enJ35qw4xIEjfj7dk0L5cOYAokOkB0iPx8uHqw7zxM+7/bprH2vks1kDiT0SEeZyefhsXS6PBqj8NKpjDHeMac+1c9fXm7wCDMyI5NlLe3D+y8sY0TGGUZ1ieeSHnY1Kd0/tn8L953VsVArcbHexOruC27/YjN3V0PbC7gn834WdT4pvyfHmbJt7LeW6uespNzt4aHznkz2UNuP7zQX8uqOIbY+MQyZEfIFAIBCchbQ60sPr9eLx+OeQ5+fnExLSgpX8FuJwSLnsdV/GARQKBWq1mhUrVjBr1iw2btyIy+Vi3Lhx9W0SExPp1q0bq1atarHocTIprbHz8p+BjUSzyy3kVFgI0SjZkmciyqDG4/Vy/7fbuH5YBh/M7E9RtQ2NSk6YTsUnq3NYtr+8/nir08O6Q5XcOaY9F7y6nCsHpPLZmhwm90ni/46KFLm8Xwr/DSB4AORUWIkyqtGpFAEFD4Av1+fy4pRezN9WhFGrpH2skc9vGMi+klriQrSkRxuID9W2TPAAiM6C6xZAVQ6+qkPIVHqp0oWlDH65D8Y+CjK5ZFQangrmUvjySrj+D4ho4hkSngJXzwNTviSmhKdIbUKP5GtbK+HPhwOPw1YllfbsO7Nl4xYIjkFprZ3Jcw/y4Ijn6D/UjrpqHx5DPBZDKuXyaG4f4yQt0sCE11fw2ZRkwv98PnBHFQeQVx2iQmfkts8LeP/Sb4lwlUBlNrKIdKo18dz0ZQFvTh1Bgm0f8gCCB4BmzSvYu10G2nQASmrs9YIHwL1DI9F8f2PgMVjKIHctl/ftyDcb8wF4b1k2Vw1K5fH5kqixvaCGIpOtXvQoqXXw2qIDAbvbX2omr8pWL3rkm2y8sThwZNfivWXcNDKTn+8YzuFyC6W1DjrEGYk2qHn2t73Y3V4W7CzB4vDw+rTelNU6sTrd9E4Np120kQhD49Qeo1bFsKxofrljOPtLzFTbnHRLCiPaoCZORHmcNfh8PrbkVXNO+5iTPZQ2JSWywdMrQVzPAoFAIDgLabXoce655/Lyyy/z7rvvAlKEhtls5pFHHuGCCy5os4F16tSJtLQ0HnroId555x0MBgMvvvgixcXFFBVJq/zFxcWo1WoiIhqbc8XFxVFcXBy0b4fDUS+qgLTidbKwuTxUWgJXGADYVVRDu2gplSc5Qsf+EjM+H7y//BAfrDjEc5f24L3lh4Lm664/XMnozrE43F7SovT8sr2Iy+Up1B5Vvs6oUVLWjGHihsNVdIoPLmjZXV7q4oV2F9UwvH0MEQYNXRP/Rgnj0EQITUSmMcKccZKXx5hHoDIbvr9R8v7QhUsPXnWlOh1B/h+NcdJPUl//fW4blPhH2dRzaLkQPdqQU2nunQysDg8ltQ7unl+ARiknJiSZGruLGttB4CCvT+uF64iHTpTaE7j6Sh3F27End2V3cS1D39hFtFFNelQ6ORUWysyS6FBqg8iSfQR9zHHUgKvBHyi3onGknlHphVL/qIw6lPmrGZQ5pF70yC63EB/a+GyHyi10Tw6X3r/TUx8BFoh9JQ0VJhwub7P3pfWHKhmYEUXSUVEYZbUONuQ0GMWuOFDOigPlRBvVaJQKft+pZc7M/gH706mVtIsx0i7mzPByaMrZPvdaQqHJTqXFSWasv6nv6UzqkbSxPcW1QvQQCAQCwVlJq0WPl156iVGjRtGlSxfsdjvTpk1j//79REdH88UXX7TZwFQqFd999x3XX389kZGRKBQKxo4dy/jx4495rM/nazaE8+mnn+axxx5rs7H+HdRKOTqVApsrQAUGoHN8KJEGNe9c3Re5DKKNGtYfrmRPcS1XD0ojI8bA/eM64gOW7C3lq/V56FQKrhmSTu/UcNQKOWqFHLkMKixOycldLpn91RmUerw+9GpFfX59U+LDtOjUwf1R5DJQK6XPOz2qjb8sKtTgsoLXJZmL1uGokUrU9pgiVa9QqCSzxtYiV0JYcnCvkNgzJ8T5VOBUmnsng1i1g0Uzk1G4LTgURr7d4+CD9Q1VRiL1Ggwa6bZs8ypBpWsQ9ZrgCU9HrWyw2Cw3+1dX0SjlyMJTgg9IoZbOcYSmZsNOrwx9SELjdLKjxxDVkewyc/3raKOaWockanSMC+HaYemkROrZkltFuF6NQaOorwYViOSjBAy1Ut7sfSktwL1Gq5KTFKEjt7KxeFP3uQzOjEKjPD28ntqas33utYStedUAZJ5hwld0iAadSsHe4lpGdYw99gECgUAgEJxhtNqUPjExkS1btnD//fdz44030rt3b5555hk2b95MbGzb/jHt27cvW7Zsobq6mqKiIn777TcqKirIyMgAID4+HqfTSVVV4xKQpaWlxMXFBe33oYcewmQy1f/k5eW16bhbQ2yIlmkDA3tGjO0ciw+Y+eF6bvxkIzd8vJFZH21g2oBUPr5uAHuLa7n0rdXM+ngDN326EYvDw6tX9uaNab3ZlFvFrI82cM0H6/hiXS5jO8fxv80FXDkwlYW7S7mwR0OZyfnbiri0T+BqN6FaJVFGNRUWJ8kRgVeIzu0Sx/L95YRolHRJbOPccEOMZGDqcYGtEiIyjpz0cYhsB/NugG9mwJfT4JuZjX0NWoIxDs55IPA+uRK6XPJ3Ri9owqk09044pgJCf72Vdl+NIO3b8XT4ZiT3eubw2RXpKOUyYowa0qMNRBnVdE0M5aPtNqw9rgncly6Cvd5kfD6ClqHsmhhKmF6FL6Jd41LOR+HqNhWPvmFfYriOxLAG4ePzXQ7sg+8JPAaFGln7cxuloFw1MI15mwoYlhXN7aOzePXP/Ux6cxWXvLmKy95exda8ah6+qEvA7iINajJjGx4244xqpvYPLNiEapV0DXCvCdGquGN0VuDxIpmSHi0UnU2c1XOvhWzJqybaqCZCH7yy0emIXCYjJVJUcBEIBALB2ctf+van0+m49tpref3113nzzTeZNWsWhYWFjB49uq3HB0BYWBgxMTHs37+fDRs2cPHFFwOSKKJSqfjjjz/q2xYVFbFjxw6GDBkStD+NRkNoaGijn5OFWiln9jntOLdLY5EmPkTL/ed15Jo56yiobljprbA4KTM7ePa3PSzZ17BC7PH6+N+WAhbtKWFTbjWrDzaExX++Lpcp/VOID9NSWG0jyqDm/G7xDM2SIif+2FXMgIxIxnRu/GAUbVTzwpRevLbwAF+vz+P9a/r5CR/90iK4vF8KC3eV8MqVveojPtoMfSRc+i7EdYNl/4ULX5CECK8bVr7ceBW8aAvMvQhMBUE6C0LWGBh0a+MqLWojTPtKigIRtBmn0tw7oVgqYN5sZPt+oz4XzOtBs+MLeu17lduHJ/DJrAEkhutQyWU8OrEru0vt7EyfiSurSXSbMZaii7/i7t/KmT5nDf+9rAddEhp/jlmxRl69sjdJ4Xp00ak4p30vpYwdhSdjNN4RD2IwNAgN8WFaPrpuQP0835RbjaPDRTh7zWw8P7Rh2K/4ljxPg+BySa8kEsN1bMqp4obhGdz99RYKTfb6/eVmJzd/toluSWFc0K1BdAWIDdHw2ayBJBwluBh0aq4bmsHYAPeludcOICWIuWjnhFD+eUFnlEd5CGmUcl6a0pPUto5EO404a+deK9icW3XGRXnUkRKhZ5eo4CIQCASCs5RWV28JxtatW+nTp09Ak9NgmM1mDhyQVuZ79+7Niy++yKhRo4iMjCQ1NZVvvvmGmJgYUlNT2b59O3feeSd9+/blu+++q+/j5ptvZv78+cydO5fIyEjuu+8+KioqWlWy9lRwsa+yOCk3O8ipsBJhUJEaqee95Yd4d1l2o3YyGbx/TT+u/2hDwH6UchlvTe/DDR9vbLTdoFZw19j2DGoXjcXhItKgwe7y4IN6ISTCoMbh9pJfZSNCryLCoKagykqIViWtAIfryKmwsKeoFqvTTXq0AbfHS5nZicPt5a0lB/j3RV0YfjxM4MylUoi9pUIyI517oWRkGoirvoX257auf3stWEqhfD+oDRCRDsZ4UKr+9tAFwTkV5t4JoXQPvDkw8D65AtfN61HFZAJwsMzM1HdWc8eY9sSHaUnROkjTmPFVZlMrC0Udlcrln+dwsEzy4vi/8Z24sGcCpTUO8qtsJEboiDFq6std12Etz0VhLsJrLkMRmY5TF4MxInBEXLHJTrHJhlGj5KZPNvKvc5MZEOuBiv3INCHYDck8s7KGPulRxIfpiA/VEqJRUOtwY3a4+XFLIR+tzgnY97gucTx+cTdMdhe5FVYiDWoSwrVBvQYKq61UW11kl1uktqFaUiN1zd7frU435WYn2WVmFHIZ6VEGYkI0aFVnZ2pLIM6auddC3B4v3R5dwKV9krmoR+KxDzjNWLCzmE/X5LD78fNRKc7OaCeBQCAQnL202tOjLdmwYQOjRo2qf33PPVIY9YwZM5g7dy5FRUXcc889lJSUkJCQwDXXXMO///3vRn289NJLKJVKpkyZgs1mY8yYMcydO7fFgsepQsQR0aF9nGQYanG42JZf7ddOr1I0a3zq9vpwuv11LIvTw5O/7OGbGwdzy+ebcXm8RBs1mKwukIHF4UajkvPZ9QM5v1t8/XEd4hobmHp9cMeXmzFolJjtLoxaFVanu77MY52RaZtjjG0I0a/MDi54ABRta73ooQ2RfqIy//oYBYJgWEqD7/N6ULkaws5NVhflZicP/7ATlUJGiFaFxeHCoNFidVp5cYq2XvAAWJldwTVD0kkM19MrNXCqC4A+OhWiG1Lpmgvgjw/TEh+m5XC5hQPlFmZ+sRe5HNIiDNQ6nJSb9wAgkyt45tLG6XkVtXZ2FgZfUd5bUovT46FDXIjf/SUQieF6EsOhSyuMkfVqJamRSlKbCD8CQTD2ltRid3nJOlMjPSL1uL0+DpVbWjTvBAKBQCA4kziposfIkSNpLtDkjjvu4I477mi2D61Wy2uvvcZrr73W1sNrU2rt0oOMyebCqFHWR1YEQ61UkBVjZE12ZaPtNpeHcH3w6AOZjKA560q5DKNWWS+ahOlU3H1uB+LDNDjdPtRKqfRtldVJhdmJ2eEm2qjG65MiUZRyGQaNErVCXt9HUwGmTSsfOGrBXCaVjtUYQB8DhihQaEAXIW0PREyHthuD4Kyk0uKgwuzE4pTmW7RRjVHzN6J+jjbhbYpMBpqGeWPUNtyWXR5f/RxzuKV/NU3md8f4kKArtz6fj5IaBxUWBy6Pj2ijmtgQDeoWmnkq5DKSwnUUVNvweiHKoOLZcVEkKM3IAGWoHqvdQWGNC7PDTZhORZhOSbsYQ6MqKkeTGqlHrz6pf3oEAj8251Yjl0FGzJmZAnV0BRcheggEAoHgbEN88zwBlNTYeWL+LuZvL6pP5++XHsHLU3uRHBF4JVKlkHPN4HQ+X5eL9yhdyOuD/CobHeNC2Fvib0p2buc4dgfJ272kdxJRBjX90iMwWV3884LOPPbTTg4fKVOpVsi5aWQmGdF67v5qK1cNTCUzxsiLf+zDfKTE7SW9Erl2aDqvLvI3DA3Xq+gc30Zh0rXF8Od/YNsX4DtS6SGpD1z2IYSmwJDbYeF//I/ThkNCz7YZg+CsJLfSym2fb2JbvgmQqhNd1jeZ+8Z1JLZJdZMWY4iRfGlKdvjv63C+tP8I0UY1vVPC2XykksTRDM6MYnuBqf61Qi5jSt8U5HJ/Lx23x8v2AhO3fLaJoiPeGjqVgofGd2Jir0TCW2DWmBKp56YR7fj3DzuZ3C2C/+tSRtSCq8F6RIzVRWAa9xLv7o7n622V3DY6C4Nayfnd4vlmYz6BNO2bR2QSZdQc89wCwYlkc2416VGGM7a6j1GrJNKgZm9xDfQ889J3BAKBQCBojhYndvbu3Zs+ffoE/Zk6derxHOdpi9Xh5vkFe/lpW1GjB4ANh6u46dONlJsdQY9NidTx1vS+GDUN2pTmSBnHd67uS6f4xqs1g9pFcu3QDC7oFk9mk9WqMZ1i6x/aXp7ai39f1IW7vtpSL3gAOD1eXl24n9IaByPaRzM4M4r/zN9VL3gAVJidDO8Qw8SeiY18DRPDpH7bpDCCywbLX4KtnzUIHgAFm+Czy8FaDr2vhj7XNDZXDE2EGT9BWDMlOgWCZiipsTPzg3X1ggdIQuPXG/J5e+lB7EFKSx8TYyxM/UwSPo4mbRhc8AJoG1I35DIZD13Qme5JjdM5+qSGc/fY9niP3EhCtUrmzOhHUpCqSgXVNq58b0294AFSpNjDP+5kSwBBJRhjOsdxw/AM7uuvJuqnmQ2CB4CtirAfr+WuPko6J4SSFK7j2d/28P2mAh6/uBuGo0pda5RyHpnQhfZxZ2b6gOD0ZlNuFVmxZ/a1mRqpY3eRqOAiEAgEgrOPFkd6XHLJJcdxGGcuZWYH328OXE1kR0ENpTUOooOseurUSkZ3imXBXedQXGPD4/WREKarN+T79PqBlNTaqTA7iTSo0akVhGpVxIRo+GL2IMrNTqotTmJDtUQb1fUru8kRejbmVGGyuQKed+6qwzx3WQ/eXHLQb98lfZK46ZONXNI7iQ9m9KfK6sSoUVJldfHojzt5bGJX4oIYErYYcyls+jDwvvJ9UJMPSX1h3JMw9C6oKQRNiPRgGSpWsAR/ncJqG9nlloD7Pluby7VDM/wMQltMZDpc/T1YysBaAYZYKcLD0Dj1pdzs5IaPN3DrqCzuGtueapuLcJ2KfSW1XD93A1/dOJiRHWKJC9MSF6JBGSS15bcdxfVeO03574K99EgOI9Jw7IiLxHAdd41MQff7/Y1FyDp8PqK3vslDYx/lqT8OA/DTtiLKLU7+e3lPvD4fXq+PtCgDmbGGv5cmJBAcB6qtTg6VWxh/lJ/VmUhKhJ6NuUHSQgUCgUAgOINpsejxyCOPHM9xnLFYHG7c3uC+JSU1drokBk8JUSnkJEXoAq7mRodoiA4J/NASG6IlNiR4KP6uwhoUchmeAGMrMtkxapRkl5n99ulUCqptLuasOMRHqw6jVMhwur31KTh7S2oZ0THW77hW4TSDO3gEDNW5kuihDZV+hPmooI3Iq7QG3edwe7E63UH3t4ijDXmDYLJJ3j9P/bJbsvtQyrG7vPXz1eXx0ictuGEpgNfrC5geU0d2mQWHO7AgEgiDzAklOxs2yGSArF4EUVfsIdkoVZ6pY/XBClYfrEClkCFDxvju8bxyRe8Wn1MgOFHUzZX2sWe210VKpJ6fthVRY3cRqhXio0AgEAjOHoSnx3HGoFEGFRcA4kKbX2mttDgoNzupsjqJMqiJ1EtlZUtrHbg8XhLDdfh8PkprHdhcHiINarQKBch8lJmdxIZoiDZqMByVIlNaa2dc1zi6JoYSolWx8mA5c1cerhdn4kI1WJ0eMqINlNQ0iA+T+ySRHqXnj5u6E6+oBWslNRhYUSTjiSWlVFtd9dVnsFWBpRxqC0GpA0O09G91DijU0oNfSALIA+RPq41SG0+TKjVRmTD0bghNhpxVEBIvrZZrTkBIstshVYypLQaZXDq3MR4UYgqdSTQXxSGllgX//zbZXFSYHZSbHYTqVEQbgouS1JaCrQJcdlDrpFLMKh0Y44g1NIigcpmMKwekcs6RikiRBjVhGhnm0sNo7OXgdeEzJqDQGFDYK/FaKvBoI6gilFnDMhjZMYa3lhwkp6KxmJMZYwCfj5wKC6W1DnQqBdFGNXGhWmQyf38QVAaI7SIZCPe/nvpcPZkcNnyARxuBQ6YlI9rAvpLGYqnL4wN89EwOD/rZCQQnk005VYTpVMf8e3y6U1fNaG9xLf3TI0/yaAQCgUAgOHGIJ7bjTLRRw8W9Epm3yT/FpXNCSLPRGAVVVu74cgsbj6qCMDAjkltGZXLLp5volBDKbaOyeODbbZQd8QZRKWTMGJJO75Rw0qMNjHtpKTMGp3PjEfPAA6VmZn+8oT6EXyaDC7ol8MKUntz91Ra8Ppg5JJ05y7OZPjCtvnrMg+M7UWVx4jUVkrrunygPLQJAD0xO7E3vK99k9o+ldIwLAVMBrH8PVr8OniMpNMY4uPR92PEdbJwrVbO4fC6kDAJlE0NFQyz0vgY2vN+wLbYLjH0Efr4PTHnSNpkc+s+Gc+4D43Eok1uH3QS758Mv94HryMOjNgwueQvajQT1men2fzaSGK4jPUrfyOumjiv6pxATRMQoqbHzyI87+W1Hcf22DnFG3r26H+nRTa6P6lz4/WGI7SjNj1WvNQh8hmiSLvuI2UOTeW9VPs9d1oOle8u47qP1+HwwtkM4z/W3YPxldkP1IqUG3zn3g9OCfMVLyIHopAGYRrzK3JVlPDqhK0/9spv9pQ1ixBOXdOOtpdl8tja3XpBNCNPy3jX96JIQ6m+MqtLC8Hvh8DL44VapshJIAuWYhzEnDOP77RU8Pbk7l7612u/z0arkjO38NyPABILjxMacKtrHGgMLfqc6HheU7pTSPBVqiMqCiDTA/70khetQyGXsEaKHQCAQCM4y2sJ2UtAMBo2Sf5zfyS9XuFdKOO9e3S/oSnClxcmdXzUWPADWHqrk3WXZXDssg9tGZXHTpxvrBQ+QVlXfX36IMrOTZfvKuOfcjry7/BC/7SimyGRj+vtrG3kW+Hzw8/Yi1h+u4qIeCdw8MpOEMC2L9pax7nAlD43vRJ/UcNQKOVqvlY6b/lMveNQhL9xM5qKb+H5GRxLDdZC9BFa81CB4gBQl8fkUGDBbem2tgE8nNwgYR6PWwYj7ofuUBqPSUf+EeTc2bu/zwrq3YfvX4P2LBpMtoWwP/HBLg+ABkhDy1XSoyjl+5xWccOJCtXx03QC6HpVyJpPBpN5J3DoqC63KPzLJ6nTz4h/7GgkeAPtKzMz8cB0lNQ1GojhqYeWrULINQhJh+QuNI5os5cg/ncQDg4zcMjKTnQUmftxaiM8nVWr59zAjkd9f2bhcs9uBbNETENOx3sRXXrCOzGV3cnXPEO7+egv3jJPKOGuUcp6d3J21hyv5eHVOowi0IpOdK99dQ6HJFvjDcdsk4c9xlBGi0wy/PoDPaeGLdXlsyzfx9ORujcrqxoVq+HzWIOneIBCcYni8PrbkVdP+dDMx9Xlg14/wzTWw4P9gzVvS390fboWf7oLi7X6HKBVyksJ1QSu8CQQCgUBwpiIiPU4AcaFanr20B/ef15Fqm4sQjZIoo7pZE8FKi4MNhwMbjq08UMFdY9qzOrsiaF7+x6sOc/XgtPrVnA9WHiI92kDx0Q9gR/Htxjz+v737Do+i3B44/t2+2fReSIVA6KEpRRQQBBuIXJoVe8eG/epV789rr9gLCtiwAXakN+mhQ4BAQk1CSO9b5/fHSmDJbgiQnvN5nn10Z2ZnzizzbnbPvO95/7j/QiL8jVRaHSx4+CJKKmyE+hoY1jmcW6avZ9rIIHTf/ek+4Kwt+NnzocACK153v421AvYuhO4TYessZ1Jk2w8w+Inq2/pGwBVvONeZSyEvDSoL3e93xRvQ5er6KWJqLoFlr7lfpzhg3Sdw2avVe6uIZisu2JsZt5xPXqmFMouNIJOeYB89vh7GwOeWmpm98bDbdfvzyskqqiD8+FS3Jdmw6Uu46FHnteOO3YJ211xu6HcPQ99YVrX4gnbBhOz9CRwe6oqsnwbJE2G583pVH17LwIEK/y63YrMrTJvUB4eioFKpeH3+Hre7KDHbSDlQUH0qbWuFs0eKuzloAe/17/Gv7pP5eFk6H17fi/ev7UWASYeXTkNwTcNmhGhku7KLKbfY6RDejOp5mIthyYuQvR2i+0Bsf/ANd3425KU7bzzMe8J546DXDc5ekf+ICTJJ0kMIIUSrc0Y9PaxWK0OGDGHPHvdfmIVnfl462ob60Cs2kPbhvqedNaG4wnPBRI1ahU6jJiPXc9HFA/nlhPgYsNmdSRFFgYNuuuwfV2l14FAUTHotQd562of50isukJggEwVlFg7ml6O3lXv80QM4a3goDijY73mbnFQIaX/iedYWsHs41+NFSqOSIXev532W59Vc+PRcWCsgv/osNlWOpTrvgIsWJcTHQFKEL71iA4kP8faY8ACotNr/qVvhXmbhSYlGaznYKsEvEgoyPAeQuRkVCmWWEz2YYvw1eBeken5NfrqzTs5JdLYSVCrYdqSIqYvSuH1mChabw6V3mJ9R69IzY5e7KS2tFc6ZkzzQFaQR76ciu7iSzMJKbpu5AZNeQ5c2/kT4e0nCQzRZKQcK0KpVtA1tJj09KvLhj0ed7f3826DrGOfniUrtHN4S1hH63gFJlzpvKqx809kr5B9xQSZ2Z5fgqKHAuhBCCNHSnFFPD51Ox/bt2+ULbAPwN1X/keVn1HLfxe1JCPGm3GonOdrf43S47UK9OVpcSXyw846tVqOibajn2hM+/xQ6Xbo7hyh/I1qNmoP55SgKxIWYaBfqg1nr7fxi5W7aSnAWJ1VrIaSDc0iIO5HJcHDNiefR59euGGhkd8/rfCNAW08F6PT/FHDMT3e/PiIZdGc5halotuwOhaPFlWQVVVBUYWP6zeexeFcOX645UC0vGHPyzEt6b+ej8BCEJEHmRvcHiDkftVqDn5e2KgGaXmClJLYHvmnz3b8mNKnacDGrzh9FKaVXbABHCsvZcriIgjILbQKMjOgSwYXtQ8kpMeNt0GCzK3y8fB/dov2rXp9fZuZYiQXFZqdDeDfUWVvcHtoS0oU9BQ5igrzILTWjVauqPlOEaMo27C8gIdQbvbYZjPa1lMH8Z5w9Pc6/C3xC3G+nUkPCIDAGwtbvnEXE+98DqIgNMlFusXOooJy4YKlHJYQQonU447/yN954I9OmTauPWMRJgr31DOpwojinj0HLOxN78vvWTG6fuYFrP11LTJDJ4w+LWwYmAAqLduUAcNegdsSHeBMX7P4H+nV9Y3n5z138vi2LlXvzuOydFdz0xXpunr6ee7/ayH1DEvlplwVLp7HuA47tD96hEBDjfrgKgMEP4i+EnXOdz3Ve0GV0Ld4NnEkPbw/FSgc/We0Od53RezsLpbpL9Km1zpksNDL1X2titTnYeKCAy6eu4F8fruaW6c52UmG1899RXV22TQr3JcL/pGLFvlHQ51bY/DX0vdP9AfTe0HEkgSYd9ww6MR3zmvR88hJGgdZD8ePzbofN31Q9tSVczIKDdsJ8DTgc0L9tCKOSo/huwyHevaYXeWUWbpmxnsd/2sp932ziv7/t5KFhHegZ60x6ZBVVcN83mxjx9nJGf7yBrE63uJ9tSaUmN/ku5mzL55YLEvh+wyH+1TuaEJ+WPROGaBnW78+nQ3OYqlZxOIeulWRD71s8JzxOFtndOfRz9x+w82cAYv/5DpDqrkeXEEII0UKdcdLDYrHw4Ycf0rt3b+68804efvhhl4eoGwEmPa/8qxuXdHbOeDBpQDyfrkhny+Giqm3eXLCHdyb2qOrNAeCt1zBleAd8DFr6xAXx0bJ9PH5pEoOTwgj3MzLzlvNJPulOrk6j4vp+sSSEerMmPY/hncP57287XWqFpGaXkJZTSnhoENs7P4y501iXMcJKu6HOmVm8//kSFtMXhr/gnNnhuKC2cP1PzvobAAFxMOnXqsKLp+UfDTf97ux1cZzWCEP+DR1Huk9K1JXg9jD+SzCdVO3eNxKun+08D9GqZBVXcP20tRSWnyjUqyjww4bDFFda6RkTAMD5CYF8NqkPoSfP0KQ3OYv5Jl7iLDR4yf85k4HHBcbDpN/APxqtRs3YPjHcNagteo0aRYEnFheSN/ZH13ZjDEC54g3Yv9JZMBiwJV7K7vNf5Lc9Fbw+LpmX5qXyzM/bGd2zDf5GLavTc/l5c6ZLr5T8MgsPzNqMzQ6lZisv/JbKqn15gHP428trzeSO+srZo+s471ByR87gnU02br0wgdJKGx0j/Hj4kg6YpKeHaOIyCyvIKqpsHvU8tv8IhzdA8gRn/Y7aiu7jvNmwfhrk7CTAS4e/l46dUtdDCCFEK6JSlJqKNFQ3ZMgQzztTqVi8eLHH9U1VcXEx/v7+FBUV4efnd/oXNKDiCit5pWYqbQ4ue2dFtfWxQSZuuzCBHjEBWG0OfL10GLRqHIpCYbmVUB8Dob4GDCfNOpFfZiavzEKFxY5Jr+HjZen8uPEwV/dsg9nq4PdtWW5juf/iRK7u2Qa9vYxgVTFaSzFaLz9nDwyvANeNLeVQkgllec4in15BziRFSSaodeAdfHa9M0qPQXmus8aAKdj55c/Tne+65LA777CV5wIqZ4LHN7J+ky2tQFNue558uXo/z/y8w+26CD8jX956PmqVimAfPQEmDwVuywug7BjYzaAxOLura43/XFeuMz1VWGwcK7VQUGZBp1Fj0qsJcuRjtBaAw06ZJoBteRClKyFQXYHRJ5Asmw8HyrSE+xm5Zfp6ckqcNTyuOT+W2y9MYOxHq8kvs7iLjDfGJdOvXRAXvrKEU4f9J0f78tiAAM4Lc6BRQaU+kCP2ADQaDYpDQatVE+qjx9sgvZ+auubY9uraL1syuf/bTXx0fW/8vZrwNZu721nHI/5C6DDizF/vcDinkbeUwlXv8+LCQ0T8M0W1EEII0Rqc8a24JUuW1EccwgM/Lx1+Xjq2HS50u/5gfjn/+XkHUyf2YMoPW6oKKmrUKlY+NoRIN9NEBnkbqgqpbjtSxA8pzpknIvyM7MoqxqhTU2mtXrfj923ZjD8vhjah4cBp7jTpTRCc6Hyc7EzuULnjE+p8NDS1BvzbOB+iVUvLKfW4Lru4EpNBS5vTTc9qCnQ+asFLryU2SEts0IkeXTNX5/LWgqO8Pi6Z22ZuOKWOSF7V/713bU+XoqUZuaVo1CqPCQ+AvTml9IkPrJbwANhyuITrvi9hxs3nMSgpDG+gQ63OQoimZ8P+fCL9jU074WG3wIq3nEn2xGFntw+1GrpPgL/fgXWfEhc8hk0HC+s0TCGEEKIpO+v+x3v37mXfvn1cdNFFeHl5ofwzFaI4d7klZrKLKzmQV0aEv5E2ASb8vHRo1CrsHiquJ4T68O8rOvPWgj0UVVh5c1x3Ss02ft2SiUatIincF7PdTsaxMqIDTQSYdOzLKSXC30iIj57JF7cnLthETKCJMb2jKbfYeWvBHrKKTsw80TnSF78aZrFoFJXFzjvmR3c475SHJoFPOOgaoPeHaJV6xwUyc/UBt+vahfpg0HgeNXi0uJIjBRVkFVUQE2Qi0t/oMvzFanNwuLCC7KIKTHotVrsDtUqFVqNif145Yb4GOgTpGNfOwXBdJQHW1Sy+qS2zdlby8dpj1Y5n0mtdEiI9YgLQaVREB3pxuMD9rEM9YwPQadQYtOpqU2Jf2T2Sq3u2odxqZ8HObDqE+xLqY5ChLKJZWpeR3/SHtmz9ztlDsv997mvq1JZXAHS8Arb/RGzngfxWaKeowtq0Ez5CCCFEHTnjb6p5eXmMHz+eJUuWoFKpSEtLo23bttx2220EBATwxhtv1EecrUZmYQV3frmBbUdOjLeN8jcy645+jOnZpqpXxskubB/C/B3ZLErN4Z2JPbA7FLYdKeLhH7ZWJUl0GhUPX9KBIwUVfLV2E50j/Xjs0iR+2ZzJxzf04d9ztrEr+0Rhs5ggL14e043HftrK0WIzGrWKOwe1w68pfUEqy4O/34bV756YSldrhKs/hvbDnb1NhKhjfeICCfLWu+0t8cRlHQnxdV/AM/1YKTd+vs4l2dA50pdPb+xDm0ATZquNzYeKeGL2Np4f1YUpP2zhkeFJzFy9v6q2xvjuQTyZeACv+ffj9c80zQkqFff1vpe2l43j8T9PzOY0JCmMDfvzq54btGoGtAvh182ZPDI8iQe/21wtxlBfA12i/Any1nNDvzg+W3liWt2bL4gn0KTnzi9TsDlO9Ch78rKOjOsTIz+eRLNSVGFld3YJF3VohJ6DtVV4ELb9CG0HVxv2dlba9IbMzcRnzALGsSurmL5tg899v0IIIUQTd8aFTB966CF0Oh0HDx7EZDrxo3LChAnMmzevToNrbYorrDwzd7tLwgMgs6iSGz5fx4PD2nPNeTFo1c4eNWoVXNo1ghv7x/HpinR2ZhXz/YZDaDVq3l6Y5tIrxGpXeGXebi7sEIqPQcvOrGI+Xp5On/ggXvtrl0vCA+BQfgX//W0n9wxOJNLfyKc39HbpXt8kHFgJq6bicivbVgk/3uT8sihEPWgTaOL7O/vRJepEHQQ/Ly0vj+nGefHuh6zklFRy24wN1XpX7Mwq4dEft1JYbiGzsJKbp6/nur6xvPRnKp0j/ViyO6cq4aHTqJjc20Dgn3eB7cSQFRQF3w3vcbHXXhJCvNGoVYxKjuT6frFM+ydpER9s4p2JPXlvyV5W7svD30vLlOEd8D2ph0b3aH/eGJcMKjDqNNw5qB23XBCPXqPGz0vLefFBvLlgT1XCA5xT977weyppR2UmCNG8bDxQgAJ0jGiqPT0UWPOhs4dGwqC62aVKBZ2vIrJyHzqVQ4qZCiGEaDXOuKfH/Pnz+euvv4iOjnZZ3r59ew4ccN/lW9ROXpmFxbtz3K47kFdObqmFZ67szE0XJLDvWCk6jYqVe/O475tNVd3QI/yMfLJ8n8dj/LI5k+Fdwpm98Qir9+Xx1OWdWJOe73bbfcfK6BETwKw7+hEX7H3uJ1iXynJh2avu1ykKbJoJw/8nhUZFvUgM8+XLW88nv8yC2eYg0KQn3NeAxsPQltwSM+m5ZW7XrdqXR36ZhU0HCym32IkP8SY1q4R7hyQy5fstVdtd1D6YkD2zwEPt6dBNU/nxxlmUavwJ9tazM6uYtyf2QK1ScbS4klfn7SI9t4w3xiXz0Pdb6BEdwEv/6oYKFXqtil3ZJUz5fguPXprE+D4xhPoaeOzSjtx0QQJ2h4OX/tjl8f34cOk+pl7jh7cMcxHNxLr9+QSYdET4NdGhkBkrIHsr9L65bqdF9wlFGz+A2LSj7NwfDBck1N2+hRBCiCbqjL+hlpWVufTwOC43NxeDwX23blE7FRa7p98zAOSWmkmOCSCnpJJ7vt7odptwfyNr0vNoG+JNYYW1Whf8rKIKesWduBtdbrbVGFO5xUbyP1NwNil2CxQf8bw+by/Yrc6ZY4SoBycXBD6dk6e3dafMYudgfjkqFVRa7QCoULnU1Aj31uBVnOFpF1B0mGAjBPt543AofPH3fv7cnl1tM6NOQ2G5laV7jrF0T/U6IBnHyly2jQ0yUVBm4Uih+xogAIcLKqi02iXpIZqNdRn5JIX7Ns1aZLZK2DDNOUV7aD2UCm43mNiMrWzfmwEMqPv9CyGEEE3MGX9Dveiii5g5cyb/93//BzinqXU4HLz22ms1TmcrTs/XqHVbPPC4CH8j5RYbkf7VZ4ZoE+DFY5cm0THClw5hvuzMKibEx4C3QcP7S/aSmuXsfp4cHUDGP3ec9Ro1PkYtWrXKpcv6yUx6LfN3ZNMu1KdqCs6jxZUczC9nX04psUEmEkK83c4SU6/03hDZE9I9TJEcf5EkPESTEV7D3WSdRoW/UUtyTACK4mxzKhVUWG2E+OjJLXUmLnfnWijq0B//tPnudxTZE/Q+AKjVKga2D3Gb9MgpqSQ+2MT+vHK3u7moQwj788rYe7SUwwXldIr0o02gFwMTQ9iR6b47fI9YfzYfLMRk0NAmwAuLTWHToQJ8DFq6tPEnzNeAUachq7CC/XllHMgrp12YD7GBJsL9m+iddtFiVVrtbD1cyLXnxzV2KO5tnwMVBdDrpvrZv9ZIfHgwKw77YDmwHn3cefVzHCGEEKKJOOOkx2uvvcbgwYPZsGEDFouFxx57jB07dpCfn8/ff/9dHzG2GmF+Bm4ZmMCHS6sPT+nfNpilu4+x/UgRFyeFMSQpjCX/DIUJ8zXwyr+6ATDlhy1sP6kmiJ+XljfGJfPmgj2kHyvjiu6R7MoqYVFqDv/q3YbNBwsZ0yua7zccqnbMwUmhzN95lPeX7CXIW8971/YkPsjEddPWVSVOjh//69v60r4hq+Ab/WHo05CxpHp3f6M/dBrZcLEIcRrBPnqGdQpjYWr14WvX9Y0j1M8AKhVxwSZW7ctleOcIZqcc5r4hiTz3604AUg4UkHPxcPwNb4H5lOSDSgUXPw3GE3VGBnUIJcCkq9bL5Nt1B3lkRBL3fbOpWiz9EgLxNmgZ88Eql15iHSN8+fD6Xvyw4RD5p+xPr1FzRbcobpuxAYvdQbtQH54d2Zn/+30nxRU2dBoVH9/Qm5hAEzdMW0d28YkZoeKCTcy85fymN3xOtGibDxVitSt0imyC9TzK82D7DxB3AXjXX5HRuJg4bIctpP0+lS53z5ShoEIIIVq0My5k2rlzZ7Zu3cr555/PJZdcQllZGWPGjGHTpk20a9euPmJsNQxaDbdckMDkixPx0jmnptOqVVzZPZJbBsbz7uI0Hv9pG3llFl4a042xvaPRqlXcPbgdG/YX8NXagy4JD4DiChuP/LCVKZd04N1revLSn7uotNl5/NIkesYG8p9ftnNxx1BuvzABo855Oeg0Kkb3aMOE82L4dHk6APllFh6YtZkjhZUEe7v2oMgpMXPLjPUcPenHTIMITYJrfwD/k+rLRPaAm/+EgNiGjUWIGgSY9LwwuhsT+kSj0zh/XBh1au4a1JZ7hyTipdMSG2Ri+k3nkXa0lJHJkVzRPYpQXwMPXdIBPy9nfvqBeXnkjf8ZIpNP7Nw/Bq79sVo3+DYBXvxwZ3+So/2rlkX4GXl0REf6JgTx+rhkl7bcv20QL47pzq3TN1QbFrcru4T/+20ns+7oT7tQn6rlbUO8eXtiDz5atg+L3dlDbd+xUj5cuo/r+zrvolvtCpmFldw+c4NLwgOctYoemLXZ7Uw4QtSXdRn5eBs0xAQ2seLcABtngloLbeu352xcgAYVCjsyiyBtQb0eSwghhGhsKkWpqYpE61BcXIy/vz9FRUX4+fmd/gX1LL/MzOaDhVTaHGjVKpbszmH2xiNVw16u7xvL81d1xWy1k1tmpqjcSmaRs86H3cMwlU9u6M3bi/awM7OExDAfpt98HuUWO2qVikCTDh+jlpxiM0UVVo4UVrAw9ShzNx3Banfd3wfX9aKk0srjP22rdozfJg+kaxv/asvrXXEWVBY6vyh6BYJ3SMPHIM5KU2t79a3cYiO31EyFxY5JryXcz4he65p7PlZSSWG5FZUKLn9nJf3aBjkTnBo1Oo2KHZnFtNGXc1k7PT46FRgDwC/S4zHzyywUlluwORT8jDrC/QyoVCrsDoWjxZWUVFrRazUEeevYfqSY6z5b63Y/KhXMf/AiAkx6iiqslFtspBwoYObqAy49v4779MY+3D5zQ7X/d2fhwxeRGNYE77q3YK2t7Z3sus/WUGGx8+iIjo0diqv8dPjlfug8EmL71/vhHllSwVDVev4btADuXA7qM74PJoQQQjQLZ1V1rqCggGnTppGamopKpaJTp07cfPPNBAUF1XV8rZLFpvD4T9s4Vmp2u/5QQQVWuwOTQUusQcvG0gLsDsVjwgMgPbeMcrOdDuE++Bq0aNQqOpwyHCUmyERZVjF3fpnicT+5pWYiPIzBL66ouVhjvfGLrPFHnxBNhUmvJTao5o/dUF8job5G0o+VYrE7WJ6Wy/K03GrblY/qwqQB8ac9ZpC3niDv6vVtNGoVUQFewIl6PMdK3H/mgHMUWaXVTqivgVBfA/N3ZPP8P0Nv3LH+0/NDp1FRbqm5YHKFxX6asxCibljtDlIOFPCvXtGn37ihbfgCfEIg+vwGOVy8v5qt5d3g6JuQ+jN0ubpBjiuEEEI0tDNO6y9btoyEhASmTp1KQUEB+fn5TJ06lYSEBJYtW1YfMbY6PkYNPWMDPK6/sH0Ixn+Gv4DzR43F5iDA5Hlau65Rfjw/qgtXdI/i4k7hlFTYKHKTpPAxavHWa9zswSk+2JtdWe6LGXpKhgjRUjkcCkcKKliwM5tPl6ezam9utSEcZ8uk1xDm63l2mG7Rdd+rql2o59oaPgYtVrvCTymHOZhXRkyQ56EBRp26qkSA1e7sYeKpZIBOo8Lfqw6n5BSiBlsPF1FpddAxoon1bjmSApkbIXEEqD3/Da5LCf5qdhUbsEX2hiUvgkOSj0IIIVqmM0563HvvvYwfP56MjAxmz57N7NmzSU9PZ+LEidx77731EWOr42PQ8eCwDmjU1X8l+HvpuLRLhMuyYG89R4sruHVggtv99YoLwGp3MOmL9by1YA+v/bWb4W8v591FaeSXud7ZDfczcPdg97VZLkgMJsik59t11YueXtE9gmAfmbJYtB6KorAjq5jLpi7n9pkp/O+PVK79bC3jP1rNwXz3M6OciXA/I49f6r77fZcov3qpRxDiY6B/W/fFEycNiOe9JXuZ8sMWhryxDI1axdCOYW63va5vHL9tyap6vvlQIaOSo9zvt388oTUkd4SoS2sz8vDSaUgIaULFcxU7pHwBgfEQ3rnBDpvgr6bSDnvb3gi5e2DHnAY7thBCCNGQzjjpsW/fPqZMmYJGc+JOhEaj4eGHH2bfvuqzjoiz0zbUm29u7+tSNLBvQiA/3tWfNoGu08P6GnVc3SuaLpF+PDo8qaoru7MgaRQvju7GvW5mavhsZUa1wqc6jYZrzo/lqcs7Vt19NWjVjO0dzb8v70ReWSVX9YjC1+Dsom/UqblpQDzPXtlF7taKViW7qJJbvlhPcYXr0I2D+eU89uMWCsvPrTinSqViaKcwXh3bndB/EopatYqrekTx6Y196iVREBngxatjuzOudzR6jfPPQ4BJx4PD2uNn1LJ4l3P2GbtD4V8frOLZkZ25tm8Mhn/qkvgZtUwZ3oGOEb78vs2Z9LiwfQije7bh31d04raTCib7GLQ8NKw9dw1uh5f+rEZaCnHG1qbn0yHcx+1NhUazbynkZ0DSZQ06i0q8vxoVsM0eA236wLJXpLeHEEKIFumMv2n26tWL1NRUkpKSXJanpqbSo0ePuoqr1TPqNPRNCGbWHf0orrSiUakIMOkIMFUfmw8Q5mtkYHsdXaL8uLRbBFabAy+9Bm+9ln/P3Ua5hzHzHy3bR8/YAHyNJxIWwT4Gbr0ggcu6RlJqdk45mV1UyeM/bWXbkWIu7hjGC1d3RadRoyjOO95hfjK0RbQumUUVHuvurEnPJ7/M4rG91laASc/YXtFc2D6EMrMdvVZNiLcek6H+kgQxQSb+M7Izdw1uh9lqZ9+xMr7fcIgVp9QVKTHb+GtHNk9e1onbBral0mrHS68lyt9IUYWVhQ8PQqdREWjS4/dPQvSxER25aUC8c1udljA/AzqNFE8UDcNmd7B+f77HXkeNwm6GTTMhomuDzzpm0qmI8lGx7ZidccnXwB9TYOfP0HVMg8YhhBBC1Lcz/uZ8//3388ADD7B371769esHwJo1a3j//fd5+eWX2bp1a9W23bt3r7tIW6njRQNrQ6/VEO7vRfhJywrLLWQVea4xkFtqxvLPrDAn02jUVWP2sworuOHzdRyf52fxrhwW78ohyFtP2xBvLkiU2VJE61NYXnPhXrObdnU21GoVkf5ep9+wDvkadfgadWQWVjD525UetztUUFG17cnCdBrC3JRM0GvVRDfFaUJFq7A9s5hyi51OkU2onsfOX6AiH3pNapTDJ/ir2ZJjh4FJENULlr0KnUfLTC5CCCFalDNOelxzzTUAPPbYY27XqVQqFEVxTolol26SDamowkJOsZnFu3Kw2BwM6RhGm0AjF7YPZevhIrevuSAxBF9jzZeBl15Dr9hAUg4UAM66Ik9f0QmNWsW2I0UEmnTsyykl3M+Aj1GGuIjWIS645qKffqdpV/Uhp7iS9GNl/L0vlwg/IwPbhxDuZ3QpfJxZWMGOzCK2Hi6ifZgPveICifL3Qu2mu7+XTkNytD9bPHx+XNQ+tOr/s4sq2JVdQsqBAuKCvembEESEnxGdVn48iaZhTXoeBq2atjUU7G1QlUWw7XuI7ttoU623DVAzK9WK1a6g6z4e5j0Be/6Ejlc0SjxCCCFEfTjjb+UZGRn1EYc4R4XlFj5fmcHUxXurlr2xYA9X9YjkgaEdmLlqPyVm19oDXjoNk/rHo9fWXCk+wKTn35d34l8frUKvUfPOxB68+Ecqe46WVm2jVat479qeDEwMkcSHaBVCfPRc3i2CP7ZlV1t3/9BEwvwatjhnZmEFt85YT2pWSdUyjVrFR9f14sIOoRh1GvYdK2XiJ2tcpqb1NWj55va+dG3jj+qUegKB3nqeubIz4z5eXdXT67jYIBNd2zhnkDmYX861n67hcEFF1XqDVs3MW8+nd2wgWhnCIpqA1fvySIrwRdtUejFs/Q4cDkgc2mghtAtQY3HA7nwHXcO7QnhXWP4aJF3eoPVFhBBCiPp0xn/54+Liav0QDSf9WJlLwuO4nzdncSC/nB/vHkD/tkFVy8+LD2T2PQOIDqxdt/mkCF++vb0ftw5MYPbGIy4JDwCbQ2Hyt5s4WuK+xoEQLU2ASc9zI7tw9+B2VdM8h/oaeGlMN8b2jkanaZhpJwEqrXamLkpzSXiAs+Do3V9vJKe4krxSM/d/u8kl4QHO2hy3zNjAUQ9T7XaO9OPrW/uSGOYsqqxVqxjZPZJvbu9LhL+R4gor/56zzSXhAc7hPbdO38DRYvlMEI3PanewLiOfzk1laEtJJuz6HRIGgb7xep7E+6vRqGBTzj89c7uNg8xNkL600WISQggh6tpZ9b/evXs37777LqmpqahUKjp27MjkyZOrFTcVDcNic/DF35574Lz+126+vrUvH93Qu6oOgb+X56Ko7ngbtPRrG0yor4FL317udhurXWF9Rr7LjDNCtGRhfkYeGtae6/vFYrEpGHVqwn2NboeK1Ke8UjNzNh1xu87mUFibkU+PmAB2ZBa73eZYiZmcEjMRbmqHmAxaBiSG8O3t/Sg129CqVQSfVEw1v8xSrcjpcaVmGxm5pdVmnBKioW07UkSF1d50kh4pM8DgA/EDGjUMvUZFnJ+KLTl2buiCs65HcCKseAPaDWnU2IQQQoi6csY9PX788Ue6du1KSkoKycnJdO/enY0bN9K1a1d++OGH+ohRnIbN7iC3zDk9pkGrpn+7YAYmhlTVFCgst2KxO/D30hMX7E1csPdZzyphdyhY7YrH9Z5msxCipdJrNbQJMJEQ4k2kh9oY9c3mUDwWTtVpVP881ASaPA89K620eVwHzl4sCSHexASZXGaPsdhrLthacJqCr0I0hNX78vDSaUhoCvU8jqXC/pWQeAlozm2Gp7rQNkBzoqeHSgVdx8L+FXA4pXEDE0IIIerIGff0eOyxx3jyySf573//67L82Wef5fHHH2fcuHF1FpyoHS+9huGdw+kZE0CPmABW7cvDZncw8bwYsosryThWhs2h8M3aA1hsDvrEB7F6Xx6HC8q5IDGEjpG+FFVY+XlzJooCQ5JCCfbRE+JjINTXWO1YCSHeZOSWuY2lb0IQmEuhJAtSf4WSbOgwHMK6gF9kQ7wdrUvhITi8Hg6shtD2zi/R/m2axBdp4VlmYQVbDxexel8uccHeXNwxjEh/I4Z/Co7aHQqZhRWs2pfLjsxiukcH0K9tEG0CvFBVFkHRYdgxF6xl0GkU4f7xdIzwZVe26/CWe/uHMiFJQ9SRmaiyivhz9CVsLo/mwd8zqbSeSFaoVBAVcPreGPllFg4XlPP71iwALu8WSaBJR7C3nrx/Eq+n6hDuw75jpczfcZSjxZUMTgqlY6QfETLNtWhAq/bl0rFJ1PNQYN1n4BcFUT0aORanxEA1Cw/YKDIr+BtUENsf/KNh5Zsw8evGDk8IIYQ4ZypFObU8Xc1MJhNbt24lMTHRZXlaWhrJycmUl5fXaYANobi4GH9/f4qKivDzayJdX8/Qofwy3l+yj1nrD7ksv6h9CM+P6sJVH/xN1yh/ru7Zhidmb8PuOPHPHuVv5OV/dee+bzZS/M/d3qt6RDGoQygXJDpnfzjZ/B3Z3PFl9TtAvWID+OyaTgQdmAc/34NL5cPQjnD9T84vUqJu5KbB9MuhNOfEMo3e+T7HDgBNw88ecqZaQts7Uwfyypj4yRqXqaS1ahWfTerDgHbB6DRqth0p4ppP1lBmOTEDlp9Ry+J7kgnZ+rHzx8jJ2g6m8NL36PHmiSnDH7wgjFsMi/Bb9ZLLprbo/mzu9yZjvzoxJO66vrE8cVnHalPPniy31MzLf+zix42HXZY/NCyRqAATj/64tdprLu0Szm0Xtq1WCDUxzIeZt5xfq0SLqB+tqe2ZbXaSn5vPv3pHc2X3qMYNZv9KWPoS9LkVQhJPv30DyCx1MGVJJTMuNzEo5p+/G2nzYdW7cO86CO3QuAEKIYQQ5+iMb3kMHjyYFStWVFu+cuVKLrzwwjoJSpy5rKLKagkPgOVpuSzdc4wofy9uviCBp+dud0l4AGQWVTJtZQbj+sRULft5cyYORWFR6lEcp2zfKzaAaZP60P6fwoYmvYabBsTzzsSeBDkKqic8AI7tghVvgs19sURxhsrzYe7drgkPALsFZl3n7GkjmpziCivP/LzDJeEBzuEpd36ZwtFiM0eLzdz5ZYpLwgPAoYA9b1/1hAdA+lL89//Fz/cMoEO4D0admnHtlWoJDwDt4dV0OPon/doGEuTtnJnpoWEdakx4gLMmwqkJD4C3Fu4lMcyHj67vRVywCQA/Ly0PDmvPU5d34trP1lT7ONibU8q7i9KotMq05qL+bT5YSKXNQZco/8YNxG6FlC+cNwGaSMIDINJbha8eNh49aYhb2yFgCoJV7zReYEIIIUQdOeNbwaNGjeLxxx8nJSWFfv36AbBmzRp++OEHnn/+eX755ReXbUX9s9gcfLFqv8f13284xNje0RwtrvQ47n952jGu7RvLtJUn7v7+uiWLLlF+HCs1u/T2CPE1MrSTkY7hvlTaHGjUKsJ99XgZdLB2QfWEx3Gbv4aBD0FAjPv1ovbK85zDWtwxF0PBfnmfm6D8cgsr0o65XWe2Odh9tIRIP2O1pAjAkA7BBO38zOO+VWs/IPnmkXxzez8cikLI8qc9buu35TM+u3ECJbpgwnyNaE5Th6TUbOPT5eke17/05y6+uKkPvWL7U2lzoFOrCPU18O26g1hs7j8PZm86wn0XJ9Im0FTjsYU4V3/vzcXXqK1KyjWa3b87E9XdJzZuHKdQqVQkBqjZePSkJKRGB51GwaavYMi/ncNxhBBCiGbqjJMe99xzDwAffPABH3zwgdt14PwjarfLXbyGYLU7qmZlcaek0oaPQUu+hzH34MxTeOk0dIr0rZr2sqTSikoFDg9JjDZBbr5AluV5DtRWCYpcE3XC7vnfEgBzSc3rRaOw2hwec4IAReVWgrzd12Px0YPOnO/5xZXF4LAT4mfA4XBAmfvkyvFtdSqFSDeztbijUhRigrxQpYOfUcf5Cc7pr9fvz6eowkpskBcOxTmbzclq+swx2xycpgaqEHVi5d5cOkX6oVY1fJHhKpYS2PwttOkNvuGNF4cH7QM1/JFuxe5QTiRBO1wGW7+HNR/A8BcaN0AhhBDiHJzx8BaHw1GrR20SHsuXL2fkyJFERUWhUqmYO3euy/rS0lLuu+8+oqOj8fLyolOnTnz44Ycu25jNZiZPnkxISAje3t6MGjWKw4erd8FuybwNWi7vGuFxff+2waxNz6N9uK/HbaIDvbDZHVyV3IZPb+xDu1Af+rUNRqtW4WfUUVJpZXd2MS/9mcqD321m3vZssooqqu3H3vZiz4FG9QK95xjEGTAGOLseexLSvsFCEbXnZ9QR6e+5gGe3Nv6E+BjQaar/OEs5XEZe/JWed544zHldAGq1GkvH0R43tSUMAa+a6zgoisLhgnJmrTvIU3O3E+HnxZx7LuD9a3vi76XD30vH1Ik9mHvPBUQHmnh67na+XnuAQ/nlHC8VNbB9iMf9d4nyw8eoqTEGIc5VqdnG1sNFdI1q5LolW793Dm9pf0njxuFBUpCaUivsKTgpE6k3QdJlsOELqChstNiEEEKIc9WoZczLyspITk7mvffec7v+oYceYt68eXz11Vekpqby0EMPMXnyZH7++eeqbR588EHmzJnDrFmzWLlyJaWlpVx55ZWtrpfJxZ3C3f6Y8jFoua5vLL9ty2JnVjFDksLcvv6+ixN5Y8EeXp63i8d+3MKzIzszMDGEK7tHoSgKszceYcTbK/h4WTpzNx3hrq9SmPjJGg4XnChcm19mZpclBFubvtUPoFLDZS+Dd3CdnXOr5hsBw190v67njeDt/t9ZNK5wfyPPjeridt3l3SII9TUQ6qPn7sHVx/vvOVqKuc0ACIir/mKdCS6c4vyR8g9V9HkQ7KZugNaAMvgpDKaa6xvsOVrKFVNX8sTsbfy8OZN3FqUx9sNV5JSY2Z9bxr6cUgrKrYz9aBVvL0zj582Z/HvOdq54dwW7/5lFJjbIm/PiAqvtW62C50d1IcjbUGMMQpyrtel52BwKXRuznkdptnM2s4QLwdA0E//tAtRoVLAh+5TvTp1Ggc3srEUihBBCNFO1mr1l6tSp3HHHHRiNRqZOnVrjtvfff//ZBaJSMWfOHEaPHl21rGvXrkyYMIFnnnmmalnv3r25/PLL+b//+z+KiooIDQ3lyy+/ZMKECQBkZmYSExPDH3/8wYgRI2p17JZSxf5QfjnvLk5j7qZMbA4Hl3SO4LERSTgUhYWpR/lm7UGmDE9ib04p36w7SH6Zhc6Rftw5qC1/783j+w0nCqF2j/bn3Yk9aBNo4kBeOUPfXOb2mBPPi+G5UV0w6jSsy8jjpi/W883EOBL3z8Jny+dgLsYR3ZfKi/8PU0x30MlsDXWmohAOrYMFzzgLxfqEO3/4drkafJpH0qOltL0zUVJpZevhIl74fSepWSUEe+u546K2jOnVpmqK6PwyC0t25fDWwj0cLqggNsjELQMTcDgcXBRWSZvtH+G18zuwm3G0H4F62PPOBMcpM/ZY8w+i/D0V/davwVaBPeFiHMOeRxXSAa3ec8Ihr9TMjZ+vY0dmcbV1PgYtr/yrO4qi8NScbVUzPp2sQ7gP39zejxAfA0eLK/lqzQFmrNpPcaWNXnEBPHNFZzpG+OKlb/ozDLVUraXtPf/rDn7bksU7E3ugaqzhLctfh8wUGDgFtE13OvH/rKykY5Cad4edMnR11buQuQke3AY6mWpaCCFE81OrpEdCQgIbNmwgODiYhIQEzztTqUhP91zsrsZA3CQ97rrrLlJSUpg7dy5RUVEsXbqUUaNG8eeffzJw4EAWL17M0KFDyc/PJzDwxN3E5ORkRo8ezfPPP+/2WGazGbPZXPW8uLiYmJiYFvHlr9Jqp6DMgoJzBgWAO2amUGa2Ma5PDIEmPQoONCo1MUEmlu85xrfrD3Iov/pQlUUPD6JdmA+fLN/Hi3/scns8g1bN4kcGE+pj4MHvNvPHtizUKhjaIZibkk2YdCo2HbUyb6+ZD67rRYiv3Nmtc6XHnHfi1BpnD5DGHLd+Gi257Z2pvFKzsxCwSkWYrwG1m2KiOSWVVFjsrE3P44tV+0nNKkGvUXNV12DGdfZCq1aRY9YzrFciWrX7jns2SyX20mOoFAWHwRejT/WeF6fam1PCsDeXe1z/9oQe6LVq7vl6o8dtFjx0UdWQOpvdQW6pGbvinO0p0NR0f/i1VK217V3y5jKiA72446J2jRNA3l749QFnMjrm/MaJoZa+3mlhfZadNdf7uCaIio7A3Ltg5NvQ+6bGCk8IIYQ4a7W6zZaRkeH2/+vb1KlTuf3224mOjkar1aJWq/nss88YOHAgANnZ2ej1epeEB0B4eDjZ2dke9/vSSy95TIg0d0adhsiAE70pCsstlFvsbDlcxJbDRVXLQ3z0PDC0A6/+tdvjvqwO59jeEjd3co8z2xw4HAoOxUGZ2bmdQ4EFu/NYsPtEUdMIPyP20+fXxNnwCW3sCGqtJbe9MxXsc/oEYJivkaNFlbz21x6OlTp/sFrsDn7Ycowftji3ubhjGBf3wONgRa3eiDbozGbysdlrbquVVjunS61ZT6pSqtWoiahl0VRRP1pj2ztaXElaTimX1lDzqn4psOFzZ8+7Nn0aKYba6xSs4bd9Ng4UK8T7n9TC/dtA3ABY+Tb0vMGZYBdCCCGakUat6XE6U6dOZc2aNfzyyy+kpKTwxhtvcM8997Bw4cIaX6coSo3dWJ988kmKioqqHocOHfK4bXPnZ9RxebfqX/jyyixEB3p57BQQ6mvA30sHwGAPdUAAescF4mPQYtRpGZXseUq7SzqHyd1d0araXl3xN+kY2slzGxzdsw16bd1+lPt56Qj2MIuMSgXhfkb8vHR4muk2wKQjQNp7k9Ia296KtFxUQNc2jVTP48hGyNoC7UeAh55YTUlSkBoVsDbLzY2OrmOhIANSf2nwuIQQQohzVaueHg8//HCtd/jmm2+edTAnq6io4KmnnmLOnDlcccUVAHTv3p3Nmzfz+uuvM2zYMCIiIrBYLBQUFLj09sjJyWHAgAEe920wGDAYWscwC7VaxRXdopi2MoOjxSe6NisKrM/I5/q+sXy55mC11z0/qgsR/0w/GRdkon/bIFanu06XqVWreHZkZwL/+XE0oF0wccHOGiAn8zNqufXCtnX+w0w0P62p7dUVo07DXYPa8dvWLErNrj9GEkK8OS/+9MNVzlSEn5FnR3bm/lmbq62beF4Mi3blYLM7uLZvHF+tOVBtm/9c2ZlwPxn735S0xra3fE8ObUO98TPqGv7gigNSpkNgPIR1avjjnwVvnYoEfxVrMu1M6HjKypD2ENkDVrwBnUc36WGUQgghxKlqlfTYtGlTrXZWl0XCrFYrVqsV9Sl3RzQaDY5/hl307t0bnU7HggULGD9+PABZWVls376dV199tc5iae7aBHrx410D+GT5PuZuysShKIxKjmLi+TF4G7QkRwfw3pK9ZBZW0jnKjycu60iXKL+qf88QXwNvT+zJnI1H+PzvDIoqrPRvG8yjlybRPsyn6jiRAV58c3s/pv+dwfcbDmOxObi0awT3D00kNtDkKTwhxGnEBpn45b4LeGdRGvN3HMWgUzOhTwyTBsQTWQ/DRtRqFYM7hvHN7X15+c9d7MoqoU2gF/cMbkeIj577v92MgrO2R8/YAN5fvJfDBRUkRfjy+GVJdG8TgMZTNxAhGoDDobAiLZdBHRpp+N/+FZCfDuff2awSBJ2CNfx9xOa+x2zXsbDgadi3yDlFthBCCNFM1KqQaX0pLS1l7969APTs2ZM333yTIUOGEBQURGxsLIMHDyY3N5f33nuPuLg4li1bxt13382bb77J3XffDcDdd9/Nb7/9xvTp0wkKCuKRRx4hLy+PlJQUNJrajTttLVXszTY7+WUWAAJNeoy6E+9PTkkldruCUaep6rlxKrtDIbfUjMOh4G3Q4ufl/u6Z5Z/jKIC/lw6TzNAgPGgtba+ulJttFFVaUQHBPnp0tfyMOxf5ZRbMVjtajYpQX6Pbz5FjJWZsdgcGnYYgD58fomlp6W1vy6FCrnr/b54d2ZmOEQ18fg4bzLkTvAKh140Ne+xztPmonVfWmVk0wZt2Aad8vigK/DEFTCFwy5+NE6AQQghxFmr9azQ9PZ2EhIQ67c2xYcMGhgwZUvX8+DCaSZMmMX36dGbNmsWTTz7JddddR35+PnFxcfzvf//jrrvuqnrNW2+9hVarZfz48VRUVDB06FCmT59e64RHU2C1O7DYHHjpNG5ncDhXZpsdhwO89BqPd4XDfF27otv+icmg01TdsdWoVbXqsq7XauqvaKHDAbYK0OhB0whdloWogcOhUGG1o9eq0WlqP5yrwmJHrQaDtubPLZNBi8lQd0lERXHGq1Wr0SuVgKratNKnJjEM2uqfI6EyK5NoYpbuPoa3XkP7MN+GP3jafCg5Ct0nNPyxz1HHYDUaFaw6Yq+e9FCpoNt4WPICHFjlLG4qhBBCNAO17umh0WjIysoiLMxZUG/ChAlMnTqV8PDweg2wITTWHa9Ss41D+eXMWLWfA/nl9EsIYnTPNkQHmuqka3huqZnd2SXMWLWfcoudMb3a0L9dcI3d4SssNg4XVPDlmgOk5ZTSIzqA8efFEB3odUY/4uqcokDhQdgxG/YtBr9o6HsnBCWAsZGK1Ilz1lLuNtsdCocLypmz8Qhr9+cTH2zixv7xxAaZ8K4hSZFZWMGqvbnM3ZyJSa/hpgHxJEX41mpml3N1pKCCpbtz6B5QQWz5Tvx3fgUqNfS5DaJ6OKc/Fi1WS2l7nox+/28MWjUPDuvQsAe2m+HH2yAwtlkmPQD+u6qSaB81n17qZliq4nBOwRuYADfOafjghBBCiLNQ66SHWq0mOzu7Kunh6+vLli1baNu2bb0G2BAa48tfpdXO71uzmHJ83sl/eOs1fH9Xf7pEndsP+dxSMy/8tpO5mzNdlrcL9ebLW/sSFVA98WG1OVi65xh3frkBx0lXhUGr5uvb+tInPuicYjonObvg8xFQWei6/LLXoMe1YPBx+zLRtLWUH147jhQx7uPVlFvsLsvfGp/MZd0iXYaSHZdZWMG1n65h/ymFf8f2iubJyzvWa+LjQF4ZYz9azXsjI+m95n60mRtcN0gYDGM+lsRHC9ZS2p47eaVm+rywkDsualvj7GP1YvtPsHE6XPAweAc37LHryJw9Vn5Pt7J5ki86jZsbMBnLYfmrcNsiiG76U/EKIYQQMp1GIzlWYubJ2duqLS+z2Hn0h63klZrdvKr20o+VVkt4AOw7VsZ36w9hszuqrcspMfPQd5tdEh4AZpuDB7/bzNHiynOK6ayVF8BvD1ZPeADMexzKcho6IiGq5Jaaefj7LdUSHgBPzN7GsZLqbdlqd/Dl6gPVEh4AP248XG0GpLpUbrHx+l+7CfM1kFi0unrCAyBjKRxaV28xCFGflu4+hgL0iAlo2ANbymDbD9CmT7NNeAB0D1NTZoWNR6t/pgEQdwEExMKSFxs2MCGEEOIs1TrpoVKpqtXzqMv6Hq3NnqMlWNwkHgB2ZhVTUG496307HArfrK0+De1x3647SN4/hQhPll1cUW1KzOMOF1RUFS9scJUFcHC1+3WKAw6vb9h4hDhJYbmV3UdL3K4z2xzsO1ZabXleqZnvNxzyuM9v13luv+eqoNzKH9uzGdPJi+CdX3recN0nYHZ/XkI0ZYt2HSUx1JsAUwMX1d35M9gqoe3FDXvcOpbgr8ZPD0sPuf8+gFoD3Sc6Z3GR5KgQQohmoNYV8RRF4aabbsJgcHa5rqys5K677sLb29tlu9mzZ9dthC2U1UPC4zjHOUyqo6Bgtnnev82h4G5Uk+3ULh51GNM5cXi423Sc7ewTREKcq9O1C4ubtqhQ82eA2WZ3P2VkHVAUBbtDQacC7DUkMh1WZ+FgIZoRs83O0t3HuKJbZAMfuBh2zoGY88GredeZUqtUJIdpWHzAxuN9PWwUPxC2fQ+LX4BJvzRofEIIIcSZqnVPj0mTJhEWFoa/vz/+/v5cf/31REVFVT0//hC10zHCD0+/Z2KDTAR4mA62NjRqNeP7xHhcf0W3CIK8q9cLiPL3Qu+hWGmwt77xpqL0CoCwzp7Xx57fYKEIcSp/Lx3Rge6LA6tVkBRRffaIQJOey7p6rpcxrk9MvfWk8zPquLB9CH+kWyjqMMbzhsnXNfsfb6L1WZOeT7nF3vA1qLbPBrvdWQ+nBegZrmF3gYMjJR4Snyo1JF8LGcsgfVnDBieEEEKcoVr39Pjiiy/qM45WJ8RHz+QhiUxdvNdluUat4uUx3QirxdSwNekc5UfvuEBSDhS4LA/y1nP7hW3Ra6snN0J8DTx1RSee+2VHtXUvjulGuO+5xXTWvENh5Dsw/XKwn9Kro+9d4N3AheqEOEm4n5GXx3Tnxs/XVquHc//Q9m4Lkhp1Gu4ZkshfO49SeMpQtvPjg0gKr79pNv28dDx9RWdGv/83mRddgX/ATOfMSCcLaQ+JQ+stBiHqy4Kd2YT7GYjxkIisFxWFkPoLxPZrMUW1k0M1aFSw8ICNSV093PCI7Q8hSbDwWbh9CR7v5AghhBCNrNazt7RkjVXFvqDMwubDhby7KI2sokp6xARw/9D2xId44+VmtoczdbSokvk7s5mx+gCVVjuXd43ghv7xxAS5mYbuH0UVVnZkFjF1URoH8srpFOHLA8M6kBjmU+PUm/XOZob8dFj+GhxaCz5hMPARiO0L3iGNF5c4Jy1lBokKq52MY2VMXZTGlsOFRPp7cf/QRJKjAwj00ENKURQOFZQzY9UB/tqRjZfOOWXtsM7hhJ9j0vN0bHYHh/LL+XbdQSYkqYlI/wnv1B+cP1p6TYJu48C/Tb3GIBpXS2l7J7M7FPq+uJC+CcFc3y+u4Q687lPYMw8uehT0nv++Njcvra7E16Di6yu9PW+UvQ3+ehL+NQ26jW244IQQQogzIEkPGv/LX2G5BbPNgY9BW+eJBUVRyCuz4HAoBHrr0Glql0wpqrBSabXjrdfiY2zEZMepLGXO4ooaPZgacQpdUScau+3VtTKzjVKzDYNWXesiihabncJyK2q1ipB6nKbWnUqrneIKK15a8HUUASowhYBaJvZq6Vpa2wNYl5HP+I9X8/yoLnSox95SLsqOwezbIWFQi+sdtWC/lZnbraRM8sXfUEMvjsUvQPFhuC8FdI3UI1QIIYSogXyzbQICTHrC/Yz10pNCpXL+kArzM9Y64QHOOgXhfsaqhEe5xUZeqZlyi4dq7g1F7w2+EZLwEE2St0FLuJ/xjGaN0Gs1hPkZ3SY8jre7Cmv9tDujznlsX5MRfMKdPagk4SGaqT+2ZRHkrScxrAGHmGyd5UzCx13QcMdsIL0jNNgVZ/Kj5g1vhpIsWPVuwwQmhBBCnKEmdAtfNEVlZhsZuWV8tGwfu7NLSIrw5a5B7Wgb4o2pMYe7CNGClZltpB8r5cOl+0jLKaVjhC93D25HfIg3Jr20OyFOZXco/Lo1k34JQagbqrZE8RFIWwDth7fIHg5BRjUdg9T8us/K2KQaErn+baDTVbDideg+HgIbcGiREEIIUQtyS094ZLM7WL7nGCPfW8lvW7NIyynlt61ZjHxvJcv2HMN2mml3hRBnzmp3sHhXDiPf+5s/tmeTllPKr1uzuOLdlaxMy5V2J4Qba9LzyCu10L9dA9Z42vyNs/dhbP+GO2YD6xul4e/DdgoqT/O5kzwR9D7wxyMgo6aFEEI0MZL0EB4dLTHz+E9bq31/URR4fPZWckrMjROYEC1YTnElT87eVm25osBjP0m7E8KdOZuOEOFnpF1oDUU361J+OqQvhbYXg+bsp5hv6vpGaXEAv6efZoidzgTn3wFp82H7Tw0SmxBCCFFbkvQQHuWVmimudP9Fp7jCRm6p/PgSoq7lllooNbtvd4XlVvLLLA0ckRBNW7nFxh/bshjYPgRVQw1tSZnunDksuk/DHK+RBBhUdA9VM2fPaep6AMQNgPiBzt4eJdn1H5wQQghRS5L0EEKIZqSBftIJ0WzM255NucXOwMQGGtqSvQWOpDhreajPfXr5pm5gtJaUo3bSC+2n37jv3c7/zr0HHDIUTwghRNMgSQ/hUYiPAX8v9912A0y6Bp9eU4jWIMTXgJ+HaaKDvPUEedd+ZhghWoNv1x2ka5Qf4X4NUExUccC6zyAgFsK71v/xmoDzIjT46OD73bXo7WH0hwsehH2LYNU79R6bEEIIURuS9BAehfkaeGNcMupTbi2rVfD62GTCfCXpIURdC/c18Nq4ZE7tpa9Wwevjkhvmh50QzcTenBLW7y9gSMewhjngvkXOeh5Jl1OtkbZQeo2KAW20/LDbisVeiyKlbXpDt3Gw6L/OuidCCCFEI5O5D4VHWo2aCxKD+eOBC/lsRUbVlLW3XZhAXJAJrUZyZkLUNa1GzYXtQ/jj/gv5dEU6aUdL6RTpy60D2xIbbEJ9ahZSiFZsxqoDBHjpOC8+qP4PZimDlBkQmdzqpmUdFq9l/n4b8zJsjEqsReHWHtdDXjp8fyPcuhBCO9R/kEIIIYQHKkWRucWKi4vx9/enqKgIPz+/xg6nSTLb7FSY7XgZNBi0LX8Ms2gY0vZqZrbaqbBIuxN1ryW0vaIKK/1eXMTl3SIY2zum/g+4/jPY9RsMfBi8Aur/eE3MC6sr0athztU+tXuBpQz+fAwcNrh1Afi3qd8AhRBCCA/kVn0rU1xhJbuo8oxngDBoNQR46+WHlxANyKBr2HanKArHSsxkF1VSYTnNFJVCNLKv1hzA5nAwrFN4/R8sPwNSf3FOUdsKEx4AlyZo2ZTjICW7lp8Nem8Y9rwz6THjSijOrN8AhRBCCA8k6dFKlJltbD5UwL3fbOTyqSu48fO1LNiZTX6ZTDsrhICjxZXMWL2fcR+t4sp3V/DszzvIOFaK3dHqOwOKJqjCYufzlRlc1D6UAFM9F/dV7LB6KphCnFOytlK9wjVE+aj4eMsZ3DTxDoHh/wNLKXw+AvL21V+AQgghhAeS9GgFFEVhdXoeV3+wihVpueSXWdh+pJjbZ6Yw/e/9lJrljq4QrVlOSSUPzNrEc7/sZH9eObmlFr5POcyV764kI7esscMTopov1+ynsMLKqOSo+j/Yzp/hWBp0uRo0rbcUmlqlYlSijvn7baTm1WL62uN8I2DEy6AoMO0SOLim/oIUQggh3JCkRytwtNjMv+dsw131lveW7CWvVHp7CNGapeeUsSY9v9ryMoudNxfsptRci6kqhWggRRVWPli6j8EdQgmr79mMCjJg40yIH9Dqipe6c0EbDeEmFW+uP8PvDT5hcOkr4BsJ06+E9dNw+6VECCGEqAeS9GgFCissHC12/wXFocC+Y6UNHJEQoin5davnsfYLdh6lpEJ6g4mm473FaVRa7YzpFV2/B7JVwrJXwSsI2o+o32M1E1q1irFJOhYcsLGhtrU9jjP6wSX/B+2Hw+8Pw0+3QWVx/QQqhBBCnESSHq2ARlXzFJd6KU4qRKvmpfP8GWDQakBmyRVNxO7sEj7/ez8ju0cR5F2ftTwUWP0elGRDj2tBU4tpWluJAW00tPVX8dzflWde80ejg353w0WPwu4/4KOBcGh9/QQqhBBC/EOSHq1AoLeeDuHup5jz0mmIDzY1cERCiKZkdE/PU0mO7xNNcL3+uBSidmx2B4//tJVwPyMj67uWx/bZsG8JdBnjHJohqqhVKiZ11bM918HXO89y6FvCIBg5FXRezgKnS14Cu/QoE0IIUT8k6dEKhPgYeGt8D7z1rndz1Sp4a0IPwnwNjRSZEKIpiA7w4vYLE6otjws2ceuFbaU3mGgS3l+yj62HC7nrorboNPX49SV9CWz4HNoOgajk+jtOM9YhSMOwOC0vra3kULHj7HbiG+Gs89F9Aix/DT4fDvnpdRuoEEIIAagURSpJFRcX4+/vT1FREX5+fo0dTr2wOxSOFFbwx9Ys1mTkkRjqw4TzYmgT6IVJ33qr0YvG1RraXnNRUGYhI6+Mr1YfqJoVo29CEJEBXo0dmqgHza3trUg7xo3T1jGmVzRje9djLY+M5c4f4FE9oeu/4DTDQ1uzcqvCk8srifRR8eMob3Sac3ivju2CFW+AuRiufBu6j6+zOIUQQghJetD8vvydK7PNjk6tRq2WL3OicbW2ttcc2B0OHA7QaaUjYEvWnNpe2tESxnywinZh3jw6vGP9/e3a/Tus+Qgik6HrWFBLGzidvQV2nv/bzPiOOv53oRHVuSSJrOWw5kNnT5ueN8DlrzmHvwghhBDnSG7xtzJWu52CMguKAt4GLX5eUpxNCHGCRq3G3ciB4gorpWYbKpVzyFy9Di8Q4h8H8sq47rO1BHrrmXxx+/pJeNgtsO5TZ2HNuAGQdIUkPGopMVDDLd31fLLFQoS3mvt7n8NwWZ0JLpwCEd1h7YeQuRkmfAlB1YfeCSGEEGdCkh6tSFZhBdNX7eebtQcps9i4IDGEJy/rRGKYt4zZF0K4ZbHZScsp5cU/Ulm1Lw9fg5br+8VxQ/84Iv3lLqyoP3uOlnD9Z2vRqlU8eVnH+hmKmZ8BK9+EwoPQeTTE9q37Y7RwQ2K1FFYqvLnBTKVN4ZHzDajPpcdH+0sgOBGWvgSfDIKxX0Di0LoLWAghRKsjw1toXt18z9bR4kpu+mIdqVklLsv1GjU/33cBnSJb5nmLpq01tL3mbseRIkZ/8DdWu+ufiq5Rfky76TzC/YyNFJk4F0297a1IO8Y9X28kyFvPE5d2JMBUxzMIWcth6yzYMRdMIdBtPPjX84wwLdxv+6x8vdPK0FgNrwz2IsTrHHvLmEthxWuQuQmG/8851a3UWBFCCHEWpP9mK5GaVVwt4QFgsTt4dd4uSirPcto5IUSLVVRh5X9/pFZLeABszyxmb05pI0QlWjKHQ+GDpXuZ9Pk62oV6858rO9dtwsNugV2/wU+3wc5foO3F0P8+SXjUgSvb6Xj0fAPrs+1cPKuUz7aaKbeew301gw9c/B9nD5y/noSf7wObpc7iFUII0XrI8JZW4s/t2R7XLU/LpbTShq9R6nsIIU4oM9tYnZ7ncf38HdlckBjSgBGJliynuJIpP2xhZVouV/Vow7je0XVXw8NugbT5sO0HKM+HqB6QeAl4BdTN/gUAvcI1vDbYi+92WXhxjZl3UsyMStRxWYKO8yI1GM50hhe1BvrcAoFxsOo9yN8HE74Cb/ncEUIIUXuS9Gglgrw93ynzNWqlx6gQohq1Crz1WkrNNrfrA2v4XBHiTPy+NYun5mxDrYInLutI9+iAutmxtQL2/AnbZ0NlEUR0c84M4hNWN/sX1fgZVNyebGB0eweLDtiYn2Hj651WjFo4P0JDvygtF7TR0jVEjaa2Sa12Q8E3Cpa+CJ8MhmtmQUTXej0PIYQQLYfU9KDpj22uC7uzixnx9gq36+4bksiDw9qjldkYRANrDW2vObPa7bz21x4+WZ7udv2Chy6ifbhvA0cl6kJTaXv5ZRb+8/N2ftuaxfkJQdw6MAG/uuh1aC6G1F9h589gq4DInpAwGHykh0BDUxSFA8UKW4/ZSc21syvfQaUdAg0qLo7TMrKdjoHRGrS1SYCU5sCS/0FJJlz1AXQdU/8nIIQQotmTnh6tRKS/F09d3okX/0h1Wd6tjR/X9YuVhIcQohqdRsPNF8SzMu0YO0+pCfTsyM5E+EsRU3F2FEXh921Z/OfnHVhtDu4bksiAdsGozrXbYUU+7JgDu/4AxQ5tzoOEC2UYSyNSqVTE+6uI91czKlGHzaGwt8DB5hw7qzNt/LTHSrhJxbWd9VzfWUdwTQVQfcLgsldg1bvw481weD0Mex600utMCCGEZ9LTg6Zzx6u+lVRYySqq5NetmRSUWbi0ayTtw31k9gXRaFpL22vujhZXsju7hPk7sgnyMXBl90gi/Y1SB6gZa8y2l1lYwX9+3s7C1BzOTwji5gHx516stDQbtv0Eexc460DE9IO4C5zFMEWTpSgK6UUOlhyws/KIDRVwTSc9d/XQE+5dQ/JDUWDXr7Dhc+eQpX9Ng+B2DRa3EEKI5kWSHsgPLyEai7Q9IRpHY7Q9q93BjFX7eXPBHgxaNZMGxNM3Ifjcdpq/z1mvI2M56LyciY7Yfs7/F81KiUXhrwwb8zKs2BxwQxc9d/fQ19zzI3cPrHgdKgpg6LNw/h3OpJcQQghxkkYd07B8+XJGjhxJVFQUKpWKuXPnuqxXqVRuH6+99lrVNmazmcmTJxMSEoK3tzejRo3i8OHDDXwmQgghhHBHURQW7zrKpW8v58U/UhmYGMLr45LPPuHhsMOBv2HeE/DL/ZC1BTpeAYMeg3ZDJOHRTPnqVYxN0jF1qBdXttPy9U4LF35TymvrKims9HB/LqQDXDnVWeh03pPwySDY/3fDBi6EEKLJa9SaHmVlZSQnJ3PzzTfzr3/9q9r6rKwsl+d//vknt956q8u2Dz74IL/++iuzZs0iODiYKVOmcOWVV5KSkoJGI9l+IYQQojEoisKKtFymLkpjw4ECOkf68b+ruxEf7H12OyzYD+lLYO8i5539wHjocS2EdQG11KVqKUw6FWOT9IxI0PHrXiufbbXwxTYL13fWc1M3PVE+p/xb64zQ9y5oOwTWfQTTL3cmQS6cAnEDkOnphBBCNJnhLSqVijlz5jB69GiP24wePZqSkhIWLVoEQFFREaGhoXz55ZdMmDABgMzMTGJiYvjjjz8YMWJErY4tXeyFaBzS9oRoHPXZ9nJKKvl9axbfrD1IWk4p7UK9+VevaHrEBJxZoVK7BY7tgiMpcHANFB0GvQkiukP0+eAXWadxi6apyKzwR7qVhfttmO0wNE7LvzroGBSjxag95XpSHM5eQFu/cybJwjo7E2Odr4KA2EaJXwghRONrNrO3HD16lN9//50ZM2ZULUtJScFqtTJ8+PCqZVFRUXTt2pVVq1bVOukhhBBCiDPncChkF1eSmlXMpoOFrNyby5ZDhWjUKnrFBfL0FZ3oHOlXc7JDcTh7bpRkQdERKNwPx9Igfy/YraD3gdAk59374ETQNJuvLqIO+BtUXNNJz+j2OlYctrHkoI0759vw0sKAKA39orR0D9XQIUhNoFEN8Rc6a7tkboa0v2DRf2H+086hMHEXQJtezt5Bwe1kVh8hhGglms03hxkzZuDr68uYMSfmZM/Ozkav1xMYGOiybXh4ONnZ2R73ZTabMZvNVc+LiooA550vIcS58/X1dfsjR9qeEPWrrtvevmNlPDp3F3uPldc6hnBfPT39S/E/NJe0AzbSav3KkwWB5gIUYyCKwQdKVVCqwL6z25toOc7TQRtfLzaWBbHoICw6aPewZeI/D6dhmRt449jH+Kd8UetjKVojFSPewJY06rTbemp7QgghGl+zSXp8/vnnXHfddRiNp59eVVGUGv/wvPTSSzz//PPVlsfExJxTjEIIJ09d5qXtCVG/6rrteXcdSsgVD9Xq2PaKYlAgsxxKCsGg6Vmr151MwTkbqXOHgOWMdyFaCT0V6AE7akpVp68Ts9DRh+0VP9BDW4q3vnbJCZWtkj9fu4OxP9xw2m1lmKYQQjRdzaKmx4oVK7jooovYvHkzycnJVcsXL17M0KFDyc/Pd+ntkZyczOjRo91+wYPqd7wcDgf5+fkEBwefUZa+uLiYmJgYDh061KL/0Ml5tiwNcZ61vdvsqe01t3+L5hYvNL+YJd7aOde215I1t2uoIch74upc3g/p6SGEEE1Xs+jpMW3aNHr37u2S8ADo3bs3Op2OBQsWMH78eMA548v27dt59dVXPe7PYDBgMBhclgUEBJx1fH5+fq3iy4KcZ8vSGOd5pm2vuf1bNLd4ofnFLPGenbr+u9ecNZV/k6ZE3hNX8n4IIUTL0qhJj9LSUvbu3Vv1PCMjg82bNxMUFERsrLPKdnFxMT/88ANvvPFGtdf7+/tz6623MmXKFIKDgwkKCuKRRx6hW7duDBs2rMHOQwghhBBCCCGEEE1PoyY9NmzYwJAhQ6qeP/zwwwBMmjSJ6dOnAzBr1iwUReGaa65xu4+33noLrVbL+PHjqaioYOjQoUyfPh2NRlPv8QshhBBCCCGEEKLpatSkx+DBgzldSZE77riDO+64w+N6o9HIu+++y7vvvlvX4Z2WwWDg2WefrdZluKWR82xZmsN5NocYT9bc4oXmF7PEK86V/JtUJ++JK3k/hBCiZWoyhUyFEEIIIYQQQggh6pK6sQMQQgghhBBCCCGEqA+S9BBCCCGEEEIIIUSLJEkPIYQQQgghhBBCtEiS9BBCCCGEEEIIIUSLJEmPWvjwww/p3r07fn5++Pn50b9/f/7888+q9Yqi8NxzzxEVFYWXlxeDBw9mx44djRjxuXvppZdQqVQ8+OCDVctawnk+99xzqFQql0dERETV+pZwjscdOXKE66+/nuDgYEwmEz169CAlJaVqfVM41+XLlzNy5EiioqJQqVTMnTvX47Z33nknKpWKt99+u8HiO1Vt4k1NTWXUqFH4+/vj6+tLv379OHjwYMMHy+njLS0t5b777iM6OhovLy86derEhx9+2CixgvNz57zzzsPX15ewsDBGjx7N7t27XbZpCtdtbeO1Wq08/vjjdOvWDW9vb6KiorjxxhvJzMxslHhbqrq4zs1mM5MnTyYkJARvb29GjRrF4cOHG/As6kZdtaHW8n7Uto22lPdDCCFaK0l61EJ0dDQvv/wyGzZsYMOGDVx88cVcddVVVV8SXn31Vd58803ee+891q9fT0REBJdccgklJSWNHPnZWb9+PZ988gndu3d3Wd5SzrNLly5kZWVVPbZt21a1rqWcY0FBARdccAE6nY4///yTnTt38sYbbxAQEFC1TVM417KyMpKTk3nvvfdq3G7u3LmsXbuWqKioBorMvdPFu2/fPgYOHEjHjh1ZunQpW7Zs4ZlnnsFoNDZwpE6ni/ehhx5i3rx5fPXVV6SmpvLQQw8xefJkfv755waO1GnZsmXce++9rFmzhgULFmCz2Rg+fDhlZWVV2zSF67a28ZaXl7Nx40aeeeYZNm7cyOzZs9mzZw+jRo1q8Fhbsrq4zh988EHmzJnDrFmzWLlyJaWlpVx55ZXY7faGOo06UVdtqLW8H7Vtoy3l/RBCiFZLEWclMDBQ+eyzzxSHw6FEREQoL7/8ctW6yspKxd/fX/noo48aMcKzU1JSorRv315ZsGCBMmjQIOWBBx5QFEVpMef57LPPKsnJyW7XtZRzVBRFefzxx5WBAwd6XN8UzxVQ5syZU2354cOHlTZt2ijbt29X4uLilLfeeqvBY3PHXbwTJkxQrr/++sYJ6DTcxdulSxflv//9r8uyXr16KU8//XQDRuZZTk6OAijLli1TFKVpXrcnOzVed9atW6cAyoEDBxowstbjbK7zwsJCRafTKbNmzapaf+TIEUWtVivz5s2r95jr09m0odb0frhzahttye+HEEK0FtLT4wzZ7XZmzZpFWVkZ/fv3JyMjg+zsbIYPH161jcFgYNCgQaxataoRIz079957L1dccQXDhg1zWd6SzjMtLY2oqCgSEhKYOHEi6enpQMs6x19++YU+ffowbtw4wsLC6NmzJ59++mnV+uZyrg6HgxtuuIFHH32ULl26NHY4NXI4HPz+++906NCBESNGEBYWRt++fWscstPYBg4cyC+//MKRI0dQFIUlS5awZ88eRowY0dihAVBUVARAUFAQ0PSv21Pj9bSNSqVy6XUl6tfprvOUlBSsVqvLdRUVFUXXrl2bxHV1Ls6mDbWm98PTNie30Zb8fgghRGshSY9a2rZtGz4+PhgMBu666y7mzJlD586dyc7OBiA8PNxl+/Dw8Kp1zcWsWbPYuHEjL730UrV1LeU8+/bty8yZM/nrr7/49NNPyc7OZsCAAeTl5bWYcwRIT0/nww8/pH379vz111/cdddd3H///cycORNoPv+er7zyClqtlvvvv7+xQzmtnJwcSktLefnll7n00kuZP38+V199NWPGjGHZsmWNHZ5bU6dOpXPnzkRHR6PX67n00kv54IMPGDhwYGOHhqIoPPzwwwwcOJCuXbsCTfu6dRfvqSorK3niiSe49tpr8fPza+AIW6/TXefZ2dno9XoCAwNdXtcUrqtzcbZtqDW9H6dy10Zb6vshhBCtibaxA2gukpKS2Lx5M4WFhfz0009MmjTJ5YeMSqVy2V5RlGrLmrJDhw7xwAMPMH/+/BrrDzT387zsssuq/r9bt27079+fdu3aMWPGDPr16wc0/3MEZ6+DPn368OKLLwLQs2dPduzYwYcffsiNN95YtV1TPteUlBTeeecdNm7c2GRiqonD4QDgqquu4qGHHgKgR48erFq1io8++ohBgwY1ZnhuTZ06lTVr1vDLL78QFxfH8uXLueeee4iMjKzW26uh3XfffWzdupWVK1dWW9cUr9ua4gVnwcSJEyficDj44IMPGji61u1sr/OmcF2di7puQy35/YAzb6PN/f0QQojWRHp61JJerycxMZE+ffrw0ksvkZyczDvvvFM188ep2f6cnJxqd1KaspSUFHJycujduzdarRatVsuyZcuYOnUqWq226lya+3meytvbm27dupGWltZi/i0BIiMj6dy5s8uyTp06Vc0i0hzOdcWKFeTk5BAbG1t1TR44cIApU6YQHx/f2OFVExISglarrfF9b0oqKip46qmnePPNNxk5ciTdu3fnvvvuY8KECbz++uuNGtvkyZP55ZdfWLJkCdHR0VXLm+p16yne46xWK+PHjycjI4MFCxZIL48GVJvrPCIiAovFQkFBgctrG/u6Ohfn0oZa0/txXE1ttCW+H0II0dpI0uMsKYqC2WwmISGBiIgIFixYULXOYrGwbNkyBgwY0IgRnpmhQ4eybds2Nm/eXPXo06cP1113HZs3b6Zt27Yt4jxPZTabSU1NJTIyssX8WwJccMEF1aYp3LNnD3FxcQDN4lxvuOEGtm7d6nJNRkVF8eijj/LXX381dnjV6PV6zjvvvBrf96bEarVitVpRq13/DGg0mqpeKw1NURTuu+8+Zs+ezeLFi0lISHBZ39Su29PFCyd+TKWlpbFw4UKCg4MbPM7WrDbXee/evdHpdC7XVVZWFtu3b28yn4e1VRdtqDW9H3D6NtqS3g8hhGi1GrZuavP05JNPKsuXL1cyMjKUrVu3Kk899ZSiVquV+fPnK4qiKC+//LLi7++vzJ49W9m2bZtyzTXXKJGRkUpxcXEjR35uTp69RVFaxnlOmTJFWbp0qZKenq6sWbNGufLKKxVfX19l//79iqK0jHNUFGf1ea1Wq/zvf/9T0tLSlK+//loxmUzKV199VbVNUzjXkpISZdOmTcqmTZsUQHnzzTeVTZs2eZzZorFnbzldvLNnz1Z0Op3yySefKGlpacq7776raDQaZcWKFU0y3kGDBildunRRlixZoqSnpytffPGFYjQalQ8++KBR4r377rsVf39/ZenSpUpWVlbVo7y8vGqbpnDd1jZeq9WqjBo1SomOjlY2b97sso3ZbG7weFuqurjO77rrLiU6OlpZuHChsnHjRuXiiy9WkpOTFZvN1lindVbqqg21lvejtm20pbwfQgjRWknSoxZuueUWJS4uTtHr9UpoaKgydOjQqoSHojingHv22WeViIgIxWAwKBdddJGybdu2Roy4bpya9GgJ5zlhwgQlMjJS0el0SlRUlDJmzBhlx44dVetbwjke9+uvvypdu3ZVDAaD0rFjR+WTTz5xWd8UznXJkiUKUO0xadIkt9s3dtKjNvFOmzZNSUxMVIxGo5KcnKzMnTu3ycablZWl3HTTTUpUVJRiNBqVpKQk5Y033lAcDkejxOsuVkD54osvqrZpCtdtbePNyMjwuM2SJUsaJeaWqC6u84qKCuW+++5TgoKCFC8vL+XKK69UDh482EhndPbqqg21lvejtm20pbwfQgjRWqkURVHqps+IEEIIIYQQQgghRNMhNT2EEEIIIYQQQgjRIknSQwghhBBCCCGEEC2SJD2EEEIIIYQQQgjRIknSQwghhBBCCCGEEC2SJD2EEEIIIYQQQgjRIknSQwghhBBCCCGEEC2SJD2EEEIIIYQQQgjRIknSowVaunQpKpWKwsLCxg6lWdm/fz8qlYrNmzef8WsXL15Mx44dcTgcdR/Yabz33nuMGjWqwY8raibt8PRuuukmRo8efVavveiii/jmm2/qNqBaMJvNxMbGkpKS0uDHbu0GDx7Mgw8+WPU8Pj6et99+u9Hiac6ee+45evTocVavveGGG3jxxRfrNqBaOu+885g9e3ajHFsIIUTzJUmPJiY7O5vJkyfTtm1bDAYDMTExjBw5kkWLFjV2aHXiXBILdelcfmy589hjj/Hvf/8btbrhm9Ttt9/O+vXrWblyZYMfu6VqLe3w+MPX15cuXbpw7733kpaWVm/Hq6t2/9tvv5Gdnc3EiRPrZH9nwmAw8Mgjj/D44483+LFbuptuusnlujz+2Lt3r9vt169fzx133NHAUdbeuSQW6pJKpWLu3Ll1sq+tW7fy+++/M3ny5DrZ35l65plneOKJJxrlBoMQQojmS5IeTcj+/fvp3bs3ixcv5tVXX2Xbtm3MmzePIUOGcO+99zZ2eMKDVatWkZaWxrhx4xrl+AaDgWuvvZZ33323UY7f0rSmdrhw4UKysrLYsmULL774IqmpqSQnJzf55M7UqVO5+eabGyXJCHDdddexYsUKUlNTG+X4Ldmll15KVlaWyyMhIcHttqGhoZhMpgaOsDqLxdLYITSY9957j3HjxuHr69sox7/iiisoKirir7/+apTjCyGEaJ4k6dGE3HPPPahUKtatW8fYsWPp0KEDXbp04eGHH2bNmjWA+zumhYWFqFQqli5d6na/06dPJyAggN9++42kpCRMJhNjx46lrKyMGTNmEB8fT2BgIJMnT8Zut1e9zmKx8Nhjj9GmTRu8vb3p27evyzGO7/evv/6iU6dO+Pj4VH1hPVuKovDqq6/Stm1bvLy8SE5O5scff6xaf3zIwKJFi+jTpw8mk4kBAwawe/dul/288MILhIWF4evry2233cYTTzxRdcftueeeY8aMGfz8889VdxJPPq/09HSGDBmCyWQiOTmZ1atX1xjzrFmzGD58OEajsWrZ8Tt8n3/+ObGxsfj4+HD33Xdjt9t59dVXiYiIICwsjP/9738u+1KpVHz88cdceeWVmEwmOnXqxOrVq9m7dy+DBw/G29ub/v37s2/fPpfXjRo1irlz51JRUXEmb7dwozW1w+DgYCIiImjbti1XXXUVCxcupG/fvtx6660uMfz666/07t0bo9FI27Ztef7557HZbFXrVSoVH374IZdddhleXl4kJCTwww8/VK0//qO1Z8+eqFQqBg8e7BLH66+/TmRkJMHBwdx7771YrVaPMefm5rJw4cJqQ7rOpu2cbTsNDg5mwIABfPvtt6d9j8WZMRgMREREuDw0Go3bbU8d3nK66/B4u501axYDBgzAaDTSpUuXam12586dXH755fj4+BAeHs4NN9xAbm5u1frBgwdz33338fDDDxMSEsIll1xyVud65MgRJkyYQGBgIMHBwVx11VXs37+/av3xHok1tY+srCyuuOKKqvP95ptvXN6X+Ph4AK6++mpUKlXV8+O+/PJL4uPj8ff3Z+LEiZSUlHiM1+Fw8MMPP1Rre/Hx8bzwwgvceOON+Pj4EBcXx88//8yxY8e46qqr8PHxoVu3bmzYsKHqNWf7eajRaLj88sul7QkhhDgzimgS8vLyFJVKpbz44os1bpeRkaEAyqZNm6qWFRQUKICyZMkSRVEUZcmSJQqgFBQUKIqiKF988YWi0+mUSy65RNm4caOybNkyJTg4WBk+fLgyfvx4ZceOHcqvv/6q6PV6ZdasWVX7vfbaa5UBAwYoy5cvV/bu3au89tprisFgUPbs2eOy32HDhinr169XUlJSlE6dOinXXnvtGcV/sqeeekrp2LGjMm/ePGXfvn3KF198oRgMBmXp0qUu59a3b19l6dKlyo4dO5QLL7xQGTBgQNU+vvrqK8VoNCqff/65snv3buX5559X/Pz8lOTkZEVRFKWkpEQZP368cumllypZWVlKVlaWYjabq2Lr2LGj8ttvvym7d+9Wxo4dq8TFxSlWq9XjOSUnJysvv/yyy7Jnn31W8fHxUcaOHavs2LFD+eWXXxS9Xq+MGDFCmTx5srJr1y7l888/VwBl9erVVa8DlDZt2ijfffedsnv3bmX06NFKfHy8cvHFFyvz5s1Tdu7cqfTr10+59NJLXY5XWlqqqFSqqvdJnB1ph4oyZ84cBVDWrl2rKIqizJs3T/Hz81OmT5+u7Nu3T5k/f74SHx+vPPfcc1WvAZTg4GDl008/VXbv3q08/fTTikajUXbu3KkoiqKsW7dOAZSFCxcqWVlZSl5enqIoijJp0iTFz89Pueuuu5TU1FTl119/VUwmk/LJJ594jH3OnDmKt7e3YrfbXZafTds523aqKIry2GOPKYMHD/YYpzhzkyZNUq666iqP6wcNGqQ88MADVc/j4uKUt956q+r56a7D49cMNfyKAAAMmElEQVR9dHS08uOPPyo7d+5UbrvtNsXX11fJzc1VFEVRMjMzlZCQEOXJJ59UUlNTlY0bNyqXXHKJMmTIEJc4fHx8lEcffVTZtWuXkpqa6jbeZ599turvzqnKysqU9u3bK7fccouydetWZefOncq1116rJCUlKWazuer9OF37GDZsmNKjRw9lzZo1SkpKijJo0CDFy8ur6n3JyclRAOWLL75QsrKylJycnKrYfHx8lDFjxijbtm1Tli9frkRERChPPfWUx/d/06ZNCqBkZ2e7LI+Li1OCgoKUjz76SNmzZ49y9913K76+vsqll16qfP/991XtsVOnTorD4VAU5ew/DxVFUT744AMlPj7eY5xCCCHEqSTp0USsXbtWAZTZs2fXuN3Z/tgClL1791a95s4771RMJpNSUlJStWzEiBHKnXfeqSiKouzdu1dRqVTKkSNHXI4/dOhQ5cknn/S43/fff18JDw8/o/iPKy0tVYxGo7Jq1SqX5bfeeqtyzTXXuJzbwoULq9b//vvvCqBUVFQoiqIoffv2Ve69916XfVxwwQUuXz7dfbk+Httnn31WtWzHjh0K4PFLraIoir+/vzJz5kyXZc8++6xiMpmU4uLiqmUjRoxQ4uPjXX6sJSUlKS+99FLVc0B5+umnq56vXr1aAZRp06ZVLfv2228Vo9FYLY7AwEBl+vTpHuMUpyftUFFSU1MVQPnuu+8URVGUCy+8sFoS6Msvv1QiIyOrngPKXXfd5bJN3759lbvvvrvG402aNEmJi4tTbDZb1bJx48YpEyZM8Bj7W2+9pbRt27ba8rNpO2fbThVFUd555x354VXHJk2apGg0GsXb27vqMXbs2Kr1tUl61OY6PDlJbbValejoaOWVV15RFEVRnnnmGWX48OEu+zh06JACKLt3766Ko0ePHqc9n5qSHtOmTVOSkpKqkgCKoihms1nx8vJS/vrrr6r3o6b2cbytrl+/vmp9WlqaAlR7X+bMmVMttlOv/UcffVTp27evx/OZM2eOotFoXGJWFOe/w/XXX1/1PCsrSwGUZ555pmrZ8faYlZWlKMrZfR4e9/PPPytqtbpa4lMIIYTwRFuHnUbEOVAUBXB2z60PJpOJdu3aVT0PDw8nPj4eHx8fl2U5OTkAbNy4EUVR6NChg8t+zGYzwcHBHvcbGRlZtY8ztXPnTiorK6t1FbZYLPTs2dNlWffu3V2OCZCTk0NsbCy7d+/mnnvucdn+/PPPZ/HixbWKw9O+O3bs6Hb7iooKl6Etx8XHx7uMew4PD0ej0bjUITj5PXd3/PDwcAC6devmsqyyspLi4mL8/Pyqlnt5eVFeXl6rcxTuSTus/h6kpKSwfv16lyEedrudyspKysvLq2oq9O/f32U//fv3r1Xh0i5durgMX4iMjGTbtm0et/fU3uDs2s7ZtlNpb/VjyJAhfPjhh1XPvb29z+j1tbkOT95Gq9XSp0+fqvosKSkpLFmyxKVNHrdv376qttinT58ziutUKSkp7N27t1ptjMrKSpchWDW1j927d6PVaunVq1fV+sTERAIDA2sVw6nX/uk+NyoqKjAYDG4/H2vT9sD5tzQiIgI488/D47y8vHA4HJjNZry8vGp1rkIIIVo3SXo0Ee3bt0elUpGamlrjrCLHv4gf/2EC1Dj+/TidTufyXKVSuV12vCK6w+FAo9GQkpJSbTz1yV9I3O3j5NjOxPFj//7777Rp08ZlncFgcHl+8nGPfwE7uZr7qV/KziSm0+37VCEhIRQUFNS4n+P7quk9r+n4tYkpPz+f0NBQj3GK05N2SNWPv+N1OBwOB88//zxjxoyptq2n5MPJcZxObdrEyTy1t1P3Vdu2c7btVNpb/fD29iYxMbFO91mb6/Dka2PkyJG88sor1bY5ngSHM0/GnMrhcNC7d2++/vrrautOvq5quhY9tfHatv2zaXvl5eVYLBb0er3HfTVE2zOZTJLwEEIIUWtSyLSJCAoKYsSIEbz//vuUlZVVW19YWAic+DJ0cpHC+pj+tWfPntjtdnJyckhMTHR5HL9LU9c6d+6MwWDg4MGD1Y4ZExNT6/0kJSWxbt06l2UnF1AD0Ov1LsXRzkXPnj3ZuXNnnezrbO3bt4/KyspqPWLEmWnt7dDhcDB16lQSEhKqrqVevXqxe/fuasdPTEx06Q1xvMjryc+P9446/gOpLtpcz549yc7O9pj4aCjbt2+X9tYE1XQdutvGZrORkpJStU2vXr3YsWMH8fHx1a73c010nKxXr16kpaURFhZW7Tj+/v612kfHjh2x2Wxs2rSpatnevXurPqeO0+l0ddL2jhcDb+y/d9u3b3fp3SKEEEKcjiQ9mpAPPvgAu93O+eefz08//URaWhqpqalMnTq1qjuul5cX/fr14+WXX2bnzp0sX76cp59+us5j6dChA9dddx033ngjs2fPJiMjg/Xr1/PKK6/wxx9/nPP+d+/ezebNm10eBoOBRx55hIceeogZM2awb98+Nm3axPvvv8+MGTNqve/Jkyczbdo0ZsyYQVpaGi+88AJbt251udsXHx/P1q1b2b17N7m5ubW6S+/JiBEjWLly5Vm/vi6sWLGCtm3bunQVFmenNbXDvLw8srOzSU9P55dffmHYsGGsW7eOadOmVfUs+c9//sPMmTN57rnn2LFjB6mpqXz33XfVzveHH37g888/Z8+ePTz77LOsW7eO++67D4CwsDC8vLyYN28eR48epaio6Kxj7tmzJ6Ghofz9999nf+J1YMWKFQwfPrxRYxDV1XQdHvf+++8zZ84cdu3axb333ktBQQG33HILAPfeey/5+flcc801rFu3jvT0dObPn88tt9xyVomDioqKan/r9u7dy3XXXUdISAhXXXUVK1asICMjg2XLlvHAAw9w+PDhWu27Y8eODBs2jDvuuIN169axadMm7rjjDry8vKr9vVu0aNE5JwtDQ0Pp1atXk/h7J21PCCHEmZCkRxOSkJDAxo0bGTJkCFOmTKFr165ccsklLFq0yGWM8+eff47VaqVPnz488MADvPDCC/USzxdffMGNN97IlClTSEpKYtSoUaxdu/aMel14MnHiRHr27OnyyMzM5P/+7//4z3/+w0svvUSnTp0YMWIEv/76a1VX+9q47rrrePLJJ3nkkUfo1asXGRkZ3HTTTS5d8W+//XaSkpLo06fPOf+Auv7669m5c2e1aXMb0rfffsvtt9/eaMdvSVpTOxw2bBiRkZF069aNJ554gk6dOrF161aGDBlStc2IESP47bffWLBgAeeddx79+vXjzTffJC4uzmVfzz//PLNmzaJ79+7MmDGDr7/+ms6dOwPOuglTp07l448/JioqiquuuuqsY9ZoNNxyyy1uhwU0lNWrV1NUVMTYsWMbLQbhXk3X4XEvv/wyr7zyCsnJyaxYsYKff/6ZkJAQAKKiovj777+x2+2MGDGCrl278sADD+Dv7+/Ss6m29uzZU+1v3W233YbJZGL58uXExsYyZswYOnXqxC233EJFRYVLrabTmTlzJuHh4Vx00UVcffXV3H777fj6+rr8vXvjjTdYsGABMTEx59w76Y477mjUtnfkyBFWrVrFzTff3GgxCCGEaH5UytkO/BaiGbnkkkuIiIjgyy+/rJf9P/bYYxQVFfHxxx/Xy/5rsn37doYOHcqePXtq3S1aiLqkUqmYM2dOjXVQ6tLRo0fp0qULKSkp1ZIvDWHcuHH07NmTp556qsGPLTw73XW4f/9+EhIS2LRpU9VQjZbm8OHDxMTEsHDhQoYOHVrn+6+srCQpKYlZs2ZVKxrbEB599FGKior45JNPGvzYQgghmi8pZCpanPLycj766CNGjBiBRqPh22+/ZeHChSxYsKDejvnvf/+b999/H7vdXq3gZH3LzMxk5syZkvAQrUZ4eDjTpk3j4MGDDZ70MJvNJCcn89BDDzXocYVwZ/HixZSWltKtWzeysrJ47LHHiI+P56KLLqqX4xmNRmbOnElubm697P90wsLCeOSRRxrl2EIIIZov6ekhWpyKigpGjhzJxo0bMZvNJCUl8fTTT7udfUIIce4auqeHEO60xp4ef/31F1OmTCE9PR1fX18GDBjA22+/3Sg9oIQQQoimSpIeQgghhBBCCCGEaJGkkKkQQgghhBBCCCFaJEl6CCGEEEIIIYQQokWSpIcQQgghhBBCCCFaJEl6CCGEEEIIIYQQokWSpIcQQgghhBBCCCFaJEl6CCGEEEIIIYQQokWSpIcQQgghhBBCCCFaJEl6CCGEEEIIIYQQokWSpIcQQgghhBBCCCFapP8Ho0XwzuZbXGoAAAAASUVORK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import seaborn as sns\n", + "\n", + "pairplot_figure = sns.pairplot(samples, hue=\"Species\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "OK so if we get to choose, we could definitely say that in this dataset, the `Flipper Length` in combination with the `Culmen Depth` leads to the hardest classification task for our machine learning model.\n", + "\n", + "Therefore, here is the plan:\n", + "- we select only those to numerical features (_iow_ we will get rid of the `Culmen Lenght` feature)\n", + "- we will apply an identical _Model evaluation_ pipeline as we did in our previous example\n", + " - Cross Validation + Evaluation on Test set\n", + "\n", + "The very difference this time is that we will use multiple metrics to evaluate our model to prove our point on _carefully selecting evaluation metrics_." + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.010659Z", + "iopub.status.busy": "2022-12-02T12:08:41.010598Z", + "iopub.status.idle": "2022-12-02T12:08:41.012415Z", + "shell.execute_reply": "2022-12-02T12:08:41.012185Z" + } + }, + "outputs": [], + "source": [ + "num_features = [\"Culmen Length (mm)\", \"Culmen Depth (mm)\", \"Flipper Length (mm)\"]\n", + "selected_num_features = num_features[1:]\n", + "cat_features = [\"Sex\"]\n", + "features = selected_num_features + cat_features" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.013669Z", + "iopub.status.busy": "2022-12-02T12:08:41.013615Z", + "iopub.status.idle": "2022-12-02T12:08:41.015540Z", + "shell.execute_reply": "2022-12-02T12:08:41.015315Z" + } + }, + "outputs": [], + "source": [ + "num_transformer = StandardScaler()\n", + "cat_transformer = OneHotEncoder(handle_unknown='ignore')\n", + "\n", + "preprocessor = ColumnTransformer(transformers=[\n", + " ('num', num_transformer, selected_num_features), # note here, we will only preprocess selected numerical features\n", + " ('cat', cat_transformer, cat_features)\n", + "])\n", + "\n", + "model = Pipeline(steps=[\n", + " ('preprocessor', preprocessor),\n", + " ('classifier', SVC()),\n", + "])" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.016889Z", + "iopub.status.busy": "2022-12-02T12:08:41.016837Z", + "iopub.status.idle": "2022-12-02T12:08:41.019785Z", + "shell.execute_reply": "2022-12-02T12:08:41.019564Z" + } + }, + "outputs": [], + "source": [ + "X, y = samples[features], samples[target[0]]\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=.7, random_state=42, stratify=y) # we also stratify on classes" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.021171Z", + "iopub.status.busy": "2022-12-02T12:08:41.021110Z", + "iopub.status.idle": "2022-12-02T12:08:41.024373Z", + "shell.execute_reply": "2022-12-02T12:08:41.024163Z" + } + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
index
Species
Adelie Penguin (Pygoscelis adeliae)102
Chinstrap penguin (Pygoscelis antarctica)47
\n", + "
" + ], + "text/plain": [ + " index\n", + "Species \n", + "Adelie Penguin (Pygoscelis adeliae) 102\n", + "Chinstrap penguin (Pygoscelis antarctica) 47" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "y_train.reset_index().groupby(\"Species\").count()" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.025770Z", + "iopub.status.busy": "2022-12-02T12:08:41.025711Z", + "iopub.status.idle": "2022-12-02T12:08:41.028799Z", + "shell.execute_reply": "2022-12-02T12:08:41.028592Z" + } + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
index
Species
Adelie Penguin (Pygoscelis adeliae)44
Chinstrap penguin (Pygoscelis antarctica)21
\n", + "
" + ], + "text/plain": [ + " index\n", + "Species \n", + "Adelie Penguin (Pygoscelis adeliae) 44\n", + "Chinstrap penguin (Pygoscelis antarctica) 21" + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "y_test.reset_index().groupby(\"Species\").count()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "In our evaluation pipeline we will be using keep record both **accuracy** (`ACC`) and **matthew correlation coefficient** (`MCC`)" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.030139Z", + "iopub.status.busy": "2022-12-02T12:08:41.030084Z", + "iopub.status.idle": "2022-12-02T12:08:41.051172Z", + "shell.execute_reply": "2022-12-02T12:08:41.050937Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "{'fit_time': array([0.00228 , 0.00207305, 0.00184989, 0.00178504, 0.00176287]),\n", + " 'score_time': array([0.00179315, 0.00131392, 0.0012598 , 0.0012362 , 0.00133991]),\n", + " 'test_MCC': array([0.37796447, 0.27863911, 0.40824829, 0.02424643, 0.08625819]),\n", + " 'test_ACC': array([0.73333333, 0.7 , 0.76666667, 0.66666667, 0.62068966])}" + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from sklearn.model_selection import cross_validate\n", + "from sklearn.metrics import make_scorer\n", + "from sklearn.metrics import matthews_corrcoef as mcc\n", + "from sklearn.metrics import accuracy_score as acc\n", + "\n", + "mcc_scorer = make_scorer(mcc)\n", + "acc_scorer = make_scorer(acc)\n", + "scores = cross_validate(model, X_train, y_train, cv=5,\n", + " scoring={\"MCC\": mcc_scorer, \"ACC\": acc_scorer})\n", + "scores" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.052554Z", + "iopub.status.busy": "2022-12-02T12:08:41.052497Z", + "iopub.status.idle": "2022-12-02T12:08:41.054435Z", + "shell.execute_reply": "2022-12-02T12:08:41.054214Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Avg ACC in CV: 0.697471264367816\n", + "Avg MCC in CV: 0.2350712993854009\n" + ] + } + ], + "source": [ + "import numpy as np\n", + "\n", + "print(\"Avg ACC in CV: \", np.average(scores[\"test_ACC\"]))\n", + "print(\"Avg MCC in CV: \", np.average(scores[\"test_MCC\"]))" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.055631Z", + "iopub.status.busy": "2022-12-02T12:08:41.055577Z", + "iopub.status.idle": "2022-12-02T12:08:41.063042Z", + "shell.execute_reply": "2022-12-02T12:08:41.062811Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "ACC: 0.7230769230769231\n", + "MCC: 0.29439815585406465\n" + ] + } + ], + "source": [ + "model = model.fit(X_train, y_train)\n", + "\n", + "print(\"ACC: \", acc_scorer(model, X_test, y_test))\n", + "print(\"MCC: \", mcc_scorer(model, X_test, y_test))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To see exactly what happened, let's have a look at the **Confusion matrix**" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.064432Z", + "iopub.status.busy": "2022-12-02T12:08:41.064358Z", + "iopub.status.idle": "2022-12-02T12:08:41.170940Z", + "shell.execute_reply": "2022-12-02T12:08:41.170658Z" + } + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from sklearn.metrics import ConfusionMatrixDisplay\n", + "fig, ax = plt.subplots(figsize=(15, 10))\n", + "ConfusionMatrixDisplay.from_estimator(model, X_test, y_test, ax=ax)\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As expected, the model did a pretty bad job in classifying *Chinstrap Penguins* and the `MCC` was able to catch that, whilst `ACC` could not as it only considers correctly classified samples!" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Time-series Validation\n", + "\n", + "But validation can get tricky if time gets involved.\n", + "\n", + "Imagine we measured the growth of baby penguin Hank over time and wanted to us machine learning to project the development of Hank. Then our data suddenly isn't i.i.d. anymore, since it is dependent in the time dimension.\n", + "\n", + "Were we to split our data randomly for our training and test set, we would test on data points that lie in between training points, where even a simple linear interpolation can do a fairly decent job.\n", + "\n", + "Therefor, we need to split our measurements along the time axis\n", + "![Scikit-learn time series validation](https://scikit-learn.org/stable/_images/sphx_glr_plot_cv_indices_013.png)\n", + "*Scikit-learn Time Series CV [[Source]](https://scikit-learn.org/stable/modules/cross_validation.html#time-series-split).*" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": { + "execution": { + "iopub.execute_input": "2022-12-02T12:08:41.172650Z", + "iopub.status.busy": "2022-12-02T12:08:41.172561Z", + "iopub.status.idle": "2022-12-02T12:08:41.175345Z", + "shell.execute_reply": "2022-12-02T12:08:41.175066Z" + }, + "lines_to_next_cell": 2 + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "TimeSeriesSplit(gap=0, max_train_size=None, n_splits=3, test_size=None)\n", + "[0 1 2] [3]\n", + "[0 1 2 3] [4]\n", + "[0 1 2 3 4] [5]\n" + ] + } + ], + "source": [ + "import numpy as np\n", + "from sklearn.model_selection import TimeSeriesSplit\n", + "\n", + "X = np.array([[1, 2], [3, 4], [1, 2], [3, 4], [1, 2], [3, 4]])\n", + "y = np.array([1, 2, 3, 4, 5, 6])\n", + "tscv = TimeSeriesSplit(n_splits=3)\n", + "print(tscv)\n", + "\n", + "for train, test in tscv.split(X):\n", + " print(\"%s %s\" % (train, test))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Spatial Validation\n", + "\n", + "Spatial data, like maps and satellite data has a similar problem.\n", + "\n", + "Here the data is correlated in the spatial dimension. However, we can mitigate the effect by supplying a group. In this simple example I used continents, but it's possible to group by bins on a lat-lon grid as well. \n", + "\n", + "Here especially, a cross-validation scheme is very important, as it is used to validate against every area on your map at least once." + ] + }, + { + "cell_type": "code", + "execution_count": 28, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:52.540859Z", - "iopub.status.busy": "2022-12-01T10:51:52.540786Z", - "iopub.status.idle": "2022-12-01T10:51:52.543451Z", - "shell.execute_reply": "2022-12-01T10:51:52.543176Z" + "iopub.execute_input": "2022-12-02T12:08:41.176696Z", + "iopub.status.busy": "2022-12-02T12:08:41.176622Z", + "iopub.status.idle": "2022-12-02T12:08:41.179172Z", + "shell.execute_reply": "2022-12-02T12:08:41.178921Z" } }, "outputs": [ @@ -734,7 +1404,7 @@ "formats": "notebooks//ipynb,python_scripts//py:percent" }, "kernelspec": { - "display_name": "Python 3.10.8 ('pydata-global-2022-ml-repro')", + "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, @@ -757,5 +1427,5 @@ } }, "nbformat": 4, - "nbformat_minor": 2 + "nbformat_minor": 4 } diff --git a/notebooks/2 - Benchmarking.ipynb b/notebooks/2 - Benchmarking.ipynb index 0e889a2..214ab8d 100644 --- a/notebooks/2 - Benchmarking.ipynb +++ b/notebooks/2 - Benchmarking.ipynb @@ -17,10 +17,10 @@ "id": "54158e1d", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:53.752314Z", - "iopub.status.busy": "2022-12-01T10:51:53.751830Z", - "iopub.status.idle": "2022-12-01T10:51:53.763582Z", - "shell.execute_reply": "2022-12-01T10:51:53.762821Z" + "iopub.execute_input": "2022-12-02T12:08:42.300568Z", + "iopub.status.busy": "2022-12-02T12:08:42.300260Z", + "iopub.status.idle": "2022-12-02T12:08:42.308258Z", + "shell.execute_reply": "2022-12-02T12:08:42.307764Z" } }, "outputs": [], @@ -36,10 +36,10 @@ "execution_count": 2, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:53.769207Z", - "iopub.status.busy": "2022-12-01T10:51:53.769019Z", - "iopub.status.idle": "2022-12-01T10:51:53.950383Z", - "shell.execute_reply": "2022-12-01T10:51:53.949593Z" + "iopub.execute_input": "2022-12-02T12:08:42.310848Z", + "iopub.status.busy": "2022-12-02T12:08:42.310696Z", + "iopub.status.idle": "2022-12-02T12:08:42.477417Z", + "shell.execute_reply": "2022-12-02T12:08:42.477157Z" } }, "outputs": [ @@ -148,10 +148,10 @@ "execution_count": 3, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:53.953198Z", - "iopub.status.busy": "2022-12-01T10:51:53.953026Z", - "iopub.status.idle": "2022-12-01T10:51:54.201640Z", - "shell.execute_reply": "2022-12-01T10:51:54.201365Z" + "iopub.execute_input": "2022-12-02T12:08:42.479128Z", + "iopub.status.busy": "2022-12-02T12:08:42.479032Z", + "iopub.status.idle": "2022-12-02T12:08:42.678687Z", + "shell.execute_reply": "2022-12-02T12:08:42.678362Z" } }, "outputs": [ @@ -315,10 +315,10 @@ "execution_count": 4, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:54.203507Z", - "iopub.status.busy": "2022-12-01T10:51:54.203401Z", - "iopub.status.idle": "2022-12-01T10:51:54.207548Z", - "shell.execute_reply": "2022-12-01T10:51:54.207330Z" + "iopub.execute_input": "2022-12-02T12:08:42.680401Z", + "iopub.status.busy": "2022-12-02T12:08:42.680316Z", + "iopub.status.idle": "2022-12-02T12:08:42.683920Z", + "shell.execute_reply": "2022-12-02T12:08:42.683655Z" } }, "outputs": [ @@ -396,10 +396,10 @@ "execution_count": 5, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:54.209424Z", - "iopub.status.busy": "2022-12-01T10:51:54.209042Z", - "iopub.status.idle": "2022-12-01T10:51:54.213292Z", - "shell.execute_reply": "2022-12-01T10:51:54.213002Z" + "iopub.execute_input": "2022-12-02T12:08:42.685291Z", + "iopub.status.busy": "2022-12-02T12:08:42.685232Z", + "iopub.status.idle": "2022-12-02T12:08:42.688440Z", + "shell.execute_reply": "2022-12-02T12:08:42.688221Z" } }, "outputs": [ diff --git a/notebooks/3 - Model Sharing.ipynb b/notebooks/3 - Model Sharing.ipynb index 8721593..7607c66 100644 --- a/notebooks/3 - Model Sharing.ipynb +++ b/notebooks/3 - Model Sharing.ipynb @@ -35,10 +35,10 @@ "id": "54158e1d", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:55.405428Z", - "iopub.status.busy": "2022-12-01T10:51:55.405062Z", - "iopub.status.idle": "2022-12-01T10:51:55.419937Z", - "shell.execute_reply": "2022-12-01T10:51:55.419492Z" + "iopub.execute_input": "2022-12-02T12:08:43.673874Z", + "iopub.status.busy": "2022-12-02T12:08:43.673366Z", + "iopub.status.idle": "2022-12-02T12:08:43.682601Z", + "shell.execute_reply": "2022-12-02T12:08:43.682091Z" } }, "outputs": [], @@ -54,10 +54,10 @@ "execution_count": 2, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:55.422779Z", - "iopub.status.busy": "2022-12-01T10:51:55.422637Z", - "iopub.status.idle": "2022-12-01T10:51:55.710708Z", - "shell.execute_reply": "2022-12-01T10:51:55.710312Z" + "iopub.execute_input": "2022-12-02T12:08:43.685392Z", + "iopub.status.busy": "2022-12-02T12:08:43.685236Z", + "iopub.status.idle": "2022-12-02T12:08:43.864460Z", + "shell.execute_reply": "2022-12-02T12:08:43.864224Z" } }, "outputs": [ @@ -166,10 +166,10 @@ "execution_count": 3, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:55.712796Z", - "iopub.status.busy": "2022-12-01T10:51:55.712702Z", - "iopub.status.idle": "2022-12-01T10:51:55.935864Z", - "shell.execute_reply": "2022-12-01T10:51:55.935362Z" + "iopub.execute_input": "2022-12-02T12:08:43.865989Z", + "iopub.status.busy": "2022-12-02T12:08:43.865895Z", + "iopub.status.idle": "2022-12-02T12:08:44.058352Z", + "shell.execute_reply": "2022-12-02T12:08:44.058058Z" } }, "outputs": [], @@ -188,10 +188,10 @@ "execution_count": 4, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:55.938373Z", - "iopub.status.busy": "2022-12-01T10:51:55.938232Z", - "iopub.status.idle": "2022-12-01T10:51:55.958495Z", - "shell.execute_reply": "2022-12-01T10:51:55.958211Z" + "iopub.execute_input": "2022-12-02T12:08:44.060021Z", + "iopub.status.busy": "2022-12-02T12:08:44.059955Z", + "iopub.status.idle": "2022-12-02T12:08:44.078070Z", + "shell.execute_reply": "2022-12-02T12:08:44.077806Z" } }, "outputs": [ @@ -234,10 +234,10 @@ "execution_count": 5, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:55.960089Z", - "iopub.status.busy": "2022-12-01T10:51:55.960002Z", - "iopub.status.idle": "2022-12-01T10:51:55.962040Z", - "shell.execute_reply": "2022-12-01T10:51:55.961724Z" + "iopub.execute_input": "2022-12-02T12:08:44.079627Z", + "iopub.status.busy": "2022-12-02T12:08:44.079563Z", + "iopub.status.idle": "2022-12-02T12:08:44.081436Z", + "shell.execute_reply": "2022-12-02T12:08:44.081146Z" } }, "outputs": [], @@ -253,10 +253,10 @@ "execution_count": 6, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:55.963654Z", - "iopub.status.busy": "2022-12-01T10:51:55.963567Z", - "iopub.status.idle": "2022-12-01T10:51:55.970837Z", - "shell.execute_reply": "2022-12-01T10:51:55.970550Z" + "iopub.execute_input": "2022-12-02T12:08:44.083437Z", + "iopub.status.busy": "2022-12-02T12:08:44.083326Z", + "iopub.status.idle": "2022-12-02T12:08:44.090630Z", + "shell.execute_reply": "2022-12-02T12:08:44.090209Z" } }, "outputs": [ @@ -300,10 +300,10 @@ "execution_count": 7, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:55.972757Z", - "iopub.status.busy": "2022-12-01T10:51:55.972653Z", - "iopub.status.idle": "2022-12-01T10:51:55.997759Z", - "shell.execute_reply": "2022-12-01T10:51:55.997500Z" + "iopub.execute_input": "2022-12-02T12:08:44.092348Z", + "iopub.status.busy": "2022-12-02T12:08:44.092252Z", + "iopub.status.idle": "2022-12-02T12:08:44.115936Z", + "shell.execute_reply": "2022-12-02T12:08:44.115392Z" } }, "outputs": [ @@ -358,10 +358,10 @@ "execution_count": 8, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:56.000844Z", - "iopub.status.busy": "2022-12-01T10:51:56.000695Z", - "iopub.status.idle": "2022-12-01T10:51:56.004812Z", - "shell.execute_reply": "2022-12-01T10:51:56.004575Z" + "iopub.execute_input": "2022-12-02T12:08:44.118328Z", + "iopub.status.busy": "2022-12-02T12:08:44.118160Z", + "iopub.status.idle": "2022-12-02T12:08:44.122130Z", + "shell.execute_reply": "2022-12-02T12:08:44.121683Z" } }, "outputs": [ diff --git a/notebooks/4 - Testing.ipynb b/notebooks/4 - Testing.ipynb index 3cdf5a9..8c681b7 100644 --- a/notebooks/4 - Testing.ipynb +++ b/notebooks/4 - Testing.ipynb @@ -21,10 +21,10 @@ "id": "54158e1d", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:57.236673Z", - "iopub.status.busy": "2022-12-01T10:51:57.236449Z", - "iopub.status.idle": "2022-12-01T10:51:57.242126Z", - "shell.execute_reply": "2022-12-01T10:51:57.241712Z" + "iopub.execute_input": "2022-12-02T12:08:45.363850Z", + "iopub.status.busy": "2022-12-02T12:08:45.363544Z", + "iopub.status.idle": "2022-12-02T12:08:45.377742Z", + "shell.execute_reply": "2022-12-02T12:08:45.376884Z" } }, "outputs": [], @@ -40,10 +40,10 @@ "execution_count": 2, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:57.244324Z", - "iopub.status.busy": "2022-12-01T10:51:57.244181Z", - "iopub.status.idle": "2022-12-01T10:51:57.448762Z", - "shell.execute_reply": "2022-12-01T10:51:57.448280Z" + "iopub.execute_input": "2022-12-02T12:08:45.382636Z", + "iopub.status.busy": "2022-12-02T12:08:45.382401Z", + "iopub.status.idle": "2022-12-02T12:08:45.558824Z", + "shell.execute_reply": "2022-12-02T12:08:45.558514Z" } }, "outputs": [ @@ -152,10 +152,10 @@ "execution_count": 3, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:57.450887Z", - "iopub.status.busy": "2022-12-01T10:51:57.450737Z", - "iopub.status.idle": "2022-12-01T10:51:57.694015Z", - "shell.execute_reply": "2022-12-01T10:51:57.693724Z" + "iopub.execute_input": "2022-12-02T12:08:45.560369Z", + "iopub.status.busy": "2022-12-02T12:08:45.560286Z", + "iopub.status.idle": "2022-12-02T12:08:45.762886Z", + "shell.execute_reply": "2022-12-02T12:08:45.762634Z" } }, "outputs": [], @@ -174,10 +174,10 @@ "execution_count": 4, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:57.695897Z", - "iopub.status.busy": "2022-12-01T10:51:57.695813Z", - "iopub.status.idle": "2022-12-01T10:51:57.720810Z", - "shell.execute_reply": "2022-12-01T10:51:57.719909Z" + "iopub.execute_input": "2022-12-02T12:08:45.764514Z", + "iopub.status.busy": "2022-12-02T12:08:45.764455Z", + "iopub.status.idle": "2022-12-02T12:08:45.781098Z", + "shell.execute_reply": "2022-12-02T12:08:45.780813Z" }, "lines_to_next_cell": 1 }, @@ -236,10 +236,10 @@ "execution_count": 5, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:57.722622Z", - "iopub.status.busy": "2022-12-01T10:51:57.722522Z", - "iopub.status.idle": "2022-12-01T10:51:57.725249Z", - "shell.execute_reply": "2022-12-01T10:51:57.724926Z" + "iopub.execute_input": "2022-12-02T12:08:45.782688Z", + "iopub.status.busy": "2022-12-02T12:08:45.782605Z", + "iopub.status.idle": "2022-12-02T12:08:45.784656Z", + "shell.execute_reply": "2022-12-02T12:08:45.784383Z" } }, "outputs": [], @@ -263,10 +263,10 @@ "execution_count": 6, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:57.727674Z", - "iopub.status.busy": "2022-12-01T10:51:57.727553Z", - "iopub.status.idle": "2022-12-01T10:51:57.731718Z", - "shell.execute_reply": "2022-12-01T10:51:57.731357Z" + "iopub.execute_input": "2022-12-02T12:08:45.786086Z", + "iopub.status.busy": "2022-12-02T12:08:45.786003Z", + "iopub.status.idle": "2022-12-02T12:08:45.789441Z", + "shell.execute_reply": "2022-12-02T12:08:45.789175Z" } }, "outputs": [], @@ -295,10 +295,10 @@ "execution_count": 7, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:57.734032Z", - "iopub.status.busy": "2022-12-01T10:51:57.733871Z", - "iopub.status.idle": "2022-12-01T10:51:58.047395Z", - "shell.execute_reply": "2022-12-01T10:51:58.046505Z" + "iopub.execute_input": "2022-12-02T12:08:45.790980Z", + "iopub.status.busy": "2022-12-02T12:08:45.790918Z", + "iopub.status.idle": "2022-12-02T12:08:45.802374Z", + "shell.execute_reply": "2022-12-02T12:08:45.802134Z" } }, "outputs": [ @@ -346,10 +346,10 @@ "execution_count": 8, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:58.052791Z", - "iopub.status.busy": "2022-12-01T10:51:58.052398Z", - "iopub.status.idle": "2022-12-01T10:51:58.058964Z", - "shell.execute_reply": "2022-12-01T10:51:58.058613Z" + "iopub.execute_input": "2022-12-02T12:08:45.803746Z", + "iopub.status.busy": "2022-12-02T12:08:45.803672Z", + "iopub.status.idle": "2022-12-02T12:08:45.807728Z", + "shell.execute_reply": "2022-12-02T12:08:45.807496Z" } }, "outputs": [ @@ -470,10 +470,10 @@ "execution_count": 9, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:51:58.061091Z", - "iopub.status.busy": "2022-12-01T10:51:58.060999Z", - "iopub.status.idle": "2022-12-01T10:52:00.476160Z", - "shell.execute_reply": "2022-12-01T10:52:00.475839Z" + "iopub.execute_input": "2022-12-02T12:08:45.809108Z", + "iopub.status.busy": "2022-12-02T12:08:45.809045Z", + "iopub.status.idle": "2022-12-02T12:08:45.892379Z", + "shell.execute_reply": "2022-12-02T12:08:45.892113Z" } }, "outputs": [], @@ -486,10 +486,10 @@ "execution_count": 10, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:00.478691Z", - "iopub.status.busy": "2022-12-01T10:52:00.478541Z", - "iopub.status.idle": "2022-12-01T10:52:00.486337Z", - "shell.execute_reply": "2022-12-01T10:52:00.485999Z" + "iopub.execute_input": "2022-12-02T12:08:45.893984Z", + "iopub.status.busy": "2022-12-02T12:08:45.893904Z", + "iopub.status.idle": "2022-12-02T12:08:45.899698Z", + "shell.execute_reply": "2022-12-02T12:08:45.899444Z" } }, "outputs": [ @@ -599,10 +599,10 @@ "execution_count": 11, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:00.488095Z", - "iopub.status.busy": "2022-12-01T10:52:00.488006Z", - "iopub.status.idle": "2022-12-01T10:52:00.494859Z", - "shell.execute_reply": "2022-12-01T10:52:00.494560Z" + "iopub.execute_input": "2022-12-02T12:08:45.901140Z", + "iopub.status.busy": "2022-12-02T12:08:45.901059Z", + "iopub.status.idle": "2022-12-02T12:08:45.906636Z", + "shell.execute_reply": "2022-12-02T12:08:45.906420Z" } }, "outputs": [], @@ -626,10 +626,10 @@ "execution_count": 12, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:00.496328Z", - "iopub.status.busy": "2022-12-01T10:52:00.496235Z", - "iopub.status.idle": "2022-12-01T10:52:00.499279Z", - "shell.execute_reply": "2022-12-01T10:52:00.498904Z" + "iopub.execute_input": "2022-12-02T12:08:45.908008Z", + "iopub.status.busy": "2022-12-02T12:08:45.907924Z", + "iopub.status.idle": "2022-12-02T12:08:45.910110Z", + "shell.execute_reply": "2022-12-02T12:08:45.909893Z" } }, "outputs": [ @@ -653,10 +653,10 @@ "execution_count": 13, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:00.500977Z", - "iopub.status.busy": "2022-12-01T10:52:00.500831Z", - "iopub.status.idle": "2022-12-01T10:52:00.503989Z", - "shell.execute_reply": "2022-12-01T10:52:00.503592Z" + "iopub.execute_input": "2022-12-02T12:08:45.911472Z", + "iopub.status.busy": "2022-12-02T12:08:45.911368Z", + "iopub.status.idle": "2022-12-02T12:08:45.914397Z", + "shell.execute_reply": "2022-12-02T12:08:45.914102Z" } }, "outputs": [ diff --git a/notebooks/5 - Interpretability.ipynb b/notebooks/5 - Interpretability.ipynb index 27aec62..f2b8895 100644 --- a/notebooks/5 - Interpretability.ipynb +++ b/notebooks/5 - Interpretability.ipynb @@ -19,10 +19,10 @@ "id": "54158e1d", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:01.742124Z", - "iopub.status.busy": "2022-12-01T10:52:01.741940Z", - "iopub.status.idle": "2022-12-01T10:52:01.747134Z", - "shell.execute_reply": "2022-12-01T10:52:01.746679Z" + "iopub.execute_input": "2022-12-02T12:08:47.022778Z", + "iopub.status.busy": "2022-12-02T12:08:47.022218Z", + "iopub.status.idle": "2022-12-02T12:08:47.033827Z", + "shell.execute_reply": "2022-12-02T12:08:47.033327Z" } }, "outputs": [], @@ -38,10 +38,10 @@ "execution_count": 2, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:01.749080Z", - "iopub.status.busy": "2022-12-01T10:52:01.748984Z", - "iopub.status.idle": "2022-12-01T10:52:01.920217Z", - "shell.execute_reply": "2022-12-01T10:52:01.919922Z" + "iopub.execute_input": "2022-12-02T12:08:47.036985Z", + "iopub.status.busy": "2022-12-02T12:08:47.036591Z", + "iopub.status.idle": "2022-12-02T12:08:47.230837Z", + "shell.execute_reply": "2022-12-02T12:08:47.230535Z" } }, "outputs": [ @@ -150,10 +150,10 @@ "execution_count": 3, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:01.921840Z", - "iopub.status.busy": "2022-12-01T10:52:01.921714Z", - "iopub.status.idle": "2022-12-01T10:52:02.166470Z", - "shell.execute_reply": "2022-12-01T10:52:02.166172Z" + "iopub.execute_input": "2022-12-02T12:08:47.232552Z", + "iopub.status.busy": "2022-12-02T12:08:47.232299Z", + "iopub.status.idle": "2022-12-02T12:08:47.450333Z", + "shell.execute_reply": "2022-12-02T12:08:47.449952Z" } }, "outputs": [], @@ -172,10 +172,10 @@ "execution_count": 4, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:02.168105Z", - "iopub.status.busy": "2022-12-01T10:52:02.168025Z", - "iopub.status.idle": "2022-12-01T10:52:02.192838Z", - "shell.execute_reply": "2022-12-01T10:52:02.192563Z" + "iopub.execute_input": "2022-12-02T12:08:47.452455Z", + "iopub.status.busy": "2022-12-02T12:08:47.452318Z", + "iopub.status.idle": "2022-12-02T12:08:47.475013Z", + "shell.execute_reply": "2022-12-02T12:08:47.474701Z" } }, "outputs": [ @@ -216,10 +216,10 @@ "execution_count": 5, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:02.194586Z", - "iopub.status.busy": "2022-12-01T10:52:02.194497Z", - "iopub.status.idle": "2022-12-01T10:52:09.816034Z", - "shell.execute_reply": "2022-12-01T10:52:09.815763Z" + "iopub.execute_input": "2022-12-02T12:08:47.476608Z", + "iopub.status.busy": "2022-12-02T12:08:47.476475Z", + "iopub.status.idle": "2022-12-02T12:08:52.964810Z", + "shell.execute_reply": "2022-12-02T12:08:52.964564Z" }, "lines_to_next_cell": 2 }, @@ -227,271 +227,271 @@ { "data": { "text/plain": [ - "{'average': array([[[[ 2.10856146, 2.1077827 , 2.1048762 , ..., 0.93125292,\n", - " 0.92666292, 0.90932505],\n", - " [ 2.12645848, 2.12755666, 2.12542525, ..., 0.9446809 ,\n", - " 0.93851936, 0.9146253 ],\n", - " [ 2.13219665, 2.13385508, 2.13196229, ..., 0.95028683,\n", - " 0.9434714 , 0.91691984],\n", + "{'average': array([[[[ 2.12806218, 2.12812882, 2.12053163, ..., 0.93051087,\n", + " 0.91885457, 0.89506308],\n", + " [ 2.17004784, 2.17488806, 2.17157337, ..., 0.99775607,\n", + " 0.97215348, 0.90332616],\n", + " [ 2.17453529, 2.17976781, 2.17684443, ..., 1.00924438,\n", + " 0.98229693, 0.9050999 ],\n", " ...,\n", - " [ 2.23949928, 2.24677985, 2.24795083, ..., 2.19554228,\n", - " 2.18991997, 2.14811507],\n", - " [ 2.23879781, 2.24612843, 2.24731028, ..., 2.19500655,\n", - " 2.1894073 , 2.14784806],\n", - " [ 2.20836602, 2.21735378, 2.21885402, ..., 2.16351668,\n", - " 2.15798824, 2.1191099 ]],\n", + " [ 2.23490226, 2.24359117, 2.24504145, ..., 2.1939063 ,\n", + " 2.18259313, 2.06587671],\n", + " [ 2.23376516, 2.24253017, 2.24399803, ..., 2.19278217,\n", + " 2.18149269, 2.06572373],\n", + " [ 2.23122066, 2.24015101, 2.24165554, ..., 2.19018788,\n", + " 2.17893132, 2.06499145]],\n", " \n", - " [[ 2.105609 , 2.1036283 , 2.10018278, ..., 0.92181649,\n", - " 0.91763893, 0.90225176],\n", - " [ 2.12394121, 2.12406387, 2.12149445, ..., 0.93391328,\n", - " 0.9283087 , 0.90691762],\n", - " [ 2.12982513, 2.13057832, 2.12827895, ..., 0.9389962 ,\n", - " 0.9327999 , 0.90898068],\n", + " [[ 2.12281642, 2.12101013, 2.11174609, ..., 0.91623534,\n", + " 0.9061637 , 0.88829476],\n", + " [ 2.16640606, 2.17028381, 2.16610168, ..., 0.97588695,\n", + " 0.95241597, 0.89527984],\n", + " [ 2.17107728, 2.17543015, 2.17171163, ..., 0.98689477,\n", + " 0.96157787, 0.89687426],\n", " ...,\n", - " [ 2.24057365, 2.24774131, 2.24888053, ..., 2.19578796,\n", - " 2.19008461, 2.14756149],\n", - " [ 2.23990193, 2.24711745, 2.24826699, ..., 2.19529492,\n", - " 2.1896173 , 2.14736423],\n", - " [ 2.21009789, 2.21892014, 2.22037062, ..., 2.16425777,\n", - " 2.15866404, 2.1192713 ]],\n", + " [ 2.23524618, 2.24378261, 2.24513838, ..., 2.19261985,\n", + " 2.18101667, 2.0612673 ],\n", + " [ 2.2341439 , 2.2427524 , 2.24412402, ..., 2.19153943,\n", + " 2.17996674, 2.06127267],\n", + " [ 2.2316691 , 2.24043444, 2.24183882, ..., 2.18902365,\n", + " 2.17749586, 2.0608303 ]],\n", " \n", - " [[ 2.09200356, 2.08529229, 2.07970064, ..., 0.89591966,\n", - " 0.89289594, 0.88292051],\n", - " [ 2.11176131, 2.10802902, 2.10370823, ..., 0.9042853 ,\n", - " 0.90022775, 0.88576923],\n", - " [ 2.11813962, 2.11531441, 2.11138832, ..., 0.90789753,\n", - " 0.90341688, 0.88716248],\n", + " [[ 2.12072449, 2.11819148, 2.1082753 , ..., 0.91159234,\n", + " 0.90203112, 0.88607823],\n", + " [ 2.1649047 , 2.16840962, 2.16388852, ..., 0.96833654,\n", + " 0.94599791, 0.89262986],\n", + " [ 2.16964468, 2.17365719, 2.16962813, ..., 0.97925226,\n", + " 0.95471255, 0.89416266],\n", " ...,\n", - " [ 2.24155214, 2.24845284, 2.24949996, ..., 2.19301407,\n", - " 2.1869534 , 2.14155852],\n", - " [ 2.24096354, 2.24790548, 2.24896143, ..., 2.19266668,\n", - " 2.18664298, 2.14161746],\n", - " [ 2.21276005, 2.22113507, 2.22243707, ..., 2.16307026,\n", - " 2.1572248 , 2.1160431 ]],\n", + " [ 2.23523077, 2.24371838, 2.24504027, ..., 2.19196472,\n", + " 2.18024785, 2.05939241],\n", + " [ 2.23413987, 2.24269814, 2.24403528, ..., 2.19090002,\n", + " 2.17921638, 2.05945785],\n", + " [ 2.23168767, 2.24039976, 2.24176833, ..., 2.18841227,\n", + " 2.17677847, 2.05912582]],\n", " \n", " ...,\n", " \n", - " [[-0.14012425, -0.16086083, -0.16685416, ..., -0.19393317,\n", - " -0.19297787, -0.18392971],\n", - " [-0.14523253, -0.16609406, -0.1720816 , ..., -0.19926561,\n", - " -0.19835291, -0.18960593],\n", - " [-0.14675178, -0.16764144, -0.17362493, ..., -0.20085752,\n", - " -0.19995994, -0.19132048],\n", + " [[-0.14948253, -0.17263516, -0.18302049, ..., -0.2003148 ,\n", + " -0.19835971, -0.16808598],\n", + " [-0.16203483, -0.18502674, -0.19509185, ..., -0.21174827,\n", + " -0.20995507, -0.1813824 ],\n", + " [-0.1631969 , -0.18616223, -0.19619348, ..., -0.21281258,\n", + " -0.21104057, -0.18269635],\n", " ...,\n", - " [ 0.97310615, 0.94862065, 0.93762868, ..., -0.17896872,\n", - " -0.17965368, -0.17906634],\n", - " [ 0.97706795, 0.95330626, 0.94269863, ..., 0.37286404,\n", - " -0.17786135, -0.17742511],\n", - " [ 1.02371827, 1.0056368 , 0.99773775, ..., 0.86231584,\n", - " 0.86111237, 0.85986918]],\n", + " [ 0.91315283, 0.87774825, 0.85869252, ..., 0.25911455,\n", + " -0.1902646 , -0.16712468],\n", + " [ 0.91719221, 0.88133926, 0.86185007, ..., 0.26079398,\n", + " -0.18860955, -0.16544992],\n", + " [ 0.9252708 , 0.88861802, 0.86828415, ..., 0.81429542,\n", + " 0.26484367, -0.1619619 ]],\n", " \n", - " [[-0.14245589, -0.1621704 , -0.16780717, ..., -0.19132731,\n", - " -0.19030503, -0.18101217],\n", - " [-0.1485767 , -0.16836803, -0.1739778 , ..., -0.19728967,\n", - " -0.1962993 , -0.18724341],\n", - " [-0.15045947, -0.17025932, -0.17585665, ..., -0.1991001 ,\n", - " -0.19812114, -0.189151 ],\n", + " [[-0.14907015, -0.17151593, -0.18152415, ..., -0.19730719,\n", + " -0.19530033, -0.16534825],\n", + " [-0.16325371, -0.18551301, -0.19516247, ..., -0.20996195,\n", + " -0.20808501, -0.17957558],\n", + " [-0.16465098, -0.18687384, -0.19648084, ..., -0.21118621,\n", + " -0.20932606, -0.18101325],\n", " ...,\n", - " [ 0.90396204, 0.87396408, 0.86436039, ..., -0.19602492,\n", - " -0.1958128 , -0.19099102],\n", - " [ 0.9082094 , 0.87767461, 0.86782325, ..., -0.19455264,\n", - " -0.19435572, -0.18957339],\n", - " [ 0.97650027, 0.9489575 , 0.93643106, ..., 0.84197801,\n", - " 0.84193334, 0.3961317 ]],\n", + " [ 0.8832024 , 0.85172775, 0.83627104, ..., -0.19793917,\n", + " -0.19666631, -0.17067148],\n", + " [ 0.88639947, 0.8545582 , 0.83880406, ..., -0.19638084,\n", + " -0.19511772, -0.16903502],\n", + " [ 0.89291214, 0.86038859, 0.84404808, ..., -0.19310058,\n", + " -0.1918577 , -0.16560897]],\n", " \n", - " [[-0.14184696, -0.16100711, -0.16646736, ..., -0.18847825,\n", - " -0.18743994, -0.17815596],\n", - " [-0.1484408 , -0.16768382, -0.17311585, ..., -0.1948184 ,\n", - " -0.19380641, -0.18472248],\n", - " [-0.15049478, -0.1697462 , -0.17516465, ..., -0.19675966,\n", - " -0.19575719, -0.18674693],\n", + " [[-0.14275612, -0.16400043, -0.17348931, ..., -0.18802435,\n", + " -0.18608428, -0.15808698],\n", + " [-0.15916131, -0.18041845, -0.18960299, ..., -0.20289901,\n", + " -0.20103683, -0.17397031],\n", + " [-0.16088801, -0.18211956, -0.19126098, ..., -0.20441021,\n", + " -0.20255819, -0.17562861],\n", " ...,\n", - " [ 0.87603835, 0.84932046, 0.84128626, ..., -0.20259276,\n", - " -0.20207181, -0.19575216],\n", - " [ 0.87923956, 0.85210878, 0.8439045 , ..., -0.20126794,\n", - " -0.20075362, -0.19443275],\n", - " [ 0.94415071, 0.91291958, 0.9022238 , ..., 0.83382846,\n", - " 0.83421198, -0.15951337]]],\n", + " [ 0.85543813, 0.82869251, 0.81665057, ..., -0.20343265,\n", + " -0.20153485, -0.17316956],\n", + " [ 0.85777955, 0.83078757, 0.81857114, ..., -0.20201519,\n", + " -0.2001097 , -0.17156798],\n", + " [ 0.86265084, 0.83519085, 0.82262634, ..., -0.19899369,\n", + " -0.19707408, -0.16819227]]],\n", " \n", " \n", - " [[[ 0.94235351, 0.36399581, -0.09531265, ..., -0.14791667,\n", - " -0.1467608 , -0.13478173],\n", - " [ 0.93364826, 0.35535294, -0.1038551 , ..., -0.15521422,\n", - " -0.15406163, -0.14213811],\n", - " [ 0.93071308, 0.3525149 , 0.34336444, ..., -0.15744697,\n", - " -0.15629508, -0.14439362],\n", + " [[[ 0.94459866, 0.45721915, -0.11366972, ..., -0.16663873,\n", + " -0.16498337, -0.1260251 ],\n", + " [ 0.92375532, 0.88888422, 0.41965229, ..., -0.17896273,\n", + " -0.17737772, -0.13972766],\n", + " [ 0.9210916 , 0.88673141, 0.41780732, ..., -0.18003573,\n", + " -0.17845377, -0.14096853],\n", " ...,\n", - " [ 0.84136929, 0.83359936, 0.83245208, ..., 0.31752302,\n", - " 0.32022673, -0.1119945 ],\n", - " [ 0.84199838, 0.83445172, 0.83340802, ..., 0.32075445,\n", - " 0.32352931, -0.10841314],\n", - " [ 0.8731447 , 0.8684991 , 0.8688803 , ..., 0.94269062,\n", - " 0.94656129, 0.96777665]],\n", + " [ 0.85625866, 0.85067344, 0.85187697, ..., 0.89827988,\n", + " 0.9034434 , -0.05966407],\n", + " [ 0.85720099, 0.85187969, 0.85334416, ..., 0.90246808,\n", + " 0.90780021, -0.05540232],\n", + " [ 0.85939519, 0.85456382, 0.85652093, ..., 0.91098125,\n", + " 0.91665002, -0.04680791]],\n", " \n", - " [[ 0.94495286, 0.36480513, -0.09513894, ..., -0.15259119,\n", - " -0.1515229 , -0.13997972],\n", - " [ 0.93668846, 0.35656479, -0.10329709, ..., -0.15973648,\n", - " -0.15868063, -0.14725607],\n", - " [ 0.93389407, 0.35386099, 0.34405233, ..., -0.16191302,\n", - " -0.16086087, -0.1494786 ],\n", + " [[ 0.9523046 , 0.46164127, -0.11132451, ..., -0.17073828,\n", + " -0.16934745, -0.13169533],\n", + " [ 0.93372086, 0.89547191, 0.4239512 , ..., -0.18231414,\n", + " -0.18105835, -0.14518168],\n", + " [ 0.93128268, 0.89357341, 0.42235199, ..., -0.1832827 ,\n", + " -0.18203865, -0.14638269],\n", " ...,\n", - " [ 0.84341283, 0.83615456, 0.83516404, ..., 0.31926126,\n", - " 0.32177041, -0.11173742],\n", - " [ 0.84394345, 0.83693685, 0.83606365, ..., 0.32270011,\n", - " 0.32528409, 0.3420592 ],\n", - " [ 0.87320002, 0.86935466, 0.87005171, ..., 0.94781259,\n", - " 0.95161938, 0.97163022]],\n", + " [ 0.8619457 , 0.85814864, 0.86045096, ..., 0.90752193,\n", + " 0.91189011, -0.05769929],\n", + " [ 0.86269822, 0.85919672, 0.86180387, ..., 0.91211907,\n", + " 0.9166693 , -0.05312282],\n", + " [ 0.86451523, 0.86155533, 0.86473189, ..., 0.92141576,\n", + " 0.92633653, -0.0438984 ]],\n", " \n", - " [[ 0.9570098 , -0.07910038, -0.09116218, ..., -0.16373012,\n", - " -0.16291839, -0.15270061],\n", - " [ 0.95069569, 0.36430767, -0.09779669, ..., -0.17039489,\n", - " -0.16962086, -0.15968656],\n", - " [ 0.94854018, 0.3621615 , -0.09992667, ..., -0.17239768,\n", - " -0.17163574, -0.16179719],\n", + " [[ 0.95531705, 0.46344585, -0.11028362, ..., -0.17198273,\n", + " -0.17068136, -0.13348214],\n", + " [ 0.93767487, 0.89814364, 0.42575083, ..., -0.18330194,\n", + " -0.18215831, -0.14688823],\n", + " [ 0.93533456, 0.89634968, 0.42424912, ..., -0.18423446,\n", + " -0.18310571, -0.14807515],\n", " ...,\n", - " [ 0.85617843, 0.85148051, 0.85130089, ..., 0.33149267,\n", - " 0.33316862, 0.34435703],\n", - " [ 0.85631146, 0.85196878, 0.85196336, ..., 0.3358008 ,\n", - " 0.3375546 , 0.34896629],\n", - " [ 0.87821551, 0.87769879, 0.87972953, ..., 0.96904086,\n", - " 0.97217399, 0.98791369]],\n", + " [ 0.86434913, 0.86130268, 0.86407289, ..., 0.91138389,\n", + " 0.91541668, -0.05676232],\n", + " [ 0.8650263 , 0.86228529, 0.86537693, ..., 0.91614477,\n", + " 0.92036329, -0.05207173],\n", + " [ 0.86669411, 0.86450957, 0.86819888, ..., 0.92575407,\n", + " 0.93035365, -0.04261905]],\n", " \n", " ...,\n", " \n", - " [[ 1.56454057, 1.09280136, 1.07924674, ..., 0.84473013,\n", - " 0.84278527, 0.83920233],\n", - " [ 1.57797648, 1.55921062, 1.09700716, ..., 0.84308495,\n", - " 0.84065254, 0.8348622 ],\n", - " [ 1.58238638, 1.56467204, 1.10296893, ..., 0.84299067,\n", - " 0.84037093, 0.83375457],\n", + " [[ 2.14276495, 1.68027094, 1.11012714, ..., 0.86547088,\n", + " 0.85775615, 0.85390535],\n", + " [ 2.17455008, 1.71998508, 1.70752094, ..., 0.88536862,\n", + " 0.87066138, 0.84544097],\n", + " [ 2.17805884, 2.17438919, 1.71282474, ..., 0.8902414 ,\n", + " 0.87411364, 0.84513552],\n", " ...,\n", - " [ 2.21326466, 2.22953466, 2.23342837, ..., 2.21331111,\n", - " 2.20908039, 2.17475907],\n", - " [ 2.21191687, 2.22866828, 2.23271324, ..., 2.21498707,\n", - " 2.21093899, 2.17810716],\n", - " [ 2.16531624, 2.19229363, 2.19961485, ..., 2.21573147,\n", - " 2.21341726, 2.19412656]],\n", + " [ 2.21618069, 2.2356562 , 2.24275885, ..., 2.22981417,\n", + " 2.22331874, 1.6866905 ],\n", + " [ 2.21449671, 2.23444913, 2.24179363, ..., 2.23016661,\n", + " 2.2238609 , 2.13986239],\n", + " [ 2.21081286, 2.23174591, 2.23958726, ..., 2.23047901,\n", + " 2.22451318, 2.14518643]],\n", " \n", - " [[ 1.56851899, 1.10101727, 1.08989541, ..., 0.856133 ,\n", - " 0.85371364, 0.84777712],\n", - " [ 1.58210957, 1.56734863, 1.10744297, ..., 0.85512992,\n", - " 0.85212023, 0.84357915],\n", - " [ 2.13657056, 1.57276794, 1.56330587, ..., 0.85529867,\n", - " 0.85206008, 0.84253524],\n", + " [[ 2.14308257, 1.68270812, 1.11511957, ..., 0.8759467 ,\n", + " 0.86726833, 0.85899037],\n", + " [ 2.17501117, 2.17192788, 1.71117064, ..., 0.89861812,\n", + " 0.88202588, 0.8502419 ],\n", + " [ 2.17854945, 2.17628576, 1.71632873, ..., 0.90407061,\n", + " 0.8858568 , 0.84990825],\n", " ...,\n", - " [ 2.22243162, 2.23740097, 2.24093354, ..., 2.22077252,\n", - " 2.21672725, 2.18372315],\n", - " [ 2.22130022, 2.2366624 , 2.24031726, ..., 2.22214065,\n", - " 2.21825192, 2.18654916],\n", - " [ 2.17970459, 2.20372905, 2.21011212, ..., 2.22036999,\n", - " 2.21799269, 2.19837493]],\n", + " [ 2.21977404, 2.23829573, 2.24497384, ..., 2.23104404,\n", + " 2.22453006, 1.68730491],\n", + " [ 2.21820555, 2.23715658, 2.24405219, ..., 2.23133563,\n", + " 2.22500123, 1.69035254],\n", + " [ 2.2147613 , 2.23459673, 2.24193965, ..., 2.23154189,\n", + " 2.2255297 , 2.14546923]],\n", " \n", - " [[ 1.56951924, 1.5542593 , 1.09441338, ..., 0.86421418,\n", - " 0.86148523, 0.85397088],\n", - " [ 1.58310839, 1.57040895, 1.5616741 , ..., 0.86356079,\n", - " 0.86017696, 0.84980836],\n", - " [ 2.13757361, 1.57576659, 1.56743562, ..., 0.86387322,\n", - " 0.86023459, 0.8487839 ],\n", + " [[ 2.1398818 , 1.68262919, 1.66881217, ..., 0.89726057,\n", + " 0.88689141, 0.86994412],\n", + " [ 2.17178687, 2.17068748, 1.71222083, ..., 0.92283678,\n", + " 0.90334711, 0.86003174],\n", + " [ 2.17536585, 2.17496322, 1.71713685, ..., 0.92885067,\n", + " 0.90750949, 0.8595432 ],\n", " ...,\n", - " [ 2.22583873, 2.24018227, 2.24354455, ..., 2.22296046,\n", - " 2.21895671, 2.18622403],\n", - " [ 2.22480469, 2.23950065, 2.24297256, ..., 2.22420939,\n", - " 2.22035227, 2.18885093],\n", - " [ 2.18554461, 2.20812483, 2.21406841, ..., 2.22143196,\n", - " 2.21900338, 2.19907765]]],\n", + " [ 2.22150825, 2.23869201, 2.24477559, ..., 2.22820374,\n", + " 2.22137119, 1.68025661],\n", + " [ 2.22008497, 2.23763001, 2.2438993 , ..., 2.22844114,\n", + " 2.22178417, 1.68331838],\n", + " [ 2.21694174, 2.23523124, 2.24188223, ..., 2.22854884,\n", + " 2.22220826, 2.13848481]]],\n", " \n", " \n", - " [[[-0.07142199, 0.51188746, 0.9817732 , ..., 2.17129902,\n", - " 2.17217167, 2.17091732],\n", - " [-0.08830166, 0.48972307, 0.95694004, ..., 2.17236604,\n", - " 2.17353781, 2.1737159 ],\n", - " [-0.0938647 , 0.48239078, 0.49870027, ..., 2.17239533,\n", - " 2.1736789 , 2.17438376],\n", + " [[[-0.0982615 , 0.38732808, 0.9856721 , ..., 2.18621307,\n", + " 2.18868332, 2.17220672],\n", + " [-0.14065009, -0.12298073, 0.35482959, ..., 2.17958459,\n", + " 2.18433269, 2.17713384],\n", + " [-0.14530845, -0.12945851, 0.34625033, ..., 2.1778117 ,\n", + " 2.18295281, 2.17720635],\n", " ...,\n", - " [-0.20580881, -0.21490484, -0.21642526, ..., 0.40874475,\n", - " 0.41698052, 0.92752695],\n", - " [-0.20488089, -0.21415697, -0.2157453 , ..., 0.40677883,\n", - " 0.41473316, 0.92305395],\n", - " [-0.16999943, -0.18241452, -0.18513704, ..., -0.14333716,\n", - " -0.13836634, -0.10353079]],\n", + " [-0.20471356, -0.21652592, -0.21925163, ..., -0.160141 ,\n", + " -0.14611178, 0.99381407],\n", + " [-0.20327071, -0.21529468, -0.21815339, ..., -0.16085789,\n", + " -0.14724712, 0.98803038],\n", + " [-0.20010272, -0.21254817, -0.21566802, ..., -0.1617002 ,\n", + " -0.14884237, 0.97773572]],\n", " \n", - " [[-0.0695367 , 0.51813011, 0.99048315, ..., 2.17834936,\n", - " 2.17913323, 2.17750376],\n", - " [-0.0870521 , 0.4946253 , 0.96410251, ..., 2.17962508,\n", - " 2.18068998, 2.18042375],\n", - " [-0.09282234, 0.48685444, 0.5052709 , ..., 2.1797325 ,\n", - " 2.18090224, 2.18113535],\n", + " [[-0.09642709, 0.39461623, 0.99925229, ..., 2.19386421,\n", + " 2.19608005, 2.17894392],\n", + " [-0.14047797, -0.11998021, 0.36166112, ..., 2.1882176 ,\n", + " 2.19252739, 2.18415098],\n", + " [-0.14530574, -0.12687124, 0.35228492, ..., 2.18661392,\n", + " 2.19128408, 2.18426492],\n", " ...,\n", - " [-0.20850394, -0.21755993, -0.21906968, ..., 0.40719093,\n", - " 0.41567374, 0.9287984 ],\n", - " [-0.20759117, -0.21683336, -0.21841304, ..., 0.40504181,\n", - " 0.41321668, 0.47385201],\n", - " [-0.17277483, -0.18529506, -0.18804767, ..., -0.14648141,\n", - " -0.1414693 , -0.10611662]],\n", + " [-0.20735956, -0.2191907 , -0.22193363, ..., -0.16256896,\n", + " -0.1482605 , 0.99805058],\n", + " [-0.20593213, -0.21798385, -0.22086428, ..., -0.16340856,\n", + " -0.14955135, 0.99174726],\n", + " [-0.20279168, -0.21528329, -0.21843377, ..., -0.16446354,\n", + " -0.15141329, 0.98022545]],\n", " \n", - " [[-0.06155031, 0.9907933 , 1.01855867, ..., 2.19568453,\n", - " 2.19625069, 2.19377238],\n", - " [-0.08096848, 0.51333301, 0.99031909, ..., 2.19743328,\n", - " 2.19823578, 2.19695001],\n", - " [-0.08736556, 0.50409983, 0.98009191, ..., 2.19772187,\n", - " 2.19861192, 2.19775779],\n", + " [[-0.0956718 , 0.39734699, 1.00405672, ..., 2.19625777,\n", + " 2.19839404, 2.18107015],\n", + " [-0.14029688, -0.11878781, 0.36426114, ..., 2.19091689,\n", + " 2.19508944, 2.186363 ],\n", + " [-0.14518357, -0.1258268 , 0.35459204, ..., 2.1893664 ,\n", + " 2.19388906, 2.18648999],\n", " ...,\n", - " [-0.21469814, -0.22369674, -0.22519239, ..., 0.4039751 ,\n", - " 0.41315921, 0.48392821],\n", - " [-0.21383198, -0.22303388, -0.22460523, ..., 0.401266 ,\n", - " 0.41005773, 0.47745278],\n", - " [-0.17933093, -0.19218542, -0.19503723, ..., -0.15454168,\n", - " -0.14945353, -0.11293154]],\n", + " [-0.20816362, -0.22000527, -0.22275587, ..., -0.16332068,\n", + " -0.14892235, 0.99949937],\n", + " [-0.20674151, -0.21880669, -0.22169631, ..., -0.16420209,\n", + " -0.15026623, 0.99302019],\n", + " [-0.20361065, -0.21612178, -0.21928444, ..., -0.16532948,\n", + " -0.15221901, 0.98110164]],\n", " \n", " ...,\n", " \n", - " [[ 1.59871862, 2.11238752, 2.13123501, ..., 2.2317425 ,\n", - " 2.23183097, 2.2282522 ],\n", - " [ 1.58264779, 1.65437477, 2.12540754, ..., 2.23528101,\n", - " 2.23549636, 2.23271225],\n", - " [ 1.57617828, 1.6507296 , 2.12264443, ..., 2.23622825,\n", - " 2.23648844, 2.23397619],\n", + " [[ 1.00992467, 1.54208539, 2.13317249, ..., 2.22934726,\n", + " 2.23021832, 2.2142571 ],\n", + " [ 0.95176593, 1.48640098, 1.54884488, ..., 2.23267815,\n", + " 2.23448304, 2.22445268],\n", + " [ 0.94352639, 1.02568678, 1.54137339, ..., 2.23252507,\n", + " 2.23447674, 2.22524838],\n", " ...,\n", - " [-0.2092948 , -0.22061482, -0.22271863, ..., 0.87291825,\n", - " 0.88677862, 1.0136662 ],\n", - " [-0.20886965, -0.22052548, -0.22275633, ..., 0.3152697 ,\n", - " 0.87791028, 0.99506506],\n", - " [-0.17839668, -0.19569825, -0.19992171, ..., -0.17513095,\n", - " -0.17034289, -0.13292719]],\n", + " [-0.19751458, -0.21304678, -0.21747714, ..., 0.39464702,\n", + " 0.86312839, 1.52734971],\n", + " [-0.19640507, -0.21230125, -0.21696571, ..., 0.39118953,\n", + " 0.85852572, 1.06798856],\n", + " [-0.19382994, -0.2104414 , -0.21555835, ..., -0.16434937,\n", + " 0.4010757 , 1.04844601]],\n", " \n", - " [[ 1.5975225 , 2.10759552, 2.12575267, ..., 2.22655111,\n", - " 2.22670693, 2.22353048],\n", - " [ 1.58322918, 1.65038228, 2.12053121, ..., 2.2304123 ,\n", - " 2.23069896, 2.22833832],\n", - " [ 1.02740793, 1.64707676, 1.66801883, ..., 2.23146181,\n", - " 2.23179465, 2.22971354],\n", + " [[ 1.00881118, 1.53667077, 2.12681118, ..., 2.22448156,\n", + " 2.22551653, 2.21068559],\n", + " [ 0.95437068, 1.03401266, 1.5437813 , ..., 2.22842318,\n", + " 2.23041982, 2.22164466],\n", + " [ 0.94653891, 1.02413567, 1.53663604, ..., 2.22833904,\n", + " 2.23048588, 2.22253052],\n", " ...,\n", - " [-0.20357975, -0.21505941, -0.21719173, ..., 0.88391165,\n", - " 0.8987604 , 1.0300875 ],\n", - " [-0.20318517, -0.2150139 , -0.21727943, ..., 0.87560624,\n", - " 0.88910361, 1.01249804],\n", - " [-0.1728603 , -0.19037667, -0.19468345, ..., -0.16964158,\n", - " -0.16474209, 0.32363472]],\n", + " [-0.19291171, -0.20842425, -0.21281768, ..., 0.85250956,\n", + " 0.8719751 , 1.53494256],\n", + " [-0.19180367, -0.20768438, -0.21231848, ..., 0.84879312,\n", + " 0.86701994, 1.52620949],\n", + " [-0.18922797, -0.20582934, -0.21092745, ..., 0.84280632,\n", + " 0.85896777, 1.05799014]],\n", " \n", - " [[ 1.5947905 , 1.65268123, 2.12045852, ..., 2.22226445,\n", - " 2.22248451, 2.21968249],\n", - " [ 1.58133311, 1.64579663, 1.66548146, ..., 2.22636602,\n", - " 2.22672058, 2.22475132],\n", - " [ 1.02592011, 1.64263506, 1.66307284, ..., 2.22749188,\n", - " 2.2278939 , 2.2262104 ],\n", + " [[ 1.00197386, 1.52071572, 1.55925903, ..., 2.21212797,\n", + " 2.21365059, 2.20205311],\n", + " [ 0.95458837, 1.02276806, 1.52762689, ..., 2.21756444,\n", + " 2.22012043, 2.21478128],\n", + " [ 0.94781113, 1.01461284, 1.52089697, ..., 2.21765316,\n", + " 2.22036804, 2.21587787],\n", " ...,\n", - " [-0.19946176, -0.21092973, -0.21304404, ..., 0.8932248 ,\n", - " 0.90896263, 1.04208135],\n", - " [-0.19907829, -0.21090349, -0.21315507, ..., 0.88440323,\n", - " 0.89867601, 1.02560817],\n", - " [-0.16875051, -0.18626631, -0.19058746, ..., -0.16470368,\n", - " -0.15968349, 0.8796544 ]]]]),\n", - " 'values': [array([35.7, 36.2, 37.6, 38.1, 38.3, 38.9, 39.7, 40.6, 40.9, 41.1, 41.3,\n", - " 43.3, 45.2, 45.3, 46.5, 46.8, 49. , 50.4, 51.3]),\n", - " array([13.4, 13.7, 13.8, 14.5, 15.7, 15.8, 16.1, 16.8, 17.2, 17.6, 17.8,\n", - " 17.9, 18. , 18.2, 18.6, 19.1, 19.2, 21.1]),\n", - " array([181., 187., 189., 191., 193., 194., 195., 197., 202., 209., 210.,\n", - " 213., 215., 216., 222.])]}" + " [-0.182481 , -0.19742661, -0.20145431, ..., 0.87437689,\n", + " 0.89687623, 1.55097587],\n", + " [-0.18135497, -0.19667386, -0.20095613, ..., 0.86995675,\n", + " 0.89093051, 1.5435827 ],\n", + " [-0.17874002, -0.1947832 , -0.19955361, ..., 0.86274997,\n", + " 0.88118655, 1.07818663]]]]),\n", + " 'values': [array([36.7, 37.3, 37.5, 37.7, 39. , 40.8, 44.1, 44.5, 45.3, 45.4, 45.5,\n", + " 45.7, 46.6, 49. , 49.3, 50.5, 51.5, 53.5]),\n", + " array([13.8, 14.6, 14.7, 16.1, 16.3, 16.8, 17. , 17.1, 17.3, 17.8, 18.4,\n", + " 18.5, 18.7, 18.8, 18.9, 19.6, 19.7, 19.9]),\n", + " array([180., 187., 191., 192., 193., 196., 199., 200., 201., 203., 205.,\n", + " 208., 211., 214., 216., 230.])]}" ] }, "execution_count": 5, @@ -511,10 +511,10 @@ "execution_count": 6, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:09.817645Z", - "iopub.status.busy": "2022-12-01T10:52:09.817532Z", - "iopub.status.idle": "2022-12-01T10:52:10.635050Z", - "shell.execute_reply": "2022-12-01T10:52:10.634703Z" + "iopub.execute_input": "2022-12-02T12:08:52.966380Z", + "iopub.status.busy": "2022-12-02T12:08:52.966303Z", + "iopub.status.idle": "2022-12-02T12:08:53.780956Z", + "shell.execute_reply": "2022-12-02T12:08:53.780633Z" }, "scrolled": true }, @@ -542,10 +542,10 @@ "execution_count": 7, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:10.637181Z", - "iopub.status.busy": "2022-12-01T10:52:10.637090Z", - "iopub.status.idle": "2022-12-01T10:52:11.134009Z", - "shell.execute_reply": "2022-12-01T10:52:11.133613Z" + "iopub.execute_input": "2022-12-02T12:08:53.782668Z", + "iopub.status.busy": "2022-12-02T12:08:53.782570Z", + "iopub.status.idle": "2022-12-02T12:08:54.211902Z", + "shell.execute_reply": "2022-12-02T12:08:54.211578Z" } }, "outputs": [ @@ -570,10 +570,10 @@ "execution_count": 8, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:11.135895Z", - "iopub.status.busy": "2022-12-01T10:52:11.135796Z", - "iopub.status.idle": "2022-12-01T10:52:11.564379Z", - "shell.execute_reply": "2022-12-01T10:52:11.563972Z" + "iopub.execute_input": "2022-12-02T12:08:54.213614Z", + "iopub.status.busy": "2022-12-02T12:08:54.213528Z", + "iopub.status.idle": "2022-12-02T12:08:54.618939Z", + "shell.execute_reply": "2022-12-02T12:08:54.618646Z" } }, "outputs": [ @@ -605,17 +605,17 @@ "execution_count": 9, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:11.566406Z", - "iopub.status.busy": "2022-12-01T10:52:11.566181Z", - "iopub.status.idle": "2022-12-01T10:52:11.619009Z", - "shell.execute_reply": "2022-12-01T10:52:11.618751Z" + "iopub.execute_input": "2022-12-02T12:08:54.620708Z", + "iopub.status.busy": "2022-12-02T12:08:54.620613Z", + "iopub.status.idle": "2022-12-02T12:08:54.672699Z", + "shell.execute_reply": "2022-12-02T12:08:54.672193Z" } }, "outputs": [ { "data": { "text/plain": [ - "0.9900990099009901" + "1.0" ] }, "execution_count": 9, @@ -648,16 +648,16 @@ "execution_count": 10, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:11.620737Z", - "iopub.status.busy": "2022-12-01T10:52:11.620607Z", - "iopub.status.idle": "2022-12-01T10:52:11.670180Z", - "shell.execute_reply": "2022-12-01T10:52:11.669851Z" + "iopub.execute_input": "2022-12-02T12:08:54.674860Z", + "iopub.status.busy": "2022-12-02T12:08:54.674592Z", + "iopub.status.idle": "2022-12-02T12:08:54.739445Z", + "shell.execute_reply": "2022-12-02T12:08:54.739180Z" } }, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAIiCAYAAAAadsgUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA0AElEQVR4nO3deXxNd/7H8feNJZGUWJNBIwtqKSqSLtKpjrZi0FHTjvrV2GmlqrXUFNWOMq2lM9VgLPVTFLWUZmbaUuqHEtI1DWGo2mMIKdqg0SC5vz+MPHolVDT3fq7c1/PxyOPhnnNu8k6PNG/f8/2e43A6nU4BAAAY8bMOAAAAfBtlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBV1jrAtcjPz9eRI0dUsWJFORwO6zgAAOAaOJ1OnT59WrVq1ZKf35XHP26IMnLkyBGFhYVZxwAAANfh0KFDuvnmm6+4/4YoIxUrVpR08ZupVKmScRoAAHAtTp06pbCwsILf41dyQ5SRS5dmKlWqRBkBAOAG83NTLJjACgAATFFGAACAKcoIAAAwRRkBAACmKCMAAMAUZQQAAJiijAAAAFOUEQAAYIoyAgAATFFGAACAKcoIAAAwRRkBAACmKCMAAMAUZQQAAJiijAAAAFNlrQN4k4gRK6wj/GIHJnSwjgAAQLEwMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADA1HWVkenTpysyMlIBAQGKiYlRcnLyNb1v8+bNKlu2rJo3b349XxYAAJRCxS4jS5cu1eDBgzVq1CilpaXpnnvuUbt27ZSRkXHV92VnZ6tHjx66//77rzssAAAofYpdRiZNmqS+ffuqX79+atSokRITExUWFqYZM2Zc9X39+/dX165d1bJly+sOCwAASp9ilZFz584pNTVV8fHxLtvj4+OVkpJyxffNnTtXe/fu1ejRo6/p6+Tm5urUqVMuHwAAoHQqVhk5fvy48vLyFBoa6rI9NDRUR48eLfI9u3fv1ogRI/T222+rbNmy1/R1xo8fr+Dg4IKPsLCw4sQEAAA3kOuawOpwOFxeO53OQtskKS8vT127dtWYMWN0yy23XPPnHzlypLKzsws+Dh06dD0xAQDADeDahir+q3r16ipTpkyhUZCsrKxCoyWSdPr0aX355ZdKS0vTwIEDJUn5+flyOp0qW7asPvroI913332F3ufv7y9/f//iRAMAADeoYo2MlC9fXjExMVqzZo3L9jVr1iguLq7Q8ZUqVdK2bdu0ZcuWgo+EhAQ1aNBAW7Zs0Z133vnL0gMAgBtesUZGJGno0KHq3r27YmNj1bJlS82aNUsZGRlKSEiQdPESy+HDhzV//nz5+fmpSZMmLu8PCQlRQEBAoe0AAMA3FbuMdOnSRSdOnNDYsWOVmZmpJk2aaOXKlQoPD5ckZWZm/uw9RwAAAC5xOJ1Op3WIn3Pq1CkFBwcrOztblSpVctvXiRixwm2f21MOTOhgHQEAAEnX/vubZ9MAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAExRRgAAgCnKCAAAMEUZAQAApigjAADAFGUEAACYoowAAABTlBEAAGCKMgIAAEyVtQ4AwLtFjFhhHaFEHJjQwToCgCtgZAQAAJiijAAAAFOUEQAAYOq6ysj06dMVGRmpgIAAxcTEKDk5+YrHbtq0SXfffbeqVaumChUqqGHDhnr99devOzAAAChdij2BdenSpRo8eLCmT5+uu+++W2+88YbatWunHTt2qE6dOoWODwoK0sCBA9WsWTMFBQVp06ZN6t+/v4KCgvTEE0+UyDcBAABuXA6n0+kszhvuvPNOtWjRQjNmzCjY1qhRI3Xq1Enjx4+/ps/x8MMPKygoSAsWLChyf25urnJzcwtenzp1SmFhYcrOzlalSpWKE7dYSsOqAVYMoKSVhp8LiZ8NwMKpU6cUHBz8s7+/i3WZ5ty5c0pNTVV8fLzL9vj4eKWkpFzT50hLS1NKSoruvffeKx4zfvx4BQcHF3yEhYUVJyYAALiBFKuMHD9+XHl5eQoNDXXZHhoaqqNHj171vTfffLP8/f0VGxurp556Sv369bvisSNHjlR2dnbBx6FDh4oTEwAA3ECu66ZnDofD5bXT6Sy07XLJyck6c+aMPv30U40YMUL16tXTY489VuSx/v7+8vf3v55oAADgBlOsMlK9enWVKVOm0ChIVlZWodGSy0VGRkqSmjZtqmPHjumll166YhkBAAC+o1iXacqXL6+YmBitWbPGZfuaNWsUFxd3zZ/H6XS6TFAFAAC+q9iXaYYOHaru3bsrNjZWLVu21KxZs5SRkaGEhARJF+d7HD58WPPnz5ckTZs2TXXq1FHDhg0lXbzvyN/+9jc9/fTTJfhtAACAG1Wxy0iXLl104sQJjR07VpmZmWrSpIlWrlyp8PBwSVJmZqYyMjIKjs/Pz9fIkSO1f/9+lS1bVnXr1tWECRPUv3//kvsuAADADavY9xmxcK3rlH+p0nA/Be6lgJJWGn4uJH42AAtuuc8IAABASaOMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwFRZ6wBAUSJGrLCO8IsdmNDBOgIA3BAYGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABg6rrKyPTp0xUZGamAgADFxMQoOTn5iscmJSWpTZs2qlGjhipVqqSWLVtq9erV1x0YAACULsUuI0uXLtXgwYM1atQopaWl6Z577lG7du2UkZFR5PEbN25UmzZttHLlSqWmpqp169b63e9+p7S0tF8cHgAA3PiKXUYmTZqkvn37ql+/fmrUqJESExMVFhamGTNmFHl8YmKinnvuOd1+++2qX7++xo0bp/r16+v999//xeEBAMCNr1hl5Ny5c0pNTVV8fLzL9vj4eKWkpFzT58jPz9fp06dVtWrVKx6Tm5urU6dOuXwAAIDSqVhl5Pjx48rLy1NoaKjL9tDQUB09evSaPsdrr72mH374QY8++ugVjxk/fryCg4MLPsLCwooTEwAA3ECuawKrw+Fwee10OgttK8rixYv10ksvaenSpQoJCbnicSNHjlR2dnbBx6FDh64nJgAAuAGULc7B1atXV5kyZQqNgmRlZRUaLbnc0qVL1bdvXy1btkwPPPDAVY/19/eXv79/caIBAIAbVLFGRsqXL6+YmBitWbPGZfuaNWsUFxd3xfctXrxYvXr10qJFi9ShQ4frSwoAAEqlYo2MSNLQoUPVvXt3xcbGqmXLlpo1a5YyMjKUkJAg6eIllsOHD2v+/PmSLhaRHj16aPLkybrrrrsKRlUqVKig4ODgEvxWAADAjajYZaRLly46ceKExo4dq8zMTDVp0kQrV65UeHi4JCkzM9PlniNvvPGGLly4oKeeekpPPfVUwfaePXtq3rx5v/w7AAAAN7RilxFJGjBggAYMGFDkvssLxscff3w9XwIAAPgInk0DAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU9dVRqZPn67IyEgFBAQoJiZGycnJVzw2MzNTXbt2VYMGDeTn56fBgwdfb1YAAFAKFbuMLF26VIMHD9aoUaOUlpame+65R+3atVNGRkaRx+fm5qpGjRoaNWqUbrvttl8cGAAAlC7FLiOTJk1S37591a9fPzVq1EiJiYkKCwvTjBkzijw+IiJCkydPVo8ePRQcHHxNXyM3N1enTp1y+QAAAKVTscrIuXPnlJqaqvj4eJft8fHxSklJKbFQ48ePV3BwcMFHWFhYiX1uAADgXYpVRo4fP668vDyFhoa6bA8NDdXRo0dLLNTIkSOVnZ1d8HHo0KES+9wAAMC7lL2eNzkcDpfXTqez0LZfwt/fX/7+/iX2+QAAgPcq1shI9erVVaZMmUKjIFlZWYVGSwAAAK5FscpI+fLlFRMTozVr1rhsX7NmjeLi4ko0GAAA8A3FvkwzdOhQde/eXbGxsWrZsqVmzZqljIwMJSQkSLo43+Pw4cOaP39+wXu2bNkiSTpz5oy+/fZbbdmyReXLl1fjxo1L5rsAAAA3rGKXkS5duujEiRMaO3asMjMz1aRJE61cuVLh4eGSLt7k7PJ7jkRHRxf8OTU1VYsWLVJ4eLgOHDjwy9IDAIAb3nVNYB0wYIAGDBhQ5L558+YV2uZ0Oq/nywAAAB/As2kAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABMUUYAAIApyggAADBV1joAAODaRIxYYR2hRByY0ME6ArwMIyMAAMAUIyMAABQTo1Qli5ERAABg6rrKyPTp0xUZGamAgADFxMQoOTn5qsdv2LBBMTExCggIUFRUlGbOnHldYQEAQOlT7DKydOlSDR48WKNGjVJaWpruuecetWvXThkZGUUev3//frVv31733HOP0tLS9Pzzz+uZZ57Ru++++4vDAwCAG1+xy8ikSZPUt29f9evXT40aNVJiYqLCwsI0Y8aMIo+fOXOm6tSpo8TERDVq1Ej9+vVTnz599Le//e0XhwcAADe+Yk1gPXfunFJTUzVixAiX7fHx8UpJSSnyPZ988oni4+NdtrVt21Zvvvmmzp8/r3LlyhV6T25urnJzcwteZ2dnS5JOnTpVnLjFlp+b49bP7wnu/m/kKZwL71EazoVUOs4H58J7cC6K9/mdTudVjytWGTl+/Ljy8vIUGhrqsj00NFRHjx4t8j1Hjx4t8vgLFy7o+PHjqlmzZqH3jB8/XmPGjCm0PSwsrDhxfVJwonUCXMK58C6cD+/BufAenjoXp0+fVnBw8BX3X9fSXofD4fLa6XQW2vZzxxe1/ZKRI0dq6NChBa/z8/N18uRJVatW7apfx5udOnVKYWFhOnTokCpVqmQdx+dxPrwH58J7cC68R2k5F06nU6dPn1atWrWuelyxykj16tVVpkyZQqMgWVlZhUY/LvnVr35V5PFly5ZVtWrVinyPv7+//P39XbZVrly5OFG9VqVKlW7ov1ilDefDe3AuvAfnwnuUhnNxtRGRS4o1gbV8+fKKiYnRmjVrXLavWbNGcXFxRb6nZcuWhY7/6KOPFBsbW+R8EQAA4FuKvZpm6NChmj17tubMmaOdO3dqyJAhysjIUEJCgqSLl1h69OhRcHxCQoIOHjyooUOHaufOnZozZ47efPNNDRs2rOS+CwAAcMMq9pyRLl266MSJExo7dqwyMzPVpEkTrVy5UuHh4ZKkzMxMl3uOREZGauXKlRoyZIimTZumWrVqacqUKXrkkUdK7ru4Afj7+2v06NGFLj/BBufDe3AuvAfnwnv42rlwOH9uvQ0AAIAb8WwaAABgijICAABMUUYAAIApyggAADBFGQEAAKau63bw+Hm5ubn6/PPPdeDAAeXk5KhGjRqKjo5WZGSkdTSfs2vXLi1evFjJycmFzkfbtm31yCOP+MzyOW+Tm5vLf3sj+/btU2Rk5A37iA2ULiztLWEpKSmaOnWq/vnPf+rcuXOqXLmyKlSooJMnTyo3N1dRUVF64oknlJCQoIoVK1rHLdXS0tL03HPPKTk5WXFxcbrjjjtUu3btgvOxfft2JScn69SpU3ruuec0ePBgfjG62erVqwuKYUZGhvLz8xUYGKgWLVooPj5evXv3/tlnWKBklClTRpmZmQoJCZF08R5SU6ZMueKjPeBeffr0uabj5syZ4+YkNigjJeihhx7SF198oa5du6pjx46KjY1VYGBgwf59+/YpOTlZixcv1tatWzV//ny1adPGMHHpFh4erj/96U/q2rWrqlatesXjPvnkE73++utq3ry5nn/+eQ8m9B3//Oc/NXz4cGVnZ6t9+/ZXLIaffPKJevXqpb/85S+qUaOGdexSzc/PT0ePHi0oIxUrVtTWrVsVFRVlnMw3+fn5KTw8XNHR0brar+V//OMfHkzlOZSREjRt2jQ9/vjjKl++/M8e++9//1tHjhyhjLjRuXPnrulcXO/xuHZ33HGHXnzxRXXo0EF+fleeqnb48GFNnjxZoaGhevbZZz2Y0PdQRrzLgAEDtGTJEtWpU0d9+vRRt27drvqPqNKGMgIAPujSE9gvjUBVrFhR6enpzGszlJubq6SkJM2ZM0cpKSnq0KGD+vbtq/j4+FI/t4cy4gFnzpxRfn6+y7Yb/ZHQN6LPP/9cH3/8sbKysgqdj0mTJhmlAmz4+fmpXbt2BfOk3n//fd13330KCgpyOS4pKckins87ePCg5s2bp/nz5+v8+fPasWOHbrrpJutYbsNqGjfZv3+/Bg4cqI8//lg//vhjwXan0ymHw6G8vDzDdL5n3LhxeuGFF9SgQQOFhoa6/CujtP+Lw9s4nU4tX75c69evL7IY8svPM3r27Onyulu3bkZJUBSHwyGHwyGn01noZ6Q0YmTETeLi4iRJgwYNKvTLT5Luvfdei1g+KzQ0VBMnTlSvXr2so/i8Z555RrNmzVLr1q2L/NmYO3euUTLA1k8v02zatEkPPvigevfurd/+9rdXnWtVGlBG3OSmm25SamqqGjRoYB0FkmrWrKmNGzeqfv361lF8XtWqVbVw4UK1b9/eOgrgNX46gbV3797q1q2bqlWrZh3LYygjbtK6dWuNGjVKDzzwgHUUSHr11Vd15MgRJSYmWkfxeZGRkfrwww/VsGFD6yiA1/Dz81OdOnUUHR191UvHpfUyJmXETfbu3auEhAR169ZNTZo0Ubly5Vz2N2vWzCiZb8rPz1eHDh30zTffqHHjxoXOR2n9AfdGb731llatWqU5c+aoQoUK1nEAr9CrV69rmr9WWi9jMoHVTb799lvt3btXvXv3Lth2aTISE1g97+mnn9b69evVunVrVatWjUmrhjp37qzFixcrJCREERERhYrhV199ZZQMsDNv3jzrCKYoI27Sp08fRUdHa/HixUVO0oNnzZ8/X++++646dOhgHcXn9erVS6mpqerWrRs/GwAkcZnGbYKCgrR161bVq1fPOgp08dbwq1evZp6CFwgKCtLq1av161//2joKAC9RutcKGbrvvvu0detW6xj4r5deekmjR49WTk6OdRSfFxYWxk3/ALhgZMRNZs2apZdffll9+vRR06ZNC10X79ixo1Ey3xQdHa29e/fK6XQyT8HYihUrNHXqVM2cOVMRERHWcQB4AcqIm1ztBjVMYPW8MWPGXHX/6NGjPZQEVapUUU5Oji5cuKDAwMBCxfDkyZNGyQBYoYwA8Ki33nrrqvsvv005gNKPMgKfw4MLAcC7sLTXjXhKrPfgwYXeJysrq8ifDW4ICPgeyoib8JRY7/LHP/5RkjRnzhzubWEsNTVVPXv21M6dO3X5wCzFEPBNXKZxE54S6114cKH3aNasmerVq6fhw4cXWQzDw8ONkgGwwsiIm/j5+enuu++2joH/uv3223Xo0CHKiBfYv3+/kpKSuCEggAKUETcZMmSIpk2bxlNivcTs2bOVkJCgw4cP8+BCY/fffz93Jwbggss0bsJTYr3Lp59+qq5du+rAgQMF23hwoY3jx4+rZ8+euuOOO4oshtwQEPA9jIy4CU+J9S48uNB7pKSkaNOmTfrwww8L7aMYAr6JkRE3qVixopYsWcJTYr0EDy70HhEREXrwwQf14osvKjQ01DoOAC/Ag/LcpGrVqqpbt651DPwXDy70HidOnNCQIUMoIgAKcJnGTS49JXbu3LkKDAy0juPzfve732nIkCHatm0bDy409vDDD2v9+vWUdQAFuEzjJjwl1rvw4ELv8corrygxMVEdOnQoshg+88wzRskAWKGMuAlPiQWKFhkZecV9DodD+/bt82AaAN6AMgIAAEwxgdUQPdC9lixZcs3HHjp0SJs3b3ZjGgDAlVBGSlCjRo20aNEinTt37qrH7d69W08++aQmTpzooWS+acaMGWrYsKEmTpyonTt3FtqfnZ2tlStXqmvXroqJidHJkycNUvqGCRMmKCcn55qO/eyzz7RixQo3JwLgTVhNU4KmTZum4cOH66mnnlJ8fLxiY2NVq1YtBQQE6LvvvtOOHTu0adMm7dixQwMHDtSAAQOsI5dqGzZs0AcffKCpU6fq+eefV1BQkEJDQwvOx9GjR1WjRg317t1b27dvV0hIiHXkUmvHjh2qU6eOOnfurI4dOyo2NlY1atSQJF24cKHgZ2PhwoXKzMzU/PnzjRMD8CTmjLhBSkqKli5dqo0bN+rAgQM6e/asqlevrujoaLVt21bdunVT5cqVrWP6lBMnTmjTpk2Fzkd0dPRVV9qg5KSnp2vatGlatmyZsrOzVaZMGfn7+xeMmERHR+uJJ55Qz5495e/vb5wWgCdRRgB4lNPpVHp6uksxbN68uapXr24dDYARyggAADDF+DQAADBFGQEAAKYoIwAAwBRlBAAAmOI+I26Un5+vPXv2KCsrS/n5+S77WrVqZZTKN+Xl5WnevHlau3Ztkedj3bp1RskAAJQRN/n000/VtWtXHTx4sNBt33lKrOcNGjRI8+bNU4cOHdSkSRM5HA7rSD7rhx9+0IQJE65YDHlQHuB7KCNukpCQoNjYWK1YsUI1a9bkl5+xJUuW6J133lH79u2to/i8fv36acOGDerevTs/GwAkcZ8RtwkKCtLWrVtVr1496yiQVKtWLX388ce65ZZbrKP4vMqVK2vFihW6++67raMA8BJMYHWTO++8U3v27LGOgf969tlnNXnyZJ6U7AWqVKmiqlWrWscA4EUYGSlB6enpBX/eu3evXnjhBf3pT39S06ZNVa5cOZdjmzVr5ul4Pufhhx92eb1u3TpVrVpVt956a6HzkZSU5MloPm3hwoX617/+pbfeekuBgYHWcQB4AcpICfLz85PD4bjiv74v7WMCq2f07t37mo+dO3euG5MgOjraZW7Inj175HQ6FRERUagYfvXVV56OB8AYE1hL0P79+60j4CcoGN6jU6dO1hEAeDFGRtxk48aNiouLU9myrn3vwoULSklJ4T4jHnbfffcpKSlJlStXdtl+6tQpderUifuMAIAhyoiblClTRpmZmQoJCXHZfuLECYWEhHCZxsP8/Px09OjRQucjKytLtWvX1vnz542S+Z6oqCh98cUXqlatmsv277//Xi1atOA+I4AP4jKNm1yaG3K5EydOKCgoyCCRb/rppOIdO3bo6NGjBa/z8vK0atUq1a5d2yKazzpw4ECRZTw3N1f/+c9/DBIBsEYZKWGXVnA4HA716tVL/v7+Bfvy8vKUnp6uuLg4q3g+p3nz5nI4HHI4HLrvvvsK7a9QoYKmTp1qkMz3vPfeewV/Xr16tYKDgwte5+Xlae3atYqMjLSIBsAYZaSEXfofrNPpVMWKFVWhQoWCfeXLl9ddd92lxx9/3Cqez9m/f7+cTqeioqL0+eefq0aNGgX7ypcvr5CQEJUpU8Ywoe+4NInV4XCoZ8+eLvvKlSuniIgIvfbaawbJAFhjzoibjBkzRsOGDeOSDHCZyMhIffHFF6pevbp1FABegjICn7Fr1y5NnTpVO3fulMPhUMOGDTVw4EA1bNjQOhoA+DQu07jJ5Td5usThcCggIED16tVTr1691Lp1a4N0vmf58uV67LHHFBsbq5YtW0q6+GTlpk2batGiRercubNxQt8xZcqUIrf/9GejVatWXD4DfAgjI24ycuRIzZgxQ02bNtUdd9whp9OpL7/8Uunp6erVq5d27NihtWvXKikpSQ899JB13FIvKipK3bp109ixY122jx49WgsWLGA5qQdFRkbq22+/VU5OjqpUqSKn06nvv/9egYGBuummm5SVlaWoqCitX79eYWFh1nEBeABlxE0ef/xx1alTRy+++KLL9pdfflkHDx7U//7v/2r06NFasWKFvvzyS6OUviMwMFDp6emFnqK8e/du3XbbbcrJyTFK5nsWL16sWbNmafbs2apbt66ki7eH79+/v5544gndfffd+p//+R/96le/0vLly43TAvAEyoibBAcHKzU1tdAvvz179igmJkbZ2dn6+uuvdfvtt+v06dNGKX1H+/bt1blz50LPq5k7d66WLFmi1atXGyXzPXXr1tW7776r5s2bu2xPS0vTI488on379iklJUWPPPKIMjMzbUIC8CjmjLhJQECAUlJSCpWRlJQUBQQESJLy8/Nd7kMC9+nYsaOGDx+u1NRU3XXXXZIuzhlZtmyZxowZ43IPjI4dO1rF9AmZmZm6cOFCoe0XLlwouCldrVq1KOmAD6GMuMnTTz+thIQEpaam6vbbb5fD4dDnn3+u2bNn6/nnn5d08cZP0dHRxkl9w4ABAyRJ06dP1/Tp04vcJ4knKntA69at1b9/f82ePbvg739aWpqefPLJghvTbdu2jRugAT6EyzRu9Pbbb+vvf/+7du3aJUlq0KCBnn76aXXt2lWSdPbs2YIVBICvOHr0qLp37661a9eqXLlyki6Oitx///1asGCBQkNDtX79ep0/f17x8fHGaQF4AmUEPufHH3+kAHqBr7/+Wt98842cTqcaNmyoBg0aWEcCYIQy4mbnzp1TVlaW8vPzXbbXqVPHKJFvysvL07hx4zRz5kwdO3ZM33zzjaKiovTiiy8qIiJCffv2tY4IAD6LOSNusnv3bvXp00cpKSku2y89zZd5CZ71yiuv6K233tKrr77q8mygpk2b6vXXX6eMeFBeXp7mzZuntWvXFlnU161bZ5QMgBXKiJv06tVLZcuW1QcffKCaNWsWeTdWeM78+fM1a9Ys3X///UpISCjY3qxZM3399deGyXzPoEGDNG/ePHXo0EFNmjThZwMAZcRdtmzZotTUVJ574iUOHz5caJm1dHF59fnz5w0S+a4lS5bonXfeUfv27a2jAPASftYBSqvGjRvr+PHj1jHwX7feequSk5MLbV+2bBnLqz2sfPnyRRZDAL6LkRE3mThxop577jmNGzdOTZs2LVjCeEmlSpWMkvmm0aNHq3v37jp8+LDy8/OVlJSkXbt2af78+frggw+s4/mUZ599VpMnT9bf//53LtEAkMRqGrfx87s46HT5/2yZwGpn9erVGjdunFJTU5Wfn68WLVroz3/+M/ey8LDf//73Wr9+vapWrapbb721UFFPSkoySgbACiMjbrJ+/XrrCLhM27Zt1bZtW+sYPq9y5cr6/e9/bx0DgBdhZAQ+wel0KjU1VQcOHJDD4VBUVJSaN2/OZQIA8AJMYHWj5ORkdevWTXFxcTp8+LAkacGCBdq0aZNxMt+yfv161a1bV3feeaceffRRde7cWbGxsapfv742btxoHc8nXbhwQf/3f/+nN954o+CBeEeOHNGZM2eMkwGwQBlxk3fffVdt27ZVhQoV9NVXXyk3N1eSdPr0aY0bN844ne/Ys2ePHnzwQUVERCgpKUk7d+7Ujh07tGzZMt18881q37699u3bZx3Tpxw8eFBNmzbVQw89pKeeekrffvutJOnVV1/VsGHDjNMBsMBlGjeJjo7WkCFD1KNHD1WsWFFbt25VVFSUtmzZot/+9rcFj0qHew0cOFA7d+7U2rVrC+1zOp164IEH1LhxY02dOtUgnW/q1KmTKlasqDfffFPVqlUr+NnYsGGD+vXrp927d1tHBOBhjIy4ya5du9SqVatC2ytVqqTvv//e84F81Mcff6zBgwcXuc/hcGjw4MFMNvawTZs26YUXXlD58uVdtoeHhxdczgTgWygjblKzZk3t2bOn0PZNmzYpKirKIJFvysjIUNOmTa+4v0mTJjp48KAHEyE/P7/Ipe3/+c9/VLFiRYNEAKxRRtykf//+GjRokD777DM5HA4dOXJEb7/9toYNG6YBAwZYx/MZZ86cUWBg4BX3BwYGKicnx4OJ0KZNGyUmJha8djgcOnPmjEaPHs0t4gEfxZwRNxo1apRef/11/fjjj5Ikf39/DRs2TH/5y1+Mk/kOPz8/rVu3TlWrVi1y//Hjx9WmTRtuQudBR44cUevWrVWmTBnt3r1bsbGx2r17t6pXr66NGzcqJCTEOiIAD6OMuFlOTo527Nih/Px8NW7cWP7+/srMzFSdOnWso/kEPz8/ORwOFfXX/NJ27ojreWfPntWSJUtc7ob7xz/+URUqVLCOBsAAZcTDtm7dqhYtWvDLz0OudT5IeHi4m5Pg5+zdu1ePP/641q1bZx0FgIdxO3iUapSMG8eZM2e0YcMG6xgADDCBFQAAmKKMAAAAU1ymKWHp6elX3b9r1y4PJQEA4MbABNYSxuoNoGjR0dFXfUpyTk6Odu/ezc8G4IMYGSlh+/fvt44AeKVOnTpZRwDgpRgZgU84duyYhg0bprVr1yorK6vQyBX/GgcAO4yMwCf06tVLGRkZevHFF1WzZs2rXi4AAHgWIyPwCRUrVlRycrKaN29uHQUAcBmW9sInhIWFFTmpGABgjzICn5CYmKgRI0bowIED1lEAAJfhMg18QpUqVZSTk6MLFy4oMDBQ5cqVc9l/8uRJo2S+5fz584qPj9cbb7yhW265xToOAC/BBFY3YfWGd0lMTLSOAEnlypXT9u3bmUAMwAUjI27Srl07ZWRkaODAgUWu3njooYeMkgG2nn32WZUrV04TJkywjgLASzAy4iabNm1i9YaX2bt3r+bOnau9e/dq8uTJCgkJ0apVqxQWFqZbb73VOp7POHfunGbPnq01a9YoNjZWQUFBLvsnTZpklAyAFcqIm7B6w7ts2LBB7dq10913362NGzfqlVdeUUhIiNLT0zV79mwtX77cOqLP2L59u1q0aCFJ+uabb1z2cfkG8E1cpnGTjz76SK+99preeOMNRUREWMfxeS1btlTnzp01dOhQVaxYUVu3blVUVJS++OILderUSYcPH7aOCAA+izLiJqze8C433XSTtm3bpsjISJcycuDAATVs2FA//vijdUSfs2fPHu3du1etWrVShQoVCh4iCcD3cJnGTVi94V0qV66szMxMRUZGumxPS0tT7dq1jVL5phMnTujRRx/V+vXr5XA4tHv3bkVFRalfv36qXLmyXnvtNeuIADyMMuImPXv2tI6An+jatauGDx+uZcuWyeFwKD8/X5s3b9awYcPUo0cP63g+ZciQISpXrpwyMjLUqFGjgu1dunTRkCFDKCOAD+IOrG60d+9evfDCC3rssceUlZUlSVq1apX+/e9/GyfzPa+88orq1Kmj2rVr68yZM2rcuLFatWqluLg4vfDCC9bxfMpHH32kiRMn6uabb3bZXr9+fR08eNAoFQBLlBE32bBhg5o2barPPvtMSUlJOnPmjCQpPT1do0ePNk7ne8qVK6e3335b33zzjd555x0tXLhQX3/9tRYsWKAyZcpYx/MpP/zwgwIDAwttP378uPz9/Q0SAbBGGXGTESNG6OWXX9aaNWtUvnz5gu2tW7fWJ598YpjMt9WtW1d/+MMf9Oijj6p+/frWcXxSq1atNH/+/ILXly6b/fWvf1Xr1q0NkwGwwpwRN9m2bZsWLVpUaHuNGjV04sQJg0S+zel0avny5Vq/fr2ysrKUn5/vsj8pKckome/561//qt/85jf68ssvde7cOT333HP697//rZMnT2rz5s3W8QAYYGTETS6t3rgcqzdsDBo0SN27d9f+/ft10003KTg42OUDntO4cWOlp6frjjvuUJs2bfTDDz/o4YcfVlpamurWrWsdD4AB7jPiJs8995w++eQTLVu2TLfccou++uorHTt2TD169FCPHj2YN+JhVatW1cKFC9W+fXvrKACAy1BG3OT8+fPq1auXlixZIqfTqbJlyyovL09du3bVvHnzmDTpYZGRkfrwww/VsGFD6yiQ9N133+nNN9/Uzp075XA41KhRI/Xu3VtVq1a1jgbAAGXEzfbu3au0tDTl5+crOjqaSZNG3nrrLa1atUpz5sxRhQoVrOP4tA0bNuihhx5SpUqVFBsbK0lKTU3V999/r/fee0/33nuvcUIAnkYZgU/IycnRww8/rM2bNysiIqLQ7fm/+uoro2S+p0mTJoqLi9OMGTMKRgjz8vI0YMAAbd68Wdu3bzdOCMDTKCNuwuoN73Lp9uN/+MMfFBoaWugZKMzh8ZwKFSpoy5YtatCggcv2Xbt2qXnz5jp79qxRMgBWWNrrJoMGDdKsWbPUunXrIn/5wbNWrFih1atX69e//rV1FJ/XokUL7dy5s1AZ2blzp5o3b24TCoApyoibLFy4UElJSaze8BJhYWGqVKmSdQxIeuaZZzRo0CDt2bNHd911lyTp008/1bRp0zRhwgSlp6cXHNusWTOrmAA8iMs0bsLqDe+yYsUKTZ06VTNnzlRERIR1HJ/m53f12xs5HA45nU45HA7l5eV5KBUAS5QRN2H1hnepUqWKcnJydOHCBQUGBhaawHry5EmjZL6nOA/DCw8Pd2MSAN6CyzRu0rlzZy1evFghISGs3vACiYmJ1hHwXxQMAJejjLhJr169lJqaqm7dujGB1Qv07NnTOgJ+YteuXZo6dWrBTc8aNmyop59+utCkVgC+gcs0bhIUFMTqDWOnTp265mOZ3Oo5y5cv12OPPabY2Fi1bNlS0sUJrF988YUWLVqkzp07GycE4GmUETdp2LCh3nnnHVYDGPLz8/vZESkmSnpeVFSUunXrprFjx7psHz16tBYsWKB9+/YZJQNghTLiJqzesLdhw4ZrPpZbkHtOYGCg0tPTVa9ePZftu3fv1m233aacnByjZACsMGfETbp166acnBzVrVuX1RtGKBje6Te/+Y2Sk5MLlZFNmzbpnnvuMUoFwBJlxE1YveFdNm7ceNX9rVq18lASdOzYUcOHD1dqaqrLTc+WLVumMWPG6L333nM5FkDpx2Ua+ISibrT10/kkzBnxnJ+76dklzOUBfAcjIyWI1Rve67vvvnN5ff78eaWlpenFF1/UK6+8YpTKN13+0EgAYGSkBLF648azceNGDRkyRKmpqdZRAMBnMTJSgtavX28dAcVUo0YN7dq1yzpGqTdlyhQ98cQTCggI0JQpU6567DPPPOOhVAC8BSMj8Ak/fRKsdHGEKjMzUxMmTND58+e1efNmo2S+ITIyUl9++aWqVaumyMjIKx7ncDi4zwjggygjbsLqDe9y6RLa5X/d77rrLs2ZM4enKwOAIcqIm7B6w7tc/qRYPz8/1ahRQwEBAUaJAACXMGfETVi94V14UqytoUOHXvOxkyZNcmMSAN6IMuImwcHBhba1adNG/v7+rN7woHXr1mngwIH69NNPCy2nzs7OVlxcnGbOnMmdP90sLS3tmo7j6daAb+IyjYft3LlTt99+u86cOWMdxSd07NhRrVu31pAhQ4rcP2XKFK1fv17/+Mc/PJwMAHAJZcRNWL3hHcLDw7Vq1So1atSoyP1ff/214uPjlZGR4eFkvmffvn2KjIxk9ANAIVymcZPmzZtfdfUGPOPYsWOFHlL4U2XLltW3337rwUS+q379+srMzFRISIgkqUuXLpoyZYpCQ0ONkwGwRhlxk/3797u8ZvWGjdq1a2vbtm2FnhB7SXp6umrWrOnhVL7p8mK+cuVKjR8/3igNAG9CGXETVm94h/bt2+vPf/6z2rVrV6gInj17VqNHj9aDDz5olA4AIEnX9vhMXLN169apcePGRT40Lzs7W7feequSk5MNkvmmF154QSdPntQtt9yiV199Vf/617/03nvvaeLEiWrQoIFOnjypUaNGWcf0CQ6Ho9B8EeaPAJCYwFriWL3hfQ4ePKgnn3xSq1evLrhU4HA41LZtW02fPl0RERG2AX2En5+f2rVrJ39/f0nS+++/r/vuu09BQUEuxyUlJVnEA2CIMlLCWL3hvb777jvt2bNHTqdT9evXV5UqVawj+ZTevXtf03Fz5851cxIA3oYyUsICAgK0ffv2K06Y3LNnj5o2baqzZ896OBkAAN6JOSMl7NLqjSth9QYAAK4oIyXs0uqNH3/8sdA+Vm8AAFAYl2lK2LFjx9SiRQuVKVNGAwcOVIMGDeRwOLRz505NmzZNeXl5+uqrr7jREwAA/0UZcQNWbwAAcO0oI27E6g0AAH4eZQQAAJhiAisAADBFGQEAAKYoIwAAwBRlBAAAmKKMAAAAU5QRAABgijICAABM/T+VtmIKeU7C0gAAAABJRU5ErkJggg==\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -676,16 +676,16 @@ "execution_count": 11, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:11.672076Z", - "iopub.status.busy": "2022-12-01T10:52:11.671985Z", - "iopub.status.idle": "2022-12-01T10:52:11.875119Z", - "shell.execute_reply": "2022-12-01T10:52:11.874778Z" + "iopub.execute_input": "2022-12-02T12:08:54.740887Z", + "iopub.status.busy": "2022-12-02T12:08:54.740803Z", + "iopub.status.idle": "2022-12-02T12:08:54.939549Z", + "shell.execute_reply": "2022-12-02T12:08:54.939228Z" } }, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -710,10 +710,10 @@ "execution_count": 12, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:11.877091Z", - "iopub.status.busy": "2022-12-01T10:52:11.876979Z", - "iopub.status.idle": "2022-12-01T10:52:11.975515Z", - "shell.execute_reply": "2022-12-01T10:52:11.975208Z" + "iopub.execute_input": "2022-12-02T12:08:54.941187Z", + "iopub.status.busy": "2022-12-02T12:08:54.941095Z", + "iopub.status.idle": "2022-12-02T12:08:55.034453Z", + "shell.execute_reply": "2022-12-02T12:08:55.034159Z" } }, "outputs": [ @@ -749,10 +749,10 @@ "execution_count": 13, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:11.977474Z", - "iopub.status.busy": "2022-12-01T10:52:11.977332Z", - "iopub.status.idle": "2022-12-01T10:52:27.194767Z", - "shell.execute_reply": "2022-12-01T10:52:27.194508Z" + "iopub.execute_input": "2022-12-02T12:08:55.036168Z", + "iopub.status.busy": "2022-12-02T12:08:55.036077Z", + "iopub.status.idle": "2022-12-02T12:08:56.328132Z", + "shell.execute_reply": "2022-12-02T12:08:56.327857Z" }, "lines_to_next_cell": 2 }, @@ -760,7 +760,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 13, @@ -783,10 +783,10 @@ "execution_count": 14, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:27.196723Z", - "iopub.status.busy": "2022-12-01T10:52:27.196629Z", - "iopub.status.idle": "2022-12-01T10:52:27.207429Z", - "shell.execute_reply": "2022-12-01T10:52:27.207178Z" + "iopub.execute_input": "2022-12-02T12:08:56.329679Z", + "iopub.status.busy": "2022-12-02T12:08:56.329593Z", + "iopub.status.idle": "2022-12-02T12:08:56.340857Z", + "shell.execute_reply": "2022-12-02T12:08:56.340459Z" } }, "outputs": [], @@ -799,10 +799,10 @@ "execution_count": 15, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:27.208834Z", - "iopub.status.busy": "2022-12-01T10:52:27.208769Z", - "iopub.status.idle": "2022-12-01T10:52:27.215624Z", - "shell.execute_reply": "2022-12-01T10:52:27.215316Z" + "iopub.execute_input": "2022-12-02T12:08:56.342520Z", + "iopub.status.busy": "2022-12-02T12:08:56.342426Z", + "iopub.status.idle": "2022-12-02T12:08:56.347698Z", + "shell.execute_reply": "2022-12-02T12:08:56.347463Z" } }, "outputs": [ @@ -858,10 +858,10 @@ "execution_count": 16, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:27.219210Z", - "iopub.status.busy": "2022-12-01T10:52:27.219032Z", - "iopub.status.idle": "2022-12-01T10:52:27.221526Z", - "shell.execute_reply": "2022-12-01T10:52:27.221296Z" + "iopub.execute_input": "2022-12-02T12:08:56.349917Z", + "iopub.status.busy": "2022-12-02T12:08:56.349845Z", + "iopub.status.idle": "2022-12-02T12:08:56.352086Z", + "shell.execute_reply": "2022-12-02T12:08:56.351861Z" } }, "outputs": [ @@ -869,7 +869,7 @@ "data": { "text/html": [ "\n", - "
\n", + "
\n", "
\n", " Visualization omitted, Javascript library not loaded!
\n", " Have you run `initjs()` in this notebook? If this notebook was from another\n", @@ -879,13 +879,13 @@ "
\n", " " ], "text/plain": [ - "" + "" ] }, "execution_count": 16, @@ -902,10 +902,10 @@ "execution_count": 17, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:27.222831Z", - "iopub.status.busy": "2022-12-01T10:52:27.222760Z", - "iopub.status.idle": "2022-12-01T10:52:27.230459Z", - "shell.execute_reply": "2022-12-01T10:52:27.230169Z" + "iopub.execute_input": "2022-12-02T12:08:56.353510Z", + "iopub.status.busy": "2022-12-02T12:08:56.353319Z", + "iopub.status.idle": "2022-12-02T12:08:56.360716Z", + "shell.execute_reply": "2022-12-02T12:08:56.360485Z" } }, "outputs": [ @@ -913,7 +913,7 @@ "data": { "text/html": [ "\n", - "
\n", + "
\n", "
\n", " Visualization omitted, Javascript library not loaded!
\n", " Have you run `initjs()` in this notebook? If this notebook was from another\n", @@ -923,13 +923,13 @@ "
\n", " " ], "text/plain": [ - "" + "" ] }, "execution_count": 17, @@ -963,10 +963,10 @@ "execution_count": 18, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:27.231920Z", - "iopub.status.busy": "2022-12-01T10:52:27.231833Z", - "iopub.status.idle": "2022-12-01T10:52:27.242427Z", - "shell.execute_reply": "2022-12-01T10:52:27.242170Z" + "iopub.execute_input": "2022-12-02T12:08:56.362233Z", + "iopub.status.busy": "2022-12-02T12:08:56.362156Z", + "iopub.status.idle": "2022-12-02T12:08:56.379039Z", + "shell.execute_reply": "2022-12-02T12:08:56.378748Z" } }, "outputs": [ @@ -1013,10 +1013,10 @@ "execution_count": 19, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:27.244012Z", - "iopub.status.busy": "2022-12-01T10:52:27.243922Z", - "iopub.status.idle": "2022-12-01T10:52:27.250432Z", - "shell.execute_reply": "2022-12-01T10:52:27.250219Z" + "iopub.execute_input": "2022-12-02T12:08:56.380654Z", + "iopub.status.busy": "2022-12-02T12:08:56.380556Z", + "iopub.status.idle": "2022-12-02T12:08:56.405525Z", + "shell.execute_reply": "2022-12-02T12:08:56.405291Z" } }, "outputs": [ @@ -1024,7 +1024,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "tensor([[0.6251, 0.5288]], grad_fn=)\n" + "tensor([[0.5559, 0.4439]], grad_fn=)\n" ] } ], diff --git a/notebooks/6 - Ablation Study.ipynb b/notebooks/6 - Ablation Study.ipynb index 6335877..ffe25e5 100644 --- a/notebooks/6 - Ablation Study.ipynb +++ b/notebooks/6 - Ablation Study.ipynb @@ -18,10 +18,10 @@ "execution_count": 1, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:28.809492Z", - "iopub.status.busy": "2022-12-01T10:52:28.809190Z", - "iopub.status.idle": "2022-12-01T10:52:28.818495Z", - "shell.execute_reply": "2022-12-01T10:52:28.817937Z" + "iopub.execute_input": "2022-12-02T12:08:57.884529Z", + "iopub.status.busy": "2022-12-02T12:08:57.883667Z", + "iopub.status.idle": "2022-12-02T12:08:57.896964Z", + "shell.execute_reply": "2022-12-02T12:08:57.896266Z" } }, "outputs": [], @@ -36,10 +36,10 @@ "id": "54158e1d", "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:28.821686Z", - "iopub.status.busy": "2022-12-01T10:52:28.821517Z", - "iopub.status.idle": "2022-12-01T10:52:28.824654Z", - "shell.execute_reply": "2022-12-01T10:52:28.824025Z" + "iopub.execute_input": "2022-12-02T12:08:57.901227Z", + "iopub.status.busy": "2022-12-02T12:08:57.900930Z", + "iopub.status.idle": "2022-12-02T12:08:57.905012Z", + "shell.execute_reply": "2022-12-02T12:08:57.904424Z" } }, "outputs": [], @@ -55,10 +55,10 @@ "execution_count": 3, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:28.827503Z", - "iopub.status.busy": "2022-12-01T10:52:28.827391Z", - "iopub.status.idle": "2022-12-01T10:52:29.065392Z", - "shell.execute_reply": "2022-12-01T10:52:29.065132Z" + "iopub.execute_input": "2022-12-02T12:08:57.908137Z", + "iopub.status.busy": "2022-12-02T12:08:57.907908Z", + "iopub.status.idle": "2022-12-02T12:08:58.127284Z", + "shell.execute_reply": "2022-12-02T12:08:58.126987Z" } }, "outputs": [ @@ -167,10 +167,10 @@ "execution_count": 4, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:29.067132Z", - "iopub.status.busy": "2022-12-01T10:52:29.066958Z", - "iopub.status.idle": "2022-12-01T10:52:29.348719Z", - "shell.execute_reply": "2022-12-01T10:52:29.348443Z" + "iopub.execute_input": "2022-12-02T12:08:58.128973Z", + "iopub.status.busy": "2022-12-02T12:08:58.128887Z", + "iopub.status.idle": "2022-12-02T12:08:58.324669Z", + "shell.execute_reply": "2022-12-02T12:08:58.324396Z" } }, "outputs": [], @@ -189,10 +189,10 @@ "execution_count": 5, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:29.350481Z", - "iopub.status.busy": "2022-12-01T10:52:29.350392Z", - "iopub.status.idle": "2022-12-01T10:52:29.377711Z", - "shell.execute_reply": "2022-12-01T10:52:29.377458Z" + "iopub.execute_input": "2022-12-02T12:08:58.326285Z", + "iopub.status.busy": "2022-12-02T12:08:58.326228Z", + "iopub.status.idle": "2022-12-02T12:08:58.344846Z", + "shell.execute_reply": "2022-12-02T12:08:58.344619Z" } }, "outputs": [ @@ -267,10 +267,10 @@ "execution_count": 6, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:29.379233Z", - "iopub.status.busy": "2022-12-01T10:52:29.379088Z", - "iopub.status.idle": "2022-12-01T10:52:29.416146Z", - "shell.execute_reply": "2022-12-01T10:52:29.415812Z" + "iopub.execute_input": "2022-12-02T12:08:58.346236Z", + "iopub.status.busy": "2022-12-02T12:08:58.346178Z", + "iopub.status.idle": "2022-12-02T12:08:58.377786Z", + "shell.execute_reply": "2022-12-02T12:08:58.377546Z" }, "lines_to_next_cell": 2 }, @@ -291,10 +291,10 @@ "execution_count": 7, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:29.418133Z", - "iopub.status.busy": "2022-12-01T10:52:29.417890Z", - "iopub.status.idle": "2022-12-01T10:52:29.422758Z", - "shell.execute_reply": "2022-12-01T10:52:29.422501Z" + "iopub.execute_input": "2022-12-02T12:08:58.379266Z", + "iopub.status.busy": "2022-12-02T12:08:58.379210Z", + "iopub.status.idle": "2022-12-02T12:08:58.382663Z", + "shell.execute_reply": "2022-12-02T12:08:58.382425Z" } }, "outputs": [ @@ -354,10 +354,10 @@ "execution_count": 8, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:29.424333Z", - "iopub.status.busy": "2022-12-01T10:52:29.424259Z", - "iopub.status.idle": "2022-12-01T10:52:29.448479Z", - "shell.execute_reply": "2022-12-01T10:52:29.448178Z" + "iopub.execute_input": "2022-12-02T12:08:58.383972Z", + "iopub.status.busy": "2022-12-02T12:08:58.383919Z", + "iopub.status.idle": "2022-12-02T12:08:58.402855Z", + "shell.execute_reply": "2022-12-02T12:08:58.402577Z" } }, "outputs": [], @@ -387,10 +387,10 @@ "execution_count": 9, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:29.450405Z", - "iopub.status.busy": "2022-12-01T10:52:29.450221Z", - "iopub.status.idle": "2022-12-01T10:52:29.453466Z", - "shell.execute_reply": "2022-12-01T10:52:29.453036Z" + "iopub.execute_input": "2022-12-02T12:08:58.404336Z", + "iopub.status.busy": "2022-12-02T12:08:58.404269Z", + "iopub.status.idle": "2022-12-02T12:08:58.407392Z", + "shell.execute_reply": "2022-12-02T12:08:58.407106Z" } }, "outputs": [ @@ -454,10 +454,10 @@ "execution_count": 10, "metadata": { "execution": { - "iopub.execute_input": "2022-12-01T10:52:29.455424Z", - "iopub.status.busy": "2022-12-01T10:52:29.455348Z", - "iopub.status.idle": "2022-12-01T10:52:29.494215Z", - "shell.execute_reply": "2022-12-01T10:52:29.493877Z" + "iopub.execute_input": "2022-12-02T12:08:58.409104Z", + "iopub.status.busy": "2022-12-02T12:08:58.409017Z", + "iopub.status.idle": "2022-12-02T12:08:58.441788Z", + "shell.execute_reply": "2022-12-02T12:08:58.441562Z" } }, "outputs": [ diff --git a/python_scripts/1 - Model Evaluation.py b/python_scripts/1 - Model Evaluation.py index 969fc55..0538e77 100644 --- a/python_scripts/1 - Model Evaluation.py +++ b/python_scripts/1 - Model Evaluation.py @@ -8,7 +8,7 @@ # format_version: '1.3' # jupytext_version: 1.14.1 # kernelspec: -# display_name: Python 3.10.8 ('pydata-global-2022-ml-repro') +# display_name: Python 3 (ipykernel) # language: python # name: python3 # --- @@ -24,13 +24,13 @@ # # So we’ll go into some methods to properly evaluate machine learning models even when our data is not “independent and identically distributed”. -# %% +# %% tags=[] from pathlib import Path DATA_FOLDER = Path("..") / "data" DATA_FILEPATH = DATA_FOLDER / "penguins_clean.csv" -# %% +# %% tags=[] import pandas as pd penguins = pd.read_csv(DATA_FILEPATH) penguins.head() @@ -48,7 +48,7 @@ #
# Tip: The i.i.d. assumption lies at the core of most machine learning and is an important concept to dive into and understand.
-# %% +# %% tags=[] from sklearn.model_selection import train_test_split num_features = ["Culmen Length (mm)", "Culmen Depth (mm)", "Flipper Length (mm)"] cat_features = ["Sex"] @@ -63,28 +63,43 @@ # Usually, our target class or another feature we use isn't distributed equally. # %% +from matplotlib import pyplot as plt + +# %% tags=[] penguins.groupby("Species").Sex.count().plot(kind="bar") +plt.show() # %% [markdown] # In this case it's not very extreme. We have around twice as many Adelie than Chinstrap penguins. # # However, this can mean that we accidentally have almost no Chinstrap penguins in our training data, as it randomly overselects Adelie penguins. -# %% +# %% tags=[] y_train.reset_index().groupby(["Species"]).count() # %% [markdown] -# We can address this by applying stratification. -# That is simply sampling randomly within a class (or strata) rather than randomly sampling from the entire dataframe. +# We can address this by applying **stratification**. +# +# That is simply achieved by randomly sampling *within a class** (or strata) rather than randomly sampling from the entire dataframe. + +# %% tags=[] +X, y = penguins[features], penguins[target[0]] +X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=.7, random_state=42, stratify=y) + +# %% [markdown] +# To qualitatevely assess the effect of stratification, let's plot class distribution in both _training_ and _test_ sets: # %% -X_train, X_test, y_train, y_test = train_test_split(penguins[features], penguins[target[0]], train_size=.7, random_state=42, stratify=penguins[target[0]]) -y_train.reset_index().groupby("Species").count().plot(kind="bar") +fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 8)) + +y_train.reset_index().groupby("Species").count().plot(kind="bar", ax=ax1, ylim=(0, len(y)), title="Training") +y_test.reset_index().groupby("Species").count().plot(kind="bar", ax=ax2, ylim=(0, len(y)), title="Test") +plt.show() # %% [markdown] # Let's quickly train a model to evaluate -# %% +# %% tags=[] from sklearn.svm import SVC from sklearn.compose import ColumnTransformer from sklearn.pipeline import Pipeline @@ -108,6 +123,7 @@ # That changes however, when we have minority classes with much less data than the majority class. # # Either way it's worth it to keep in mind that stratification exists. The `stratify=` keyword takes any type of vector as long as it matches the dimension of the dataframe. +# # ## Cross-Validation # Cross-validation is often considered the gold standard in statistical applications and machine learning. # @@ -118,17 +134,174 @@ # # Cross-validation is particularly useful when we don't have a lot of data or the data is highly heterogeneous. -# %% +# %% tags=[] from sklearn.model_selection import cross_val_score + scores = cross_val_score(model, X_train, y_train, cv=5) scores -# %% +# %% tags=[] print(f"{scores.mean():0.2f} accuracy with a standard deviation of {scores.std():0.2f}") # %% [markdown] # Now we know there are some folds this support-vector machine will do exceptional on and others it does quite well on only getting a few samples wrong. +# %% [markdown] +# ## Model Evaluation + +# %% [markdown] +# Brilliant! So let's recap for a moment what we have done so far, in preparation for our (final) **Model evaluation**. +# +# We have: +# +# - prepared the model pipeline: `sklearn.pipeline.Pipeline` with `preprocessor + model` +# - generated **train** and **test** data partitions (with stratification): `(X_train, y_train)` and `(X_test, y_test)`, respectively +# - stratification guaranteed that those partitions will retain class distributions +# - assessed model performance via **cross validation** (i.e. `cross_val_score`) on `X_train`(!!) +# - this had the objective of verifying model consistency on multiple data partitioning +# +# Now we need the complete our last step, namely "assess how the model we chose in CV" (we only had one model, so that was an easy choice :D ) will perform on _future data_! +# And we have a _candidate_ as representative for these data: `X_test`. +# +# Please note that `X_test` has never been used so far (as it should have!). The take away message here is: _generate test partition, and forget about it until the last step!_ +# +# +# Thanks to `CV`, We have an indication of how the `SVC` classifier behaves on multiple "version" of the training set. We calculated an average score of `0.99` accuracy, therefore we decided this model is to be trusted for predictions on _unseen data_. +# +# Now all we need to do, is to prove this assertion. +# +# To do so we need to: +# - train a new model on the entire **training set** +# - evaluate it's performance on **test set** (using the metric of choice - presumably the same metric we chose in CV!) + +# %% +# training +model = Pipeline(steps=[ + ('preprocessor', preprocessor), + ('classifier', SVC()), +]) +classifier = model.fit(X_train, y_train) + +# %% +# Model evaluation +from sklearn.metrics import accuracy_score + +y_pred = classifier.predict(X_test) +print("TEST ACC: ", accuracy_score(y_true=y_test, y_pred=y_pred)) + +# %% [markdown] +# Now we can finally say that we have concluded our model evaluation - with a fantastic score of `0.96` Accuracy on the test set. + +# %% [markdown] +# ## Choosing the appropriate Evaluation Metric + +# %% [markdown] +# Ok, now for the mere sake of considering a more realistic data scenario, let's pretend our reference dataset is composed by only samples from two (out of the three) classes we have. In particular, we will crafting our dataset by choosing the most and the least represented classes, respectively. +# +# The very idea is to explore whether the choice of appropriate metrics could make the difference in our machine learning models evaluation. + +# %% [markdown] +# Let's recall class distributions in our dataset: + +# %% +y.reset_index().groupby(["Species"]).count() + +# %% [markdown] +# So let's select samples from the first two classes, `Adelie Penguin` and `Chinstrap penguin`: + +# %% +samples = penguins[((penguins["Species"].str.startswith("Adelie")) | (penguins["Species"].str.startswith("Chinstrap")))] + +# %% +samples.shape[0] == 146 + 68 # quick verification + +# %% [markdown] +# To make things even harder for our machine learning model, let's also see if we could get rid of _clearly_ separating features in this toy dataset + +# %% +import seaborn as sns + +pairplot_figure = sns.pairplot(samples, hue="Species") + +# %% [markdown] +# OK so if we get to choose, we could definitely say that in this dataset, the `Flipper Length` in combination with the `Culmen Depth` leads to the hardest classification task for our machine learning model. +# +# Therefore, here is the plan: +# - we select only those to numerical features (_iow_ we will get rid of the `Culmen Lenght` feature) +# - we will apply an identical _Model evaluation_ pipeline as we did in our previous example +# - Cross Validation + Evaluation on Test set +# +# The very difference this time is that we will use multiple metrics to evaluate our model to prove our point on _carefully selecting evaluation metrics_. + +# %% +num_features = ["Culmen Length (mm)", "Culmen Depth (mm)", "Flipper Length (mm)"] +selected_num_features = num_features[1:] +cat_features = ["Sex"] +features = selected_num_features + cat_features + +# %% +num_transformer = StandardScaler() +cat_transformer = OneHotEncoder(handle_unknown='ignore') + +preprocessor = ColumnTransformer(transformers=[ + ('num', num_transformer, selected_num_features), # note here, we will only preprocess selected numerical features + ('cat', cat_transformer, cat_features) +]) + +model = Pipeline(steps=[ + ('preprocessor', preprocessor), + ('classifier', SVC()), +]) + +# %% +X, y = samples[features], samples[target[0]] +X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=.7, random_state=42, stratify=y) # we also stratify on classes + +# %% +y_train.reset_index().groupby("Species").count() + +# %% +y_test.reset_index().groupby("Species").count() + +# %% [markdown] +# In our evaluation pipeline we will be using keep record both **accuracy** (`ACC`) and **matthew correlation coefficient** (`MCC`) + +# %% +from sklearn.model_selection import cross_validate +from sklearn.metrics import make_scorer +from sklearn.metrics import matthews_corrcoef as mcc +from sklearn.metrics import accuracy_score as acc + +mcc_scorer = make_scorer(mcc) +acc_scorer = make_scorer(acc) +scores = cross_validate(model, X_train, y_train, cv=5, + scoring={"MCC": mcc_scorer, "ACC": acc_scorer}) +scores + +# %% +import numpy as np + +print("Avg ACC in CV: ", np.average(scores["test_ACC"])) +print("Avg MCC in CV: ", np.average(scores["test_MCC"])) + +# %% +model = model.fit(X_train, y_train) + +print("ACC: ", acc_scorer(model, X_test, y_test)) +print("MCC: ", mcc_scorer(model, X_test, y_test)) + +# %% [markdown] +# To see exactly what happened, let's have a look at the **Confusion matrix** + +# %% +from sklearn.metrics import ConfusionMatrixDisplay +fig, ax = plt.subplots(figsize=(15, 10)) +ConfusionMatrixDisplay.from_estimator(model, X_test, y_test, ax=ax) +plt.show() + +# %% [markdown] +# As expected, the model did a pretty bad job in classifying *Chinstrap Penguins* and the `MCC` was able to catch that, whilst `ACC` could not as it only considers correctly classified samples! + # %% [markdown] # ## Time-series Validation # diff --git a/rendered_notebooks/0 - Basic Data Prep and Model.html b/rendered_notebooks/0 - Basic Data Prep and Model.html index eeec39a..f6110de 100644 --- a/rendered_notebooks/0 - Basic Data Prep and Model.html +++ b/rendered_notebooks/0 - Basic Data Prep and Model.html @@ -15300,39 +15300,39 @@

Machine LearningMachine LearningMachine LearningModel Training -
0.9914163090128756
+
0.9871244635193133

@@ -15805,7 +15805,7 @@

Model Training -
1.0
+
0.9900990099009901
diff --git a/rendered_notebooks/1 - Model Evaluation.html b/rendered_notebooks/1 - Model Evaluation.html index b24fdd3..319bc9a 100644 --- a/rendered_notebooks/1 - Model Evaluation.html +++ b/rendered_notebooks/1 - Model Evaluation.html @@ -14896,15 +14896,32 @@

Stratification +