From b60bf0db47aa931a7e89286eda0eeb9bba8c560d Mon Sep 17 00:00:00 2001 From: Bashar Ammari <96192809+bammari@users.noreply.github.com> Date: Mon, 18 Sep 2023 10:53:43 -0400 Subject: [PATCH] Linear Tree Implementation (#108) * added initial linear-tree files. Created linear-tree model class * initial commit * LinearTreeModel * public vs private test * private vs public testing * public vs private testing * Initial Linear Tree Commit * ltmodel testing * ltmodel testing * ltmodel testing * ltmodel testing * recursion test find children splits * testing * recursion test find all children splits * recursion test number find all children splits * Recursion test 1 * Globalizing Function Test * Globalizing Function Test * Determine if parent information is even needed * LinearModel Testing * Lt Model Testing and commit * Cleaning Up LinearTreeModel Class code * Linear Tree Init Commiit * Raise exceptions for missing bounds/wrong transformation * changed parse_tree_data name * raise errors on unsupported GDP transformations * Bounds changes * Implemented output variable bound calculation function * adding comments to output variable bound function * Initial Linear Tree Documentation * linear tree documentation * Added Notebook for Linear Tree in Docs * Documentation * Added more documentation on comments * Updated docstrings * docstring updates * Docstring Updates * Docstring Updates * Upload the script for testing * Upload script for testing LinearTreeModel Test dictionary * Upload the test for bigm transformation * Pass pytest: fix some bus, e.g. len * Test hull formulation * Test the slope and bounds: Note that bounds may be None for other dataset * Added Hybrid Big-M Formulations * Added Multiple BigM Test * Add more comments * Added multivariate input testing for linear model decision trees * Added Hybrid Big M Formulation Tests * Docstring Updates * Added option to pass in summary rather than linear-tree instance * Added test to ensure model summary argument functions * Added Hybrid Big-M Tests * docstring updates * Docstring updates * Docstring updates * Added code to ensure input dict is correct * Added hybrid big-m representation to docs * Added testing for raised exceptions * Reassigned none bounds in lt model. Added unscaled_input_bounds karg * Initial formulation consolidation code commit * docstring updates * Docstring updates * Docstring update * Update Variable Names * Updated LinearTreeModel to LinearTreeDefinition, made helper functions internal, and added cbc as solver for MILP tests * Updated Notebook to use LinearTree Definition * Code Cleanup * Updated lineartree to linear_tree, changed _setup_scaled_inputs due to maxpool int/float issue * Install pyscipopt in main.yml * Update linear_tree notebook to use SCIP rather than gurobi * Changed quadratic formulation solver from gurobi to scip * Skip if solvers unavailable * omlt.lineartree to omlt.linear_tree * Added 'custom' transformation option to LinearTreeGDPFormulation * Ran through pylint and black * Cleaning up for linting * Fixing pylint issues * Linting * Linting * Added test for scaling LinearTreeDefinition * Linting * Edit docstring * Added properties to definition. Docstring Updates. * Removing unused properties * Docstring Updates * Linear Tree Notebook Updates * Linting * docstring update * For code coverage * Addressing requested changes * Addressing changes * Addressing ruth comments * Notebook Update * Updating README.rst. Also updated OMLT paper citation. * Modifying citation * Notebook modification * Notebook modification * Notebook docstring * Attempt tests on Python 3.10 * Testing on Python 3.10 --------- Co-authored-by: Shumeng Lin <54089164+linshumeng@users.noreply.github.com> --- .github/workflows/main.yml | 4 +- README.rst | 37 +- docs/api_doc/omlt.linear_tree.rst | 20 + docs/api_doc/omlt.rst | 3 +- docs/notebooks.rst | 4 +- .../notebooks/{ => trees}/bo_with_trees.ipynb | 29 +- .../trees/linear_tree_formulations.ipynb | 1463 +++++++++++++++++ setup.cfg | 1 + src/omlt/dependencies.py | 2 + src/omlt/linear_tree/__init__.py | 24 + src/omlt/linear_tree/lt_definition.py | 398 +++++ src/omlt/linear_tree/lt_formulation.py | 381 +++++ tests/linear_tree/test_lt_formulation.py | 765 +++++++++ tox.ini | 3 +- 14 files changed, 3120 insertions(+), 14 deletions(-) create mode 100644 docs/api_doc/omlt.linear_tree.rst rename docs/notebooks/{ => trees}/bo_with_trees.ipynb (99%) create mode 100644 docs/notebooks/trees/linear_tree_formulations.ipynb create mode 100644 src/omlt/linear_tree/__init__.py create mode 100644 src/omlt/linear_tree/lt_definition.py create mode 100644 src/omlt/linear_tree/lt_formulation.py create mode 100644 tests/linear_tree/test_lt_formulation.py diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index 47b13f4f..55870dbc 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -15,7 +15,8 @@ jobs: strategy: matrix: - python-version: ["3.7", "3.8", "3.9"] + # python-version: ["3.7", "3.8", "3.9"] + python-version: ["3.8", "3.9", "3.10"] steps: - uses: "actions/checkout@v2" @@ -35,6 +36,7 @@ jobs: python -m pip install --upgrade pip setuptools wheel python -m pip install --upgrade coverage[toml] virtualenv tox tox-gh-actions conda install -c conda-forge ipopt + conda install -c conda-forge pyscipopt - name: "Run tox targets with lean testing environment for ${{ matrix.python-version }}" run: "tox -re leanenv" diff --git a/README.rst b/README.rst index 19c03687..a5432beb 100644 --- a/README.rst +++ b/README.rst @@ -27,17 +27,32 @@ OMLT: Optimization and Machine Learning Toolkit OMLT is a Python package for representing machine learning models (neural networks and gradient-boosted trees) within the Pyomo optimization environment. The package provides various optimization formulations for machine learning models (such as full-space, reduced-space, and MILP) as well as an interface to import sequential Keras and general ONNX models. -Please reference the `preprint `_ of this software package as: +Please reference the paper for this software package as: :: - @misc{ceccon2022omlt, + @article{ceccon2022omlt, title={OMLT: Optimization & Machine Learning Toolkit}, - author={Ceccon, F. and Jalving, J. and Haddad, J. and Thebelt, A. and Tsay, C. and Laird, C. D. and Misener, R.}, - year={2022}, - eprint={2202.02414}, - archivePrefix={arXiv}, - primaryClass={stat.ML} + author={Ceccon, F. and Jalving, J. and Haddad, J. and Thebelt, A. and Tsay, C. and Laird, C. D and Misener, R.}, + journal={Journal of Machine Learning Research}, + volume={23}, + number={349}, + pages={1--8}, + year={2022} + } + +When utilizing linear model decision trees, please cite the following paper in addition: + +:: + + @article{ammari2023, + title={Linear Model Decision Trees as Surrogates in Optimization of Engineering Applications}, + author= {Bashar L. Ammari and Emma S. Johnson and Georgia Stinchfield and Taehun Kim and Michael Bynum and William E. Hart and Joshua Pulsipher and Carl D. Laird}, + journal={Computers \& Chemical Engineering}, + volume = {178}, + year = {2023}, + issn = {0098-1354}, + doi = {https://doi.org/10.1016/j.compchemeng.2023.108347} } Documentation @@ -152,6 +167,10 @@ Contributors - Alexander Thebelt - This work was supported by BASF SE, Ludwigshafen am Rhein. + * - |bammari|_ + - Bashar L. Ammari + - This work was funded by Sandia National Laboratories, Laboratory Directed Research and Development program. + .. _jalving: https://github.com/jalving .. |jalving| image:: https://avatars1.githubusercontent.com/u/16785413?s=120&v=4 @@ -172,3 +191,7 @@ Contributors .. _thebtron: https://github.com/ThebTron .. |thebtron| image:: https://avatars.githubusercontent.com/u/31448377?s=120&v=4 :width: 80px + +.. _bammari: https://github.com/bammari +.. |bammari| image:: https://avatars.githubusercontent.com/u/96192809?v=4 + :width: 80px diff --git a/docs/api_doc/omlt.linear_tree.rst b/docs/api_doc/omlt.linear_tree.rst new file mode 100644 index 00000000..6aa807ac --- /dev/null +++ b/docs/api_doc/omlt.linear_tree.rst @@ -0,0 +1,20 @@ +Linear Model Decision Trees +============================ + +.. automodule:: omlt.linear_tree.__init__ + +Linear Tree Definition +----------------------- + +.. automodule:: omlt.linear_tree.lt_definition + :members: + :undoc-members: + :show-inheritance: + +Linear Tree Formulations +------------------------- + +.. automodule:: omlt.linear_tree.lt_formulation + :members: + :undoc-members: + :show-inheritance: diff --git a/docs/api_doc/omlt.rst b/docs/api_doc/omlt.rst index 0df96e32..df9fa475 100644 --- a/docs/api_doc/omlt.rst +++ b/docs/api_doc/omlt.rst @@ -8,4 +8,5 @@ API Documentation omlt.scaling omlt.io omlt.gbt - omlt.neuralnet \ No newline at end of file + omlt.neuralnet + omlt.linear_tree \ No newline at end of file diff --git a/docs/notebooks.rst b/docs/notebooks.rst index 0a919772..4bf3dfd9 100644 --- a/docs/notebooks.rst +++ b/docs/notebooks.rst @@ -18,4 +18,6 @@ github `page `_. * `auto-thermal-reformer-relu.ipynb `_ develops a neural network surrogate (using ReLU activations) with data from a process model built using `IDAES-PSE `_. -* `bo_with_trees.ipynb `_ incorporates gradient-boosted-trees into a Bayesian optimization loop to optimize the Rosenbrock function. \ No newline at end of file +* `bo_with_trees.ipynb `_ incorporates gradient-boosted-trees into a Bayesian optimization loop to optimize the Rosenbrock function. + +* `linear_tree_formulations.ipynb `_ showcases the different linear model decision tree formulations available in OMLT. \ No newline at end of file diff --git a/docs/notebooks/bo_with_trees.ipynb b/docs/notebooks/trees/bo_with_trees.ipynb similarity index 99% rename from docs/notebooks/bo_with_trees.ipynb rename to docs/notebooks/trees/bo_with_trees.ipynb index 71686e3d..11801d96 100644 --- a/docs/notebooks/bo_with_trees.ipynb +++ b/docs/notebooks/trees/bo_with_trees.ipynb @@ -2,6 +2,7 @@ "cells": [ { "cell_type": "markdown", + "id": "df681f98", "metadata": {}, "source": [ "# Bayesian Optimization with Trees in OMLT\n", @@ -11,6 +12,7 @@ }, { "cell_type": "markdown", + "id": "4d6f8819", "metadata": {}, "source": [ "## Library Setup\n", @@ -31,6 +33,7 @@ }, { "cell_type": "markdown", + "id": "e691f915", "metadata": {}, "source": [ "## Define Black-Box Function and Initial Dataset\n", @@ -40,6 +43,7 @@ { "cell_type": "code", "execution_count": 49, + "id": "8f96917d", "metadata": {}, "outputs": [], "source": [ @@ -71,6 +75,7 @@ }, { "cell_type": "markdown", + "id": "ac51f219", "metadata": {}, "source": [ "## Training the Tree Ensemble\n", @@ -80,6 +85,7 @@ { "cell_type": "code", "execution_count": 50, + "id": "92bac6ec", "metadata": {}, "outputs": [], "source": [ @@ -110,6 +116,7 @@ }, { "cell_type": "markdown", + "id": "7b07ba8a", "metadata": {}, "source": [ "## Handling Trees with ONNX\n", @@ -119,6 +126,7 @@ { "cell_type": "code", "execution_count": 51, + "id": "26bb7771", "metadata": {}, "outputs": [], "source": [ @@ -137,6 +145,7 @@ }, { "cell_type": "markdown", + "id": "8eeefc86", "metadata": {}, "source": [ "You can use tools like [Netron](https://netron.app/) to inspect the model. Use the `write_onnx_to_file` function and provide a path and file name to export the `ONNX` model." @@ -145,6 +154,7 @@ { "cell_type": "code", "execution_count": 52, + "id": "c54d0dad", "metadata": {}, "outputs": [], "source": [ @@ -157,6 +167,7 @@ }, { "cell_type": "markdown", + "id": "a4d04727", "metadata": {}, "source": [ "## Build the Pyomo Model\n", @@ -166,6 +177,7 @@ { "cell_type": "code", "execution_count": 53, + "id": "f8f86aa0", "metadata": {}, "outputs": [], "source": [ @@ -186,6 +198,7 @@ }, { "cell_type": "markdown", + "id": "bc3a01e2", "metadata": {}, "source": [ "We also define an uncertainty metric according to [[1]](#1) to incentivize proposals far away from data used to train the tree ensemble. We implement `add_unc_metric` in a similar fashion to `add_tree_model` where relevant constraints are added to the optimization model." @@ -194,6 +207,7 @@ { "cell_type": "code", "execution_count": 54, + "id": "7e549e57", "metadata": {}, "outputs": [], "source": [ @@ -223,6 +237,7 @@ }, { "cell_type": "markdown", + "id": "7faf6bf6", "metadata": {}, "source": [ "## Running the Experiment\n", @@ -232,13 +247,14 @@ { "cell_type": "code", "execution_count": 55, + "id": "376f0893", "metadata": { "scrolled": false }, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -285,6 +301,7 @@ }, { "cell_type": "markdown", + "id": "3be35144", "metadata": {}, "source": [ "The main BO loop that minimizes the black-box function consists of:\n", @@ -299,6 +316,7 @@ { "cell_type": "code", "execution_count": 56, + "id": "5c09d3ee", "metadata": {}, "outputs": [], "source": [ @@ -333,6 +351,7 @@ }, { "cell_type": "markdown", + "id": "f0c72603", "metadata": {}, "source": [ "After going through 80 iterations we plot the progress to see how the algorithm converges to the global minimum of the black-box function." @@ -341,6 +360,7 @@ { "cell_type": "code", "execution_count": 57, + "id": "04352480", "metadata": {}, "outputs": [ { @@ -366,7 +386,7 @@ }, { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -419,6 +439,7 @@ }, { "cell_type": "markdown", + "id": "2ab589c4", "metadata": {}, "source": [ "The optimization model incentivizes exploration to find new promising solutions when activating uncertainty. To obtain the following results, we can install a solver capable of solving non-convex MIQP problems and change `has_unc=True` in the `minimize_model` function." @@ -427,11 +448,12 @@ { "cell_type": "code", "execution_count": 58, + "id": "e6ac3905", "metadata": {}, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "" ] @@ -452,6 +474,7 @@ }, { "cell_type": "markdown", + "id": "4fa7a32e", "metadata": {}, "source": [ "## References\n", diff --git a/docs/notebooks/trees/linear_tree_formulations.ipynb b/docs/notebooks/trees/linear_tree_formulations.ipynb new file mode 100644 index 00000000..f98373e1 --- /dev/null +++ b/docs/notebooks/trees/linear_tree_formulations.ipynb @@ -0,0 +1,1463 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "id": "547d2783", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "# Using Linear Tree Formulations in OMLT\n", + "\n", + "In this notebook we show how OMLT can be used to build different optimization formulations of linear model decision trees within Pyomo. Additional information on the formulations utilized in this notebook can be found in [[1]](#1). This notebook specifically demonstrates the following examples:\n", + "\n", + "1. A linear model decision tree represented using the GDP formulation and a Big-M transformation
\n", + "2. A linear model decision tree represented using the GDP formulation and a Convex Hull transformation
\n", + "3. A linear model decision tree represented using the GDP formulation and a custom transformation applied to the overall Pyomo model
\n", + "4. A linear model decision tree represented using the Hybrid Big-M Formulation
\n", + "\n", + "After building the OMLT formulations, we minimize each representation of the function and compare the results." + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "5cfa4057", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "## Library Setup\n", + "The required Python libraries used this notebook are as follows:
\n", + "- `pandas`: used for data import and management
\n", + "- `matplotlib`: used for plotting the results in this example\n", + "- `linear-tree`: the machine learning language we use to train our linear model decision tree\n", + "- `scikit-learn`: another machine learning language used to for the Linear Regression models\n", + "- `pyomo`: the algebraic modeling language for Python, it is used to define the optimization model passed to the solver\n", + "- `omlt`: The package this notebook demonstates. OMLT can formulate machine learning models (such as neural networks) within Pyomo\n", + "\n", + "**NOTE:** This notebook also assumes you have a working MIP solver executable (e.g., CBC, Gurobi) to solve optimization problems in Pyomo. For the Hybrid Big-M formulation, the solver executable must also handle mixed-integer quadratically constrained programs (MIQCPs). We use SCIP (users can install `pyscipopt`) in this notebook" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "7a82ca0e", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [], + "source": [ + "#Start by importing the following libraries\n", + "#data manipulation and plotting\n", + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import matplotlib\n", + "matplotlib.rc('font', size=24)\n", + "plt.rc('axes', titlesize=24)\n", + "\n", + "#linear-tree objects\n", + "from lineartree import LinearTreeRegressor\n", + "from sklearn.linear_model import LinearRegression\n", + "\n", + "#pyomo for optimization\n", + "import pyomo.environ as pyo\n", + "\n", + "#omlt for interfacing our linear tree with pyomo\n", + "from omlt import OmltBlock\n", + "from omlt.linear_tree import LinearTreeGDPFormulation, LinearTreeHybridBigMFormulation, LinearTreeDefinition\n", + "import omlt" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "1f8f945b", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "## Import the Data" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "6d66c118", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "We begin by training linear trees that learn from data given the following imported dataframe. In practice, this data could represent the output of a simulation, real sensor measurements, or some other external data source. The data contains a single input `x` and a single output `y` and contains 10,000 total samples" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "6d5f1146", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [], + "source": [ + "df = pd.read_csv(\"../data/sin_quadratic.csv\",index_col=[0])" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "9406ec5f", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "The data we use for training is plotted below (on the left figure). We also scale the training data to a mean of zero with unit standard deviation. The scaled inputs and outputs are added to the dataframe and plotted next to the original data values (on the right)." + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "e0ab963e", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#retrieve input 'x' and output 'y' from the dataframe\n", + "x = df[\"x\"]\n", + "y = df[\"y\"]\n", + "\n", + "#calculate mean and standard deviation, add scaled 'x' and scaled 'y' to the dataframe\n", + "mean_data = df.mean(axis=0)\n", + "std_data = df.std(axis=0)\n", + "df[\"x_scaled\"] = (df['x'] - mean_data['x']) / std_data['x']\n", + "df[\"y_scaled\"] = (df['y'] - mean_data['y']) / std_data['y']\n", + "\n", + "#create plots for unscaled and scaled data\n", + "f, (ax1, ax2) = plt.subplots(1, 2,figsize = (16,8))\n", + "\n", + "ax1.plot(x, y)\n", + "ax1.set_xlabel(\"x\")\n", + "ax1.set_ylabel(\"y\")\n", + "ax1.set_title(\"Training Data\")\n", + "\n", + "ax2.plot(df[\"x_scaled\"], df[\"y_scaled\"])\n", + "ax2.set_xlabel(\"x_scaled\")\n", + "ax2.set_ylabel(\"y_scaled\")\n", + "ax2.set_title(\"Scaled Training Data\")\n", + "\n", + "plt.tight_layout()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "842d4475", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "## Train a Linear Model Decsion Tree using the linear-tree package" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "a8d1d8a1", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [], + "source": [ + "#Build the linear-tree model\n", + "regr = LinearTreeRegressor(LinearRegression(), \n", + " criterion='mse', \n", + " max_bins=120, \n", + " min_samples_leaf=30, \n", + " max_depth=8)" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "9c6736bd", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [], + "source": [ + "#Data needs to be in array and reshaped\n", + "x_scaled = df[\"x_scaled\"].to_numpy().reshape(-1,1)\n", + "y_scaled = df[\"y_scaled\"].to_numpy().reshape(-1,1)\n", + "\n", + "#train the linear tree on the scaled data\n", + "history1 = regr.fit(x_scaled,y_scaled)" + ] + }, + { + "cell_type": "markdown", + "id": "550a5302", + "metadata": {}, + "source": [ + "### Saving your linear-tree model\n", + "\n", + "To save your model, you can use `pickle`. For example\n", + "\n", + "```\n", + "import pickle\n", + "\n", + "with open('lt_regr.pickle', 'wb') as handle1:\n", + " pickle.dump(regr, handle1)\n", + "```\n", + "\n", + "To load your model, you would run the following code:\n", + "\n", + "```\n", + "with open('lt_regr.pickle', 'rb') as handle2:\n", + " regr = pickle.load(handle2)\n", + "```" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "7b9963c7", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "## Check the predictions\n", + "Before we formulate our trained linear model decision trees in OMLT, we check to see that they adequately represent the data. While we would normally use some accuracy measure, we suffice with a visual plot of the fits." + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "5e63ccea", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [], + "source": [ + "#note: we calculate the unscaled output for each neural network to check the predictions\n", + "y_predict_scaled_lt = regr.predict(x_scaled)\n", + "y_predict_lt = y_predict_scaled_lt*(std_data['y']) + mean_data['y']" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "ef4fff6d", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuEAAALNCAYAAACBLr/fAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAAC/bElEQVR4nOzdd3yb1dk//s8tyZIsyZL33ivbSciADEgCCSPsTdij7C76bQvd9Pc8LX3apw8tpWUUkrBnmGWHJBASsrfjDO+9hyzJkjXu3x8mihVLHrGsW+Pzfr3yQjr3kXR5YF8+Oue6BFEURRARERERUcDIpA6AiIiIiCjSMAknIiIiIgowJuFERERERAHGJJyIiIiIKMCYhBMRERERBRiTcCIiIiKiAGMSTkREREQUYEzCiYiIiIgCTCF1ADR6LpcLjY2NiImJgSAIUodDRERERKcQRRG9vb1IT0+HTOZ7vZtJeAhpbGxEVlaW1GEQERER0Qjq6uqQmZnp8zqT8BASExMDYOCLqtfrJY6GiIiIiE5lNBqRlZXlztt8YRIeQk5sQdHr9UzCiYiIiILYSFuHeTCTiIiIiCjAmIQTEREREQUYk3AiIiIiogBjEk5EREREFGBMwomIiIiIAoxJOBERERFRgDEJJyIiIiIKMCbhREREREQBxiSciIiIiCjAmIQTEREREQUYk3AiIiIiogBjEk5EREREFGAKqQMgovDicrngcDjgcrmkDoWIwoxMJoNCoYBMxjVECn1Mwolo3FwuF0wmE4xGI0wmE0RRlDokIgpTgiBAp9NBr9dDp9MxIaeQxSSciMbF5XKhvr4eZrMZarUaSUlJUKvVkMlkEARB6vCIKEyIogiXywWr1Qqj0YiGhgZotVpkZmYyEaeQxCSciE7biQTcYrEgOzsbWq1W6pCIKMxptVokJCTAbDajrq4O9fX1TMQpJPE7lohOm8lkgtlsRlZWFhNwIgoorVaLrKwsmM1mmEwmqcMhGjMm4UR02oxGI9RqNRNwIpKEVquFWq2G0WiUOhSiMWMSTkSn5cRhTL1eL3UoRBTB9Ho9TCYTKzJRyGESTkSnxeFwQBRFqNVqqUMhogimVqshiiIcDofUoRCNCZNwIjotJ1adeBiKiKR04mcQV8Ip1PC3JxGNC8sQEpGU+DOIQhVLFJJXoiii2WgFAKQZoiWOhoiIiCi8MAknrzYda8O+2m4AQEmmAedNSZE2ICIiIqIwwu0oNITV7nQn4ABwoL4HVrtTuoCIiIiIwgyTcBrCaLUPGdtd0yVBJEREREThiUk4jcqOqk6pQyAiIiIKG0zCaQgBAtT2bmT07MbMpjcxr36t1CERkcSWLl0KQRAgCAI2bdokdThERCGPBzNpiNpdn+D+Hbe57/fLNNiZcauEERERERGFF66E0xBmfb7HfaXLAr2tWaJoiCgSbdq0yb3yvnTpUqnDISLyOybhNIRTkwKrQu8xlmCpkCgaIiIiovDD7Sg0hEwuQ7umAKm9pejU5KJDkw+rwoB+hwtKBf9uIyIiIhovJuE0hE4lx38m/w+sihiIwslvkW8rO7CkOEnCyIiIiIjCA5c1aYjpGQb0RcV5JOAAsIe1womIiIj8gkk4DaFSyKUOgYgCxOVy4YUXXsCKFSuQmpoKtVqN3NxcXH755XjvvffG9Fw9PT147bXXcO+99+LMM89EYmIilEol9Ho9CgoKsGrVKrz55ptwuVw+n+PRRx+FIAhYtmyZe+yrr75yH9Ic/C83N9frx7N582b89re/xfnnn4/s7GxoNBqoVCqkpaXh3HPPxR/+8Ae0t7eP6WMjIvI3bkchIopQzc3NuOKKK7B9+3aP8ZqaGtTU1OCDDz7AlVdeiRdeeGHE53rnnXdw4403wmazDblmt9vR29uLyspKvP7665g5cybeffdd5OXl+e1jOfE6eXl5aGho8Hq9ubkZzc3N2LhxIx577DE8/fTTuPnmm/0aAxHRaDEJJyKKQN3d3Tj33HNRVlbmHsvLy8OCBQugUqlQWlqKHTt24N1334VMNvKbpq2tre4EPDMzE1OnTkVqaio0Gg1MJhPKysqwZ88eiKKI/fv345xzzsG+ffuQkJDg8Tzz58/Hgw8+iIaGBvdKfHp6Oq688sohr3nqY51OpzsB1+l0mDZtGvLz86HX62G321FfX49t27bBaDTCbDbjlltuQVRUFK6//voxfe6IiPyBSTh5dWZePLazVT1R2Pp//+//uRNwpVKJp556CnfeeafHnB07duC6667DunXroFQqh32+jIwMPPbYY7jmmmtQWFjodU5VVRXuv/9+fPbZZ6ivr8fDDz+M5557zmPOypUrsXLlSmzatMmdhBcVFeHJJ58c8WOSyWS44447cOutt2LRokWIiooaMsdms+GJJ57AL3/5SzgcDtx33324+OKLodPpRnx+IiJ/YhJOXs3Li8ee8nokWCqQaK5AgqUSfVEGWO1/hDqKe8ZpbERRRJ/dKXUYQSs6Sg5BEAL2eseOHcOaNWvc95999lncdtttQ+bNnz8fn3/+OWbPng2LxTLsc1566aW49NJLh52Tl5eHDz/8EHPnzsWBAwfwyiuv4C9/+Qvi4uJO7wM5hVKpxOrVq4edo1Kp8LOf/QwulwuPPPIIuru78dJLL+H+++/3SwxERKPFJJy8ipLLMKntM6yo+KN7rCM6D69svw93LfbvPk4Kf312J575qlLqMILWvUvyoVEG7sfx888/D1EUAQwk2t4S8BOKi4vx4x//GH/84x99zhmLqKgo3HTTTThw4ACsViu++eabEZP3iXDHHXfgkUceAQCsX7+eSTgRBRyTcPKpQ1PgcT+urxZms1miaIjIXzZu3Oi+fcstt4w4/7bbbhtTEt7d3Y1t27ahtLQUHR0dMJlMHhVRjhw54r69b9++CUnCXS4Xdu/ejX379qG+vh5GoxF2u93r3H379vn99YmIRsIknHzq0OR73JfBibi+GgDTpQmIiMbtxMHIExYsWDDiY4qLixEfH4/OzuHPidTX1+ORRx7B22+/7bVKijf+LhXocDjwxBNP4PHHH0d9fb0kMRARjQaTcPIpKz0VPap0OGRKdGry0a4pgE0RA5dLhEwWuP2rROQ/PT096O/vd9/Pzs4e1eOys7OHTcL37t2L8847D11dY2vq1dvbO6b5w7HZbLjsssvw+eefSxYDEdFoMQknn6al67FmzjsQBc+DmM1GK9JjoyWKikJRdJQc9y7JH3lihIoO4GFnk8nkcV+j0YzqcVqt1uc1m82Gq6++2p2AJyUl4d5778V5552HwsJCxMfHIzo62n34dO3atbjjjjsAYNjGPWP1+9//3p2AC4KA6667DldffTVmzJiB9PR0REdHe1RMORHPif3xRESBxCScfJILwpAEHADqu/qYhNOYCIIQ0IOH5NuppfgsFsuwCfYJw50HWbduHaqqqgAMlCrcuXMn0tLSfM6fiJVnm82Gf/zjH+77a9euxa233hrQGIiIxoJt68knjdL76ly3pd/rOBEFP4PB4FHzu7a2dlSPq6ur83ntyy+/dN/+8Y9/PGwCDgx05PS3HTt2uFf5p02bNmwCPlExEBGNBZNw8ikpRuV1vLTRGOBIiMhfBEHAzJkz3fe3bds24mOOHz+Ojo4On9cbGxvdt2fMmDHi83399dcjzhlr3fSJiIGIaCIxCSefAtk8hIgCZ9myZe7bL7/88ojzX3zxxWGvD25rP1JTn927d2Pnzp0jvqZarXbf9lVa8HRjcLlcePbZZ0d8TiKiicQknIgowtx1113u29u2bRs2ES8vL8fjjz8+7PPl5588dPvBBx/4nGexWHDPPfeMKsaEhAT37YaGhhHnD47hq6++Qk9Pj8+5f/nLXzzKNBIRSYFJOA3rrPwEGPrqMbn1YyyqfhKXH34Ic+tfZDUBohBWXFyM22+/3X3/e9/7Hl544YUh83bt2oUVK1bAbDZ77CM/1eBmOy+88AL++te/wul0eswpLy/H+eefjz179ozqIGheXp67cktNTc2Iq+ezZ89GRkYGgIEyjNdee63HFhVg4PDmb3/7WzzyyCOjioGIaCIJIrOpkGE0GmEwGNDT0wO9Xh+Q17Tandj3wk9xVv1q91hV7AJYrnsD09INAYmBgpPVakVVVRXy8vI8tg5QaOjq6sKCBQtw9OhR91h+fj4WLFgAlUqF0tJS7NixA6Io4qqrrkJHRwe++uorAAMdN5cuXerxfEuWLPHYZ52Xl4czzjgDBoMBx48fx9atW+F0OpGRkYEf/ehH+PnPfw5goBvn2rVrvcZ400034dVXXwUwUErxwgsvRHZ2NuTygUPj8fHx+OUvf+mev2bNGtx5553u+1qtFgsXLkROTg46OjqwadMmdxnFV155BTfddJN7Ln8Vhi7+LKJgM9p8jTXDaFjqKPmQ9vWJlgo8V9rCJJwohMXFxWHDhg24/PLLsWvXLgBAZWUlKisrPeZddtllWLt27Yit5d98802sXLkSe/bsAQBUVVW5yxaeMHXqVLz11lvYsWPHqGL84x//iA0bNqC5uRkWiwXvvPOOx/WcnByPJPyOO+5AeXk5/vjHPwIYKKv4xRdfeDxGrVbjb3/7G2688UaPJJyIKNC4HYVG1KH1TMJj+luhcrDGLlGoS09Px7Zt27B69Wqcd955SEpKglKpRFZWFi655BK89dZbeO+99xATEzPic6WkpGDr1q148sknsXjxYsTGxkKpVCIzMxPnnXcenn32WezcuRNTp04ddXw5OTnYv38/fvOb3+DMM89EXFwcFIrh147+8Ic/YPPmzbj++uuRkZEBpVKJhIQEzJw5Ew8//DAOHDiAe++9d9QxEBFNFG5HCSFSbEcBgJrWbuhXL0ZXdA46vmtfX5GwFN+/cFbAYqDgw7eAiSgY8GcRBRtuRyG/yU4y4G9z3hkybne6ECXnmylEREREY8UMikbkq154r9UR4EiIiIiIwgOTcDpt35S3Sx0CERERUUhiEk6nraLVJHUIRERERCGJSTgRERERUYAxCSciIiIiCjBWR6FRSYpRQVO7EenGA4jvq0FcXzX2ZNwE4CdSh0ZEREQUcpiE06gkx6iQ2/IBiju+dI8lmo+jy9yPOK1SwsiIiIiIQg+3o9CoZCdo0Bmd4zEW31eNw01GiSIiIiIiCl1MwmlUipNj0KnJ8xiLt1RDp+KbKURERERjxQyKRkUmE9CuLUJV7AJ0avLcLewLXS6pQyMiIiIKOUzCadQ6NAV4b9oTHmONx9oxJydeooiIiIiIQhO3oxARERERBRiTcBq1pBiV1CEQERERhQUm4TRq09L1Xsfbem0BjoSIiIgotDEJp1GL1XivB360uTfAkRARERGFNibhNGo58Zqhg6ILdqcz8MEQERERhTBWR6FRk8kECKID8+vXIs5Sjfi+asT31eAl62tYNjlF6vCIiIiIQgaTcBoTEXLMbnwd0Y4e91h8X5WEERERERGFHm5HobERBHR5aV/vcLJpDxEREdFoMQmnMVk+JQUd0UPb1x9s6PHxCCIiIiI6Fbej0JhMz9Djo7gFcMqU6NTkojM6Fx2aAsS29GJ2dpzU4RERERGFBK6ET7Cf/OQnEATB/S83N1fqkMZFEAQcTzwPGwt+jv1p16Eudj4sygQ0dlulDo2IiIgoZDAJn0A7duzA3//+d6nDICLykJub614YqK6uHnL99ttvd19fu3ZtwOMjIooETMIniN1ux/e+9z24XDywSERERESemIRPkP/5n//BwYMHAQA33nijxNEEhiiKUodARBTSli5d6n4XYtOmTVKHQ0QTiEn4BDhy5Aj++7//GwBw0003YcWKFRJH5F9n5sd7Ha/psAQ4EiIiIqLQxCTcz0RRxPe+9z3YbDbExcXh//7v/6QOye8KknSeA6IIlcOII81GaQIiIr9au3YtRFGEKIq4/fbbpQ6HiCgssUShnz311FPYsmULAOAvf/kLkpOTJY7I/zRKOeItVZjb8ALiLTWI76uCTRGD5xUf4sLpaVKHR0RERBT0mIT7UV1dHR555BEAwNlnn40777xT4ogmRow6CnJXP6a1fuQeUznNUDj7JIyKiIiIKHRwO4ofPfDAA+jt7YVSqcQzzzwDQRCkDmnCdEXnQITnxxffVyNRNETkT6MpUfjoo4+65zz66KMAAIfDgRdffBHLly9HRkYGVCoV0tLScMUVV+A///nPmOPYuXMnHnroIcyaNQtJSUlQKpVITU3FkiVL8D//8z/o6uoa1fO0trZizZo1uO222zB79mzEx8cjKioKsbGxmDx5Mu644w589tlno3oubx93X18fnn/+eZx//vnIzs6GUqmEIAjYt2/fqD/WE8/51VdfuceWLVvm0WfC19fEW8nJiooK/OpXv8Ls2bORlJQEmUyGWbNmeX1ts9mMp556CpdeeilycnKg0WgQExODoqIi3HnnndiwYcOoP44T/PW1IwpnXAn3k9dff939S+bhhx/GlClTJI5oYjnkahhVaTDYGt1j8ZYqCSMiIik1NDTguuuuw9atWz3Gm5ub8f777+P999/HHXfcgeeeew4y2fDrP11dXbj77ruxbt26IddaWlrQ0tKCr7/+Gn/605/w73//G9dcc43P53riiSfwk5/8BE6nc8i1np4e9PT04OjRo1i7di3OPfdcvPnmm0hISBjlRw2UlZXh2muvRWlp6agfM9GeffZZ/OhHP4LVOnITtbfeegs//OEP0dzcPORaeXk5ysvLsWbNGlxyySV4+eWXYTAYhn0+f37tiMIdk3A/6OjowA9/+EMAQHFxMX71q1/55XltNhtsNpv7vtEYXAcf96SvgiC6vmtfn4deVQrmm2xI0KmkDo2IAshkMuHCCy/EoUOHoNFocPbZZyMrKwu9vb3YuHEjWltbAQBr1qzBpEmT8PDDD/t8rubmZpx77rkoKytzj02bNg0zZ86ETqdDa2srNm/ejI6ODnR3d+O6667DSy+9hJtuusnr8zU2NroT8Pz8fEyZMgVJSUlQq9Xo7u7GwYMH3Qn0hg0bsHz5cmzbtg0q1cg/xzo6OnDhhReitrYWarUaixcvRk5ODkwmE7Zt2zbqzx8APPjggwCAd999F42NA4sbV1xxBTIyMobMHW6R56233sLPf/5zAEB6ejoWLVoEg8GAxsZGdHZ2esx9/PHH8f/+3/9zl5fV6/VYsGABMjMz4XQ6UVpail27dkEURfznP//B0qVLsWXLFmg0Gq+v7e+vHVHYE2ncbrnlFhGACEDcsGHDkOtr1qxxX8/JyRn18/7ud79zP27wv56eHj9Gf3oq20zi/31+dMi/Nd9USh0aBUhfX594+PBhsa+vT+pQaIxycnLcP0+qqqqGXL/tttvc19esWeP1OQb/fFKpVCIA8bbbbhM7Ojo85pnNZnHVqlXuuTqdTjSZTF6f0+l0isuWLXPPnT9/vrhnz54h8/r6+sRHH31UFARBBCBqtVqxstL7z57nn39e/Mc//iHW19f7/Hzs379fnDt3rvt1/+u//svn3MEft0KhEAGI11xzjdja2jrkY+nv7/f5PL4sWbLE/fwbN24c1WMGfz0VCoWoVCrFZ599VnS5XB7zrFar+/b69etFmUwmAhCVSqX4pz/9STSbzUOee+/eveLUqVPdz3///fd7jWEivnajxZ9FFGx6enpGla9xT/g4ff7553jppZcAALfddhuWLVvmt+f+xS9+4X67tKenB3V1dX577vHKS9R6He+y2AMcCYUcc7vnP6djDHOH+f4yd3jOdfT7nmvpPGWuzT9z+7o859qH2Q7Q1+37Woix2WxYtWoV1q5di/h4zz4CGo0Gq1evRlZWFoCBVXNf+8NfeeUVbNy4EQBw1llnYdOmTZg9e/aQeWq1Gr/73e/w29/+FsDAnuY///nPXp/zzjvvxPe//32vK8onlJSUYP369UhNTQUA/Otf//K6feVUDocD559/Pt544w0kJSV5XJPJZIiKihrxOfzN4XBg9erVuPvuu4ecSzqxuu9yuXD//fe7Ozq//vrrePjhh72ucM+aNQtffvklUlJSAADPPfcc6uvrh8ybiK8dUbhjEj4OZrMZ9957LwAgISEB//u//+vX51epVNDr9R7/iELeXwo8/3Uc9z33bzM85zYf8D33n/M859Zt9z33mSWecys3+Z675iLPuUc/9j33pSs95x4aui/W7fXweQteqVQO2xNBrVZj1apV7vs7duzwOm/wczz99NOIjo4e9nUfeeQRxMbGAgBee+01d1J5OgwGA6688koAQFNTEw4fPjyqx/3tb38bcY97IM2fP3/E7R0ffvghjh8f+P/uiiuucH/cvqSmpuLHP/4xAMBut+PNN98cMkfKrx1RqOKe8HH41a9+5T6J/te//hWJiYnSBkREJIHFixe7V5F9GbwqeuLn5mBNTU3uaiJTp07FzJkzR3xdtVqNBQsW4JNPPkFPTw8OHTqEkpISn/NbW1uxbds2lJWVoaurC2az2b0fGgB27drlvr1v3z7MmDFj2NcvKSkJukP4N9xww4hzPv745B+SN95446ie99xzz3Xf/uabb/CTn/zEfT8QXzuicMQk/DTt2bMH//jHPwAMlJG67bbbJI6IiEgaIyWrADwqjng7ZP7tt9+6b/f19eH73//+qF67oqLCfbuurs5rInf48GE8/PDD+OSTT0a1zQQA2tvbR5wzZ86cUT1XII0mpsGf63Xr1nmURfSlp6fHffvUrZET+bUjCmdMwk/TgQMH3G+f1dbW4qyzzvI5t62tzX27qanJY+5vfvMbXHzxxRMXaACp7d2Qu/rhcBZCIQ+et2eJaGKNVLYOgMf+aLt96N7+ExVBAKCqqgr//Oc/xxyHt9rTn332GS6//HKPSlOj0dvbO+KcU/eBD3b8+HH8/e9/H/bxK1euxMqVK8cU13hiOmHw5/qNN94Y82uc+nmeqK8dUbhjEu4HFRUVHn/RD6e/vx/bt5/cqzo4QQ8118/Lwv6PnkFJy7uIt1Qh2tGD0uRLUNc12+fBTSL87JT/V9Sxvuf++OApc4dJ9h7ciYHCDN9RDXOG4t6vAHHQHlRVjO+5d3wy+rm3vAu4Bq20KnW+597wiu9rIcYfjckGr7SeLofD85BvW1sbrr/+encCnpOTg/vuuw9nn3028vPzERsbC7Va7Y7/0Ucfxe9//3sAGNUe5eH2PTc0NIyYjCYmJvo9CR9pLzYw/s/1qZ/nifjaEUUCJuF02tIMahyzdyPDuM89FtdXg2ZLP/LAJJx80I7h7MSY5o6+wQo08SPPOZ250XFjmBs7+rkRQKs9+TPjsssuw/vvvz/u5/z3v//tThBnzpyJr7/+etgD7qNZ/Q4HWq3W/XnZs2eP1yomY32+E/z1tSOKBNwzcJpuv/12iKI4qn9r1qxxPy4nJ8fj2u233y7dBzFOgiCgS5PjMRZvqYLoEn08gojIuxMl8AB47d54Or788kv37V//+tcjVpiqqanxy+sCwNKlS0f83XCi7X2g+ftzPRFfO6JIwCScxqUzOs/jvsJlg7135ANNRESDnXnmme7b+/btg9lsHvdzDt6rPNLhUafTiS1btoz7NcfLH1t7RjL4c+2Pj3kivnZEkYBJOI2LUZWKr3N+iPen/BVrzliHJxd8jW+5EEJEY3SipTwwcHbm+eefH/dzDq7fbbFYhp373nvvBcUqrlqtdt/2doDVHy655BL37dWrV8NqHaap1ChMxNeOKBIwCafxEWTYnXkLKuPPQXd0NkRh4JjB4Nq7RESj8fDDD7tv//rXv8bBgweHme3JWwKdn5/vvv3BBx/4fGxbWxseeuihUb/WRBpcyrGhoWFCXuPqq69GYWEhgIGKXQ888MCof2abTCavK93+/toRRQIm4TQu+UneD2AeazEFOBIiCnU333yzuylMb28vFi9ejGeeeQb9/f1e5xuNRrzyyitYunQpfvCDHwy5fumll7pvP/bYY3j55ZeHzNmzZw+WLFmCuro6jwOGUpk+fbr79ttvvz0hCxpyuRxPPfUU5HI5AGDNmjW4+OKLUVZW5vMx+/btw8MPP4ysrCxUVVUNue7vrx1RJGB1FBqXOTlxqGwbuiqyvqwFk1KHKeVGRHQKuVyON998EytWrMDevXthNBpx33334ec//zkWLFiAjIwMyOVydHV14ejRoygrK3OXtrv66quHPN9tt92Gv/71rzh27BhsNhtuueUW/PGPf8TMmTOhVqtx6NAhd5fMmTNn4oILLsCf//zngH7Mp7rqqqvwy1/+EqIo4qOPPkJJSQkWLlyImJiTP09vuOEGzJ07d1yvs3z5cjz11FO4//774XQ68cknn+DTTz/F1KlTUVJSAr1eD4vFgqamJuzfv3/Ecrr+/toRRQIm4QFw++23h3QVlOFkxHqvSdvvGLnGLhHRqRISErBlyxb85Cc/wXPPPQeHwwGj0YjPPvvM52Oio6O9dopUqVT48MMPcdFFF6GyshIAUFZWNmTFd9GiRXjjjTfw73//278fzGkoLi7GI488gsceewwAcOjQIRw6dMhjzvTp08edhAPA3XffjcLCQtx77704fvw4RFFEaWkpSktLfT5m2rRpiI/3XrbTn187okjAJJzGJRAn+YkoskRHR+Opp57Cww8/jJdffhkbNmzAsWPH0NHRAZfLBYPBgPz8fMycORPnnXceLrzwQp/lB4uLi7F3717885//xDvvvIOjR4+iv78fqampmDFjBm688UZcd9117q0ZweCPf/wjFi9ejDVr1mD37t1oaWkZ8WDp6Vq2bBnKysrw3nvv4aOPPsK2bdvQ3NwMo9EIjUaDlJQUTJ48GQsXLsRFF12EWbNmDft8/vzaEYU7QeQJupBhNBphMBjQ09MTVD+0Hv/imPu20mFCfF81etQZuG/lmcM8ikKd1WpFVVUV8vLyPCo6EBEFEn8WUbAZbb7GlXDyi/OPP4qcru3Q2QdqhH9S9HsATMKJiIiIvGF1FBq3GRkGRNt73Ak4AMT3VaPXOjE1bomIiIhCHZNwGreiFB06NZ6dM+P7qrG+rEWiiIiIiGgiiaKItl4b2nptaDfZ0GGy4VBDD7ZWtON4Sy/7hYwCt6PQuGXGaXA0Otd93wU55K5+VLdPzEEiIiIikobD6cLTX1XA7hw5yT5/WgqmpRsCEFVoYhJO4yaXCaiJnY8PJ/8POqNz0a3OgksWJXVYRERE5Ced5n68sLV6TI/5vLQFn5e2oCTTgHMnJ7Oi2imYhJNfmFSpKFelSh0GERER+ZnN4RxzAj7YgfoeHKjvQWGyDhfPSINMxmQc4J5wIiIiIvKhw2TDvzZWDDtH4ezDRUd/hXt2XIALj/0W0f2dXueVt5rw3r6GiQgzJDEJJ79I1qu8jptsjgBHQkRERP7Q1mvDi9/WjDivpHkdJrd/Dq29E1PaPsEt+25EmvGA17k1HRY8/sUxtBqtcLki+/Amk3Dyiwumed+K0mXuD3AkRERENB79Dhd213Ti5W0jJ+AAEGutx6dFj7rva+0dMCsThn3MK9tr8fcvj2NbZcd4Qg1pTMLJLxJ13lfCWaGIiIgodLhcIv65sRxfH2v3GNdbG1HQsdHrY7bkPACncLIgQ4t2EozqjFG93rcVHdha0T7yxDDEJJz8TuHsQ7LpCCa1fYZWY5/U4RAREdEo/Xtzpcf9RPNxPLRlHu7afTkuO/JzaPuHJsw2hR71hjPc9z8r/v2QOQqnFZcceRhJpqNDrm2v7ERZk9EP0YcWVkchv5mZ4MCcz66CwdbkHntWfwbm5i2SMCoiIiIajcONRlj6ne77MdYmXHfwbo85C2qexvqiXw95rEWZiK1Z96A6biE6NAVDri+ofQZFHRtQ1LEBAPBqyVq0xExzX//0UDMy4qKhV0dOiWOuhJPfZKZnINre7TEW31clTTAUMOyKRkRS4s8g/2jrteGz0maPMYsyHtVxCz3GZrS+j0TzMa/PsT37bo/E+oSU3sOY2/iyx1iqqXTIvOc3V+HxL45FTFEHJuHkNyJk6IrO8RiL76uWJhiacDLZwI8Pl8slcSREFMlO/Aw68TOJxs7S7/B6CNMpKIdsP9mbdj1MyuQxPf+cUxLwAylXoVU7yef8f39diVajdUyvEYq4HYX8JkatQKcmFynmIxAhwKhKgwgW5A9XCoUCgiDAarVCq9VKHQ4RRSir1QpBEKBQMKU5Xc98Ven9giDgo0l/xKymt2CwNmBn5q1o1xaP+fk/K3oUndG5WFD3bwBAXtc3EEQnmvQzfT7mle21uOmsbCTHqMf8eqGC37HkN2kGNb7IvAO7Mm5BlzobTvnA/zjzrPaI2uMVKWQyGXQ6HYxGIxIShi9FRUQ0UYxGI3Q6HVfCT9PgWt0KpxVn1j2HmP5W7ElfhVbdlIG93jn3j+s1nDIltmXfg+MJ52Ja64foVaXgUMoVIz7ulW21mJKmx4XTw7MjN79jyW8EQUCnJh/t2mJ3Ag4M7PGi8KTX62G1WmE2m6UOhYgikNlshtVqhV6vlzqUkPXkxnIAgCA6cXnZQ5jf8AKmtH2Caw/eC6XDNKrnWDE1BbcsyBlxXoe2EF/nPYS96TfCLtcMua7p70C6cb/HWFmTEdXt4fk7hivhRHTadDodtFot6urqkJWVxW0pRBQwZrMZdXV10Gq10Ol0UocTktpNNji/Wwmf0/Aysnt2oUU7Ba3aYlijDIi2d6Nf4f1ze/msdOQneV57aMXAVhWH04Uvj7TicOPoyw7KXf249MjPkGIqw5cFj6A05XL3tXf3NiA/SYuLpqdBqQif9WMm4UR02mQyGTIzM1FfX4/a2lqo1Wro9Xqo1WrIZDIIAs8EEJF/iKIIl8sFq9UKo9HoPo+SmZnJrSin6aXvWtKn9B7GwtqnBm6by6Bw9eFQyhXoic4c8pj7lxZAHSUf9nkVchkumJaKC6alwukSYbI6sHrLMO+KiyKWl/8R6b0HAQDnl/83EiyV2Jz7Q4jCwGtVtpnxz43lo3r9UMEknPxqUWEitpQPLeTvcLqgkPOHZDg6kYibTCYYjUa0tbWxZBgRTRhBEKDT6ZCQkMC94OPw9bE29+0ks2cDnbi+WrTopnqMCQJw35KxJ8BymQCDJgp3LsrzmYgXdXyJqW0feYwVdmzEjsw7YI2K9Rh/e3c9bj5r5K0voYBJOPlVXqLWaxJ+vNWEKWncsxeuZDIZ9Ho99Ho9XC4XHA4HSxcSkd/JZDIoFAom3uMkiiJ213S57x9KvRJt2mJcdOw3iLPWYWfm7WgwzHZf98fhSIMmCj84txD/2FA+5Fp5wjLsSr/ZXUu8XxaND6b8dUgCDgzUMz/SbERxcgxkstB+t1UQuWQVMoxGIwwGA3p6eoL2EEprrxWvbKuF3NWP2L5axPdVQxBdaM6+GHctzpM6PCIiooj39bE2jyT8hCinBbMbX8OujNvgkg2s056ZH4+FBYl+j+HZrytgtjk9xqa2fIjzKv8HHxX/AZUJS0Z8jktnpqMgSRt0Wx9Hm69xJZz8SiGToaj9S6w8+kvIMLAS2qnOxgtJ50scGREREVntTq8JOADY5RrsyLrLfV8mCBOSgAPA1Wdk4sVvPRsEHU65FDVxC2BWju41P9zfiGS9CjedGZrbU/h+DvlVnCYKvaoUdwIOALHWBshcdgmjIiIiIgA+E3Bv7jknf8LiSNCp8MCyAiTqlB7jo03AT2g12vD4F8dQ2tgTcueRmISTXwmCgJmz53mMyeBErLVOooiIiIgIGNgLvqOqExBHPrNz75J8RCsntgqJSiHHLQtyx3RmTG9tgNw5tKX956UteGNnaOUa3I5CfmeIS4QpKhE6ezt6lcnojM6FTHRIHRYREVFEK2vqBQDMr1+D/M7NKEteiaOJ5w85APnj5UUB3Wd94fRUzM6Oxavba71eX1L5f0gyH4PBWg99fwv6FAa8VrIGPdFZHvOaeqx4/ItjSDOoce3cLMiD/OAmk3DyO4VMwNvTn4JJmQS74mTzFpYpJCIiks5npc2AKGJq68eIs9YizVSKJVWPY2P+z3Aw9Sr3PCkOOqbo1bh/aQGe2lQx5Fqq6ZC7hjgARDt6sPLYr/BayQsDtRNP0dRjxRNfHkdGbDQy46MxLc0AgyZqQuM/HcyIyO+SdCp0aXI9EnAAONw0+s5ZRERE5D8n9kunmg4hznpyxVkuOtCmLXLfv3bu0AY9gaKOkuPK2RlDxndk3gEXTm6NcQpyfFnwS68J+GAN3X3YXtmJ1VuqUN5q8nu848UknPzOV93Ous6+AEdCREREAPDJoWYAQHH7eo/xLnU2mnXTAQDLp6QgM04T8NgGy03U4qEVxVAMyiWq4s/G6yXPo0s9sP3k2+z70KqbPKbn/XB/o1/j9Acm4RQwDd0WqUMgIiKKSEebB/aDb879Ad6d8jccTVwBh6DE4eSVgCDg7nPyMSPTIHGUJ33/3EKP+y0x0/DKrJfxdc4PsSvjFp+PS+09iGi79wowdZ3BlYdwTzgFzKlF+YmIiGjilTb2uG+LggLV8YtQHb8IKkcvgIFtKjpVcKWEgiDgx8uL8Lf1x91jdrkGuzN9J+AylwMXH/0lou1dKE2+FHvSb0JP9MntNQcbepAVL+1K/2BcCSciIiIKY5+XtngdtyliYFPokZ+k9XpdaoIg4IfnFY088TtFHeuhtzUjymXDrOa3cfueqxFnqXZfdwVZHfHg+rOHworM5YDBWo/4vmrE91WjRTcFQLHUYREREUWM1t6BmtoxtmYUta9HtzoLlfHneBxqPH9qqlThjUguE/DQimJsOtqKvbXdvieKIuY2vOgx1BwzDV3ROYOnBBUm4TQhFhclQvvR9zG17SP32P7Uq+Fy3ezz4CYRERH51yvbaqG2d2PV/tuhtXcAADbl/QR701e550x0Ux5/WDopGXKZgF3V3vd7K1xWtGmLkWCphFwc2P66K+MWjz82gm0lnNtRaELMzYlD56C/PgEg3lKNirbgKxFEREQUjk6sgi+qecqdgAPApLbP3LfPKR5bm3gpnV2UNOTA5gkOeTQ+L3oUq+e8h13pN6FFO2VgxX+QIMvBmYTTxBAEAZ2aXI+x+L4q/OdAkzQBERERRZhXttUCooiSlnc8xtNMpe7bc3LiAx3WuETJZbh/aYHP6yZVKjbn/RivzlwLUfBc4RcRXFk4k3CaMJ3ReQAAc1Q86vRnoCJ+KQS2ryciIgocQcBHxX/wGCpLuhAAMCs7VoKAxk8dJcdDK4qhVQ2zjUYYmuKm6NUTGNXYcU84TZiEnCn4V9R62KKCp+4oERFRJBjcIfJY4nKUNL+DLONuAEBV3CIAwLJJyZLE5i/3nDOwIi6KIjYda8O+4Q5uAlApgmvtmUk4TZilk1PxXNvQLpl2pwtR8uD6H4GIiCiceHSIFGR4Z9o/kGw+AktUAozqdOkCmwCCIGDZpGQsKUrCK9tr0G7qHzJnZpYBqYZoCaLzjUk4TRhfhf+PNPUGVVcuIiKicCJ6OYHokkWhOWaG+/6di/ICGVJAyGQCbjozB8dae/HJwWYAgD46ClfNzkCcVilxdEMxCacJIwjeSxHur+9mEk5ERDRBtpR3DHtdqZDBoIkKUDSBJZMJmJyqx+RUPVwu0T0WjJiEU8C19dqkDoGIiChs7azuHPb6vefkBygSaQVr8n0CN+YSERERhZmi9vXI6/zGa1UyBc9lBQWuhNOEM1jrEWepQXxfFeL7qrEz43awfT0REZH/iaIIiC6cXf0EDLYmmKISUZa8EvvSroVJlYoz80OrLng4YxJOE+7ag/cipr/Vfb8m9iyI4jKfe8aJiIjo9Gw62oasnl0w2Aaa4+ns7ZjX8CKOJS6HSZWKs/ISJI6QTuD7ETSh5uXGo+vU9vV91ei1sWkPERGRP/X1O7GvrhvTWv/jMd6mKUSrdjLm58UH/T7pSMIknCbUrOxYdGg8yyDFW6qxs2r4QyNEREQ0NqWNPQAAvbUR/TLNyfHkSwFBwMysWIkiI2+4HYUmlFYpR1d0DqzyGHRqctEZnYd6wxyU1ffgvCkpUodHREQUNjYfbwcAvFnyHAAgymmBpr8TFuXAPnBf/TtIGvxq0IQSBAEHUq/C/tRrAe4BJyIimhDeGvTY5Rr0RA+siM/P44HMYMMknCacKPDbjIiIaCLVdlqGvT47OzYwgdCocU84ERERUYh7Z0+Dz2uG6CholFwQCzZMwmnCpceqvY47XUPfOiMiIiL/StarpA6BvGASThMuK17jdXxfXXdgAyEiIgpD/Q4XAHjtjgkACwsSAxkOjRKTcJpwMzNjvY5/fawtsIEQERGFIbPNAZnLjrt2XY6LjzyCwvYNkDut7uvxWqWE0ZEv3CBEE06rUkDh7EN67wHEWwZa12v72/HhlP+VOjQiIqKQt3ZrNXK7dyCmvxUxHV+iuONL2ORaPD/nfdx/0VypwyMfmIRTQOj623B16fc9xtT2bmmCISIiChN9/U4AwKT2LzzGOzT5sEUZILA8cNDidhQKiB51OhxClMdYfF+NRNEQERGFh9LGHgiiA3ld33iMH0tcIVFENFpMwikgREGB7ugsj7F4SxUrpBAREY3DjupOiIICa894G18U/BK1hrlwCnIcS1iOs4t4IDOYcTsKBURRig4tuilwyFTois5FR3QemmOmobrDjIIkndThERERhSSbfaAyijUqFodSr8Sh1CuhchhhU+hR4qMwAgUHJuEUEIsKErG25dEh4w1dfUzCiYiI/Mim0AMAlApueAhm/OpQQMT5KI90tLk3wJEQERGFhy3l7VKHQOPAJJwkZbJ5byxAREREvomiiB1VnT6vnzs5OYDR0OlgEk5EREQUYtpN/cNeL8k0BCgSOl1MwilgtCq513EXK6QQERGNyRs7a6GzNSO7e7tHd8wTWB88+PFgJgXM7Ow4fHN86P61ynYTCpNjJIiIiIgoNNmdIkrav8A51U/AISjRqC/BscQVOJh6FZZxK0pIYBJOAZNmUCO2rxbZ3TsR31eFuL4adEXnoCrjv5iEExERjZIoDryDnNW9EwCgEPuR3bML3eqBfhzT0vWSxUajxyScAiYjNhrZ3TtxXuWf3GMaewc2NfRgxdQUCSMjIiIKHQ6XCJnLjkzjXo/xuth5uHZuJqLk3G0cCvhVooARBAEdmlyPsfi+WgiiU5qAiIiIQtA7e+qhtXegXVMI16BUrs4wFzHqKAkjo7HgSjgFVFd0rsd9hcsGva0JTtdkyGU8REJERDSSxm4roErF6zPXQOkwIbNnNxL6KtEXFQdDNJPwUMEknALKEhWPJt109KpS0aHJR7umABZFHKp4OJOIiGjM+hU6VCYsQSWWIFHnvTEeBScm4RRQOnUUXp+5Zsh4TYeFSTgREdEIbA7fWzhXzc8OYCQ0XtwTTgF1+ex0r+NWuyvAkRAREYWenVVdPq8peCAzpPCrRQGVHKP2On6spTfAkRAREYWendW+W9VTaOF2lHFqb2/Hli1bsGPHDhw8eBAVFRVobGyEyWRCVFQU4uLiMH36dCxduhS33norMjIypA6ZiIiIwsw5xYlSh0BjxCR8nG6//XZ89NFHXq85HA709fWhsbERn3/+OX7/+9/jF7/4BX7zm99AJuObEERERDR6VrsT0fYuLKn6P9Qa5qEudj56VakAwHNVIYhJuB8lJiZiypQpyMnJgU6ng8ViQXl5OXbs2AGHwwGbzYZHH30UlZWVeOGFF6QON7h81/2LiIiIvPtwfyOyundiStunmNL2KQCgRTsJr858CTEqpnShhl+xcVq6dCkuvfRSnHfeeSgsLPQ6p6WlBQ899BBee+01AMCLL76ISy+9FNdcc00gQw0a0zMMMJZ+juzuHUiwVCLBUoFDKZdDFP8EQWCtcCIiIm/qu/qwvGenx5hZmQgIAmTstRFymISP009/+tMR56SkpOCVV15BS0sLNmzYAAB45plnIjYJPyM7FrWbN2N20xvusURLBY63mlCcwrfTiIiITiV+945xdvcOj/E6wzycU5wkRUg0TtyYHCCCIOCOO+5w39+7d6+E0UgrQadChybfc8xSgY8ONEkUERERUXCr6+yDIDqxK+MWHE84F30KAwCgNnYe+CZyaOJKeAAlJZ38S7W3N7JL8nVoCjzux/XVQOaySxQNERFRcFtf1gJRkONA2jU4kHYNBNGJJPMxtGsKUZCokzo8Og1MwgPo8OHD7tu5ubnSBRIE2jUFqIxbjHZNAdq1hejQFMAlyKUOi4iIKCj19HkuVImCHK26KQAAgyZKipBonJiEB0hjYyP+93//130/UveDn9Cv0OH9qY9LHQYREVHQczjZVToccU/4BLJYLDh8+DD++te/Yvbs2WhsbAQATJkyBY888ojE0UlrVnas13GRpQqJiIg87K7x3ar+3iX5Pq9RcONKuB998803OPvss4eds3LlSrzyyiuIiYnsKiCpeu/t68tbTShihRQiIiK3rRUdPq9plEzlQhVXwgMkLi4Or732Gj766CPExsaO6jE2mw1Go9HjX7jIS9R6Hf8PK6QQERENJXJLSrjhn09+lJ6ejgcffBDAwLaK3t5eHD16FHv27EFXVxdWrVqFZ599Fk8//TSKi4tHfL7HHnsMv//97yc6bEmoo3gIk4iIaCQntmleePxRxPbVoTZ2Pmpj56MpZgacMqXE0dF4CCI34U64xsZG/OpXv8LatWsBDKyKb9q0CSUlJcM+zmazwWazue8bjUZkZWWhp6cHer1+IkMOiMe/OOZxX+60QgDwgwuH/7wQERFFinaTDS9trcK9Oy+Exn5yb/j6gl+ge+rNuGZOpoTRkTdGoxEGg2HEfI0r4QGQnp6ONWvWQK/X44knnkBXVxduuOEGHDx4EHK57xVhlUoFlUoVwEgDr6RpHbJ7diDBXI5Yaz2+LPgFRHEG29cTEREBcLpEJFrKPRJwAKiNnY+rpiRLFBX5A/eEB9Bjjz3m/ouorKwMn3zyicQRSS+rZyeKOjYg3loLGVxIsFTA4eKbM0RERMBAEn5qq/oeVTp61JmI1XA7SihjEh5AGo0GCxcudN/fsmWLhNFIb8mkJLSf0jkz0VKBndWdEkVEREQUXD491Iz9qddg3bQnsTPjVrRoJ6Mm9izIZXzHONRxO0qAxcXFuW93dPguORQJZmQY8KnWMwlPsFRiXWUnFhYkShQVERFRcBBFcaBTplyN2tgzURt75ncXXLiOe8FDHpPwAGtqOlmCLz4+XsJIpBcll6FFNxV70lahXVuAdk0hOjV5UodFREQUFCrbzd4vCDJkxEYHNhjyOybhAdTR0YFvv/3WfX/KlCkSRhMcelWp+Cr/J1KHQUREFHQ+Zu+MsMY94ePQ2Tn6vcsulwvf//733SUHVSoVLrnkkokKjYiIiEIcCxWENybh4/Diiy9i3rx5ePHFF4ftZnngwAGsXLkSr7/+unvsZz/7GRISEgIRZlA7IyfO63iHyeZ1nIiIKBI4nL47ZF46Mz2AkdBE4XaUcdq1axduu+02KBQKTJ48GZMmTUJcXBwEQUBHRwcOHDiA8vJyj8dcffXV+N3vfidRxMHlzLx47KnpGjK++Xg7rpidIUFERERE0qvv6oOhrx4y0YGu6BxgUP+MwmSdhJGRvzAJH4fBjXQcDgcOHTqEQ4cO+ZwfExODRx99FD/60Y+GbdITSXy1r6/ydRiFiIgoApQ2GjGn8WXMbF6HXmUyamPnoyxpJepi50kdGvkJk/BxuP/++3Heeedh/fr12L59O0pLS1FbW4vu7m4AgF6vR1paGmbNmoXly5fj6quvhk7Hv16HI4gOxPXVoVfJLmBERBS5jrX0YmH3TgBATH8rprX+B53RuUzCwwiT8HEqLi5GcXExHnjgAalDCWnLKv6M9N59iLdUQyHa8f6UvwKYLXVYREREklDbuxFnrfUYq42dj6x4jUQRkb/xYCYFhURLOZLNx6EQ7QCABEsFLP0OiaMiIiIKPJvDiVhrHewytXvMLlOhTVuEi6anShgZ+RNXwkly0Uo52jUFyDTudY8lmiuwtbwDy6emSBgZERFR4O2v60FzzAz886yNiLdUI8V8BGp7Ny6YkQmtiqlbuOBXkiR31ewMHKzxbF+vsXegptMiUURERETS2VLeDgAQBQU6tIXo0BYCAB5K00sZFvkZk3CSXFKMCrWx87Ah/2do1xSgQ1MAa1QsDFIHRkRERDRBmIST5ARBQHd0Drqjc04ZlyggIiIiognGg5kUtLotdqlDICIiIpoQTMIpKGTERnsdF0UxwJEQERFJp8Vo9Tp+Vn5CgCOhicYknILC7OxYr+PHW02BDYSIiEhCr26vRV7nZqQZD0DhPJmQn5UfL2FUNBG4J5yCQm6i1uv4Z4eaUZwSE+BoiIiIpHNexZ8Q098KF+To0OThy4JHIAjFUodFfsaVcAoK8sGnMEUX9NZG5Hd+DdHeJ11QREREAabpb0dMfysAQAYnkizlsClYmjAccSWcgoIgABBduO7gPUgyH4fSNVAj/NWStQBKpAyNiIgoYFJMRzzu98s06IrOligamkhcCaegIAgCIMgQ7eh2J+DAQDt7IiKiSNBhskEmOtCuyYfruxStVTcJ0SqlxJHRROBKOAUNrUqODk0B4vtq3GOJ5nK4XCJkMhYNJyKi8PbxoWa0JyxFRcJSRDktSDIfgyC6cNfiPKlDownAlXAKGpfNzEC7ptB936rQQxRkKG9jhRQiIgp/7b029227XING/Sw0GM6AQs50LRxxJZyCRqpBjU+SzkdzzFS0aYpgViYBgoCU6i5WSCEiIqKwwiScgoq39vW+GhcQERGFi75+p9QhUIDx/Q0KKjFq/l1IRESR5+mvKryO8/di+GISTkFFHx3ldZzt64mIKBLNz2OnzHDFJJyCikfTnkHqu9i0h4iIwpgoYlHNPzGl9WPEW6ogiAPbU3LivXeUptDH9zgoqMzMMqC20zJk/O3d9XhoBVv2EhFR+HE4XdD2t2F+/Vr3WL9MgxfOeBMGDX/3hSuuhFNQKUz+rgqKKELT347s7u2Y3fCqe0WAiIgo3DR2W5FiKvMYEwUBJmWSRBFRIHAlnIKO0mHCHbuvhMbR7R6ril8MYIpkMREREU2UbyvbkXNKEt6imwIIXCsNZ/zqUtDpV+ggnvKDh+3riYgoXLX12tChyUd5/BL0KpMBAK3ayZiarpc4MppIXAmnoHPF7Ax0HCqAtqfTPZZorkC/wwWlgn83EhFR+BBFEXaniGNJ5+NY0vkAAG1/OyCKuHNyssTR0URiRkNBJy9R625fb5ep0aybCktUHA419kgcGRERkX+1DWpVf4JZmQizKont6sMcV8IpKO1JX4V9adeiR51xck/c0TackR0nbWBERER+9NbueqlDIIkwCaeg1KtOkzoEIiKiCdfvcHkdP39aSoAjoUBjEk5EREQkEbmrH1eXPohW7SS06KagRTcVndE5mJZukDo0mmBMwimkiKIIwUdXTSIiolDicLqQaC5HhnEfMoz73OP/PHOjdEFRwHDHP4UUlyh1BERERP5R22kZ0qSnU52NfoVOoogokJiEU1C68cxsr+MmmyPAkRAREU2MYy29SDEd9hhriZmKKWkxEkVEgcTtKBSUopVyAECUw4wky3Ekm45AY+/EajyAh1YUSxwdERHR+JU19aI75XL0qDOQYipDiukwWnRTUJzCJDwSMAmnoBSjUiCl9zBWHbgdAgb2oLggx/asuySOjIiIyH+a9CVo0pe47wuiE9PUURJGRIHC7SgUlARBQKcm12NMBicSzeXckkJERCHPV2lCUZAjKUYV4GhICkzCKWjFxcWhK9pzb3iy+SjavXQXIyIiCiWWfu8LSlefkRngSEgqTMIpaKXq1WjVToYLcrRpilCafAm61VnYfLxN6tCIiIjGxVe1L0M0t6JECu4Jp6CVEReNTXk/wedFv4FTNuitOVO/dEERERH5wTof7eoNGibhkYIr4RS0ipNj0KeM90zAiYiIwoDJ2g+IbH4RybgSTkFLJvPdGdPlEoe9TkREFKxae63INO7FxUd/4W5V3xQzHdVxi6QOjQKIK+EU1HITNV7HqzvMAY6EiIjIP8w2J1J6D0Nj70Je11acVfccFtY8LXVYFGBMwimoLSpI9Dp+vNUU4EiIiIj8o6zJOLRTpm6qRNGQVJiEU1DzVSv1cKMxwJEQERH5x9HmXqSYyzzGWnRTWJ4wwnBPOAU1QRAgiA4kWKqQbDqCZPNRJFgq8c60f0gdGhER0Wl7Z+qTSDEddrerb4qZjhUJ3rdgUnhiEk5BL9reg1v23egxFttXB5ujGCqFXKKoiIiIxu5Ek56e6Ez0RGfiWNL5EkdEUuF2FAp6FmUCTFGee8OTzUdhtjklioiIiOj07K7pkjoEChJMwinoyWUCWnWTPcaSTUewv65bmoCIiIhOU3U7q3vRAG5HoaCXn6RFq3YSEiyVaNVNQqt2Mmpj56O5rhvLJidLHR4REdGo9fTZvY7LBPa+iDRMwinozc+LxyvN9+DbnPukDoWIiGhc7E7vXTJvW5gT4EhIatyOQkEvSacCBH6rEhFR6FM6TIhyWoaMx2qUEkRDUuJKOAU9QRCQEReNhq6+Idf6HS4oFUzQiYgo+ImiiJnNb2NRzb/Qpc5Cm7YYlfFnwzz5GqlDIwkwe6GQkKTz3rSnnJ0ziYgoRLyzpwFJ5mMQICLeWotJHeuR3nsA3A0emZiEU0iIUXt/0+az0uYAR0JERHR6ajstSDIf8xhr0xbjrIIEiSIiKTEJp5CQHc8uYkREFNrkLhsM1gaPsVbtJKTp1RJFRFLinnAKCUkxKsTYmjGp7TNo7J3Q9HcgymXDh5P/LHVoREREo+KUqfDPMzci0VKBJPMxJJuPol1TAJmMG1IiEZNwCgmCIEAmOnB2zZMe47HWOohiMQTWVyUiohDglKvREjMNLTHTpA6FJMbtKBQyelQZMEd57ptLN+5HTcfQUk9ERETBxOnyXh+cIheTcAoZWrUCjfqZHmPpvfvR2DO0dCEREVEw+c+BRq/jZxclBjgSChbcjkIh4/JZGThUPQ8aeycaY0rQoJ+FppgZsFZ2YmEBf4gREVHwqmwzex2fkxMX4EgoWDAJp5CRpFPhQNo1OJDGpgZERBQeeKYpcnE7CoWM4U6Pc68dEREFM721AQnmcshcDqlDoSDBlXAKC0eajZiWbpA6DCIioiHsThdKmt/BvIYX4RCi0KEpwOHki9E1406pQyMJcSWcQsqtC3K8jn9e2hLgSIiIiEZn09E2JJuPAgAUoh0p5iPQ2LuQFKOSODKSEpNwCilaFd+8ISKi0HKovttru/qSjFhpAqKgwIyGQorMywEWna0FTplSgmiIiIhGpnKaYJd5tqZv1RZDp2YaFsn41aeQolQMvHlT1L4ehR2bkN67H3pbM77O+SGaekqQZoiWOEIiIqKTLP0O2BQxWD33A6gcRiSZjiHJcgw96gzI2a4+onE7CoWkrJ5dmNz+GfS2ZgBARu8+bDzSJnFUREREnnZUdbpv2xR61MfOxd70GwGBKVik43cAhZzLZqWjMabEYyzduB8t7JxJRERB5mB9j9fxy2elBzgSCjZMwinkpBnUQ9rXC6IL2n6uhBMRUXBx+OhjkZ+kC3AkFGy4J5xCjkapgFGVjn2p16JdW4hG/Ux0ROcBggw2hxMqhVzqEImIiIiGxSScQpMgYGPBz4cMf17agktn8i0+IiIiCm7cjkIhaVJqjNfx8lZTgCMhIiLyLaW3FAUdG6G3NgDiwNaUE5W+KLJxJZxC0qTUGBxt7pU6DCIiIp/sThdmtLyLGS3vAwCsch12Z9yCqdf/fxJHRsGAf4pRSMpP1Pq85vRxCIaIiCiQvixr9eiUqXaa4JCp2KSHADAJpxAlCILX7pkA0NjNUoVERCS9I42dSDRXeIy1aYsh9/H7iyILk3AKWRlxJ7tjKpxWZPTsQU7Xt2hgEk5EREFAbTeiUT8TfQqDe6xNWwQZO2USuCfcL6qrq/HFF1/gq6++wsGDB1FbWwuTyYSYmBhkZmZiwYIFuPHGG7FkyRKpQw0rsdFRQPcOLK55EknmY5CLTrRpivBy3AKclZ8gdXhERBTh+pTxWDf9X4AoQtffggRLJaxRsVKHRUGCSfg47N27F/fddx927Njh9XpXVxe6urpw8OBBPPvss1i6dCleeOEFZGdnBzjS8DQvNx6flimRaipzjyVYKqB0sEIKERFJq2lwF2dBgEmVCpMqFdfNy5IuKAoqTMLH4ejRo0MS8OLiYkyfPh2JiYno7u7G1q1bUV9fDwDYtGkTFixYgM2bNyM/P1+KkMOKQROFFu1kOAUF5KIDACCDC6mmUgBnSBscERFFtNd31Hkdj9NEBTgSClZMwv2gsLAQ3/ve93DzzTcjIyPD45rL5cLatWvxgx/8ABaLBY2NjbjpppuwdetWCDyYMW5OuRqt2slIMx2CC3K0aQshEx0obezBtHTDyE9AYyaKIuxOEQ6XC/vrehAlFzArKxYKOY+YEBGNRGQBL/oOk/BxSEtLw5o1a3DLLbdALvfeKl0mk+HOO+9EXFwcrrrqKgDAtm3b8Pnnn+OCCy4IZLhha2vOfXBBjpaYqbDLNQCA6tIWJuF+1m3pR31XH7443DLk2ubj7e7bqigZbl2QC52KP16IiE6lUXrPFyjy8LfkOCxZsmTUhy2vvPJKzJ8/37195aOPPmIS7gdn5sVje9WZUocR1qx2J57aVDHyxO/Y7C78++tKnJWfgAUFPCBLRDQY3wWnE/j+cQAtWrTIfbu6ulq6QMLIcFVQLP2OAEYSvsaSgA+2rbIDW8rb4WLzJCKKMFa7EwUdmzCn/iVkd29HtL0LAGCI5n5wOokr4QE0+K9fp9MpYSThY7haq8dbTJiZFRu4YMJIv8OFLRXt2FfbPa7n2VHViR1Vnbjr7Dzo1fzlQ0SRYWtFO6a0fYyijo3usR2ZtyPtqsckjIqCDVfCA+jgwYPu21lZLFHkL7E+TppvqWj3Ok7D67Xa8c+N5aefgIuuIUPPb65Cr9U+vsCIiELE/roej3b1ANClzkLmoCZzREzCA6S2thYbNmxw31++fLmE0YSX86eleh232YcmgzQ8p0vEc5urhl4QRRS2b8CVpT/Ahcd+6/sJRBHf23UpLjr6a2j7Pf8Iem5zFbcIEVFEUDpMiLU2eIy1aSdxPzh54HaUAPnJT37i3oKSnZ2NSy+9VOKIwofD+V2yLYqI66tBWu9BpPcewDc5D8LudCGKpfNGxWxz4NmvK4eMZ/bsxrLKvyDRMrA33KrQD6x2C0M/r3pbI2L6WzG5/TPY5WqsL/y1x/VnvqrEA8sKoFKwOgARhS+Fy4YDKVchyXwMiZbjkIkOdGrypA6LggyT8AB44YUXsG7dOvf9xx57DCqVasTH2Ww22Gw2932j0Tgh8YW66Cg5ILpw167LoO8/WT6vIv4cHG2ehOkZLFU4Eku/9wQcAGSiw52AA4DaYUSS+TjadJM85sX21SKt9+SWqxkt7w9JwgHgXxsr8OCyQigV/OOIiMKPw+mCRZmALwt/AQAQRCf0tiYUZyRKHBkFG/4WnGC7du3Cfffd576/atUq3HjjjaN67GOPPQaDweD+x33k3iXFqABBhh61Z6OktN6DXmta01DDfZ5qDfPRpJvmMZbVs2vIPF1/Gy48/qj7/uGki30+54vfVo85RiKiUHC81eRxXxTk6FFnIj1WLVFEFKyYhE+gqqoqXHrppbBarQCAkpISPP3006N+/C9+8Qv09PS4/9XVeW+BG+lO7LFripnhMZ5uPCBFOCHn8S+OobLN7HuCIGBH1p0AgM7oHHyV+2OUJl8yZFqXOtvj/p70VT6fstfqgMi2cUQUhj491Ox1XMb94HQKbkeZIE1NTVixYgWamwf+Z8zPz8enn34KvV4/6udQqVSj2rZCwIPLCvHJuhKgATBFJaJJX4Jaw1ypwwp6++q6RzWvMu5svDflcVTHLYAoeN/PbVYlYUv2/Tiz7nkoxH6YlcO/9bq9qnPYOu9EROGkMFkndQgUZJiET4COjg6sWLECFRUD+2jT0tKwfv16pKWlSRxZ+FIqZKg1zMNzcz5AryoVGLTiYLU7oY7iQcBT9fU7sfFI68kBUfT4vHkQBFTFLx7xOXdk3Yk96asQY2uBJSrO86IoItV0CM3fvWPxbUUH0gxq5CRoT/dDICIKGfw9RKfidhQ/MxqNuOCCC1BaWgoASExMxPr165GXx1PRE80hj0avOm1IIrlhcKJJAACXS8TTXw3qhCmKuOTIzzGvbg0EcXyNpBzyaHRpcj2qpwiiE0ur/opVB+7E1JYP3ePv7GngthQiChud5n6pQ6AQwiTcj8xmM1auXIndu3cDAAwGAz799FNMnTpV4sgig95HO+Cjzb0BjiT4bTzq+YdJSfM7KOrchMW1/8I1h+6H3to4qucpShn57VW5y4aVR3+F2U1vAABWlP8BuV1b3Nf/tv74GCInIgpeXxxuxsymN3Fe+R9R0vQ20oz7EeW0SB0WBSluR/ETq9WKyy67DFu2DCQXGo0GH330EebMmSNxZJHjkpI0vLq9Vuowgp7N4cSB+h73/ThLNc6pftx9P9O4F1eW/gAvnvHmkP3fMkHAfUvzfdb5fnNnHRq6+zzGCjq/RnHHlyefA05cfOSXWD33PfR9t2Wl12pHDNvaE1GIa+y24qyOr5DTs8M9ti3zTuhX/l7CqChY+WUl/Kc//SmOHj3qj6cKSXa7HVdffbW7I6ZKpcL777+PRYsWSRxZZBnu3Dk7NQ4QRRHrdnt2ccsw7oPc5dlS/qu8nwxJwG9dkIMfLS8attHOdfOycNvCXI+xY4krsCvjFvd9p6DA+sJfuhNwAN67dBIRhRpRRJLF8929Nu0kZMazXT0N5Zck/P/+7/8wdepULF68GC+88AL6+vpGflCYcDqduPHGG/Hxxx8DABQKBd588022pZeAr+0owECnRgL+takCLUarx9ih1Cvw1oxn0aNKBwDsS70W1fEn/4C8c1EefnheERJ0o6vUE69V4kfnFeGiGanusc0530dZ0oWwybV4d+rfcTTpgiGPKz+lti4RUajR9rdDY+/yGGvTFiNGxY0HNJRf94R/++23uPPOO5GWlob7778fu3YNbegRTkRRxF133YW3334bACCTyfDSSy/hsssukziyyHTi5LnK0Yucrq1YUPM0rj70ABbW/EviyILDoYYe9DtcXq816mfi5VmvYEfm7fg694fu8WvmZMKgiYJcNrb6tjKZgMmpety/tGBgQJDh88Lf4rWStaiLne/1MR/ub+QhTSIKaU5ZFL7O/SHKki5EuyYfVoUePep0dz8LosH88qfZLbfcgnXr1sFiGTh8YDQa8eyzz+LZZ5/FjBkzcPfdd+Omm25CbGysP14uaDz11FN44YUX3PcLCgrwzTff4JtvvhnV45988smJCi2izWx6E4tqTzZF0tg7sDXnAQkjkp4oiie7Yooizmh8FWVJF6FPGe+e06/QYUvOg+77l5SkISteM67XVUfJkRijQnuvDS5Z1EDVlGF8fbwdS4qTxvWaRERSaDVaYY2Kxe5B2+9kLodHpSiiwQTRT0tPvb29ePXVV7F69Wrs3Lnz5At899efSqXC1VdfjbvuugtLly71x0tK7tFHH8Xvf3/6hy3G+qk3Go0wGAzo6ekZU9OfSPKfA43oLf8Wqw7c6TH+zLxPcN68GShMjpEoMml9WdbiPoxZ2LEBlx55GFaFHt/kPIiDKVcM+SVx0YxUTE71z/eYKIpjqoDy0Ipiv7wuEVEg7a7pxNfH2oeMXzMnc9wLGhRaRpuv+e3Ps5iYGNx7773Yvn07Dhw4gB/+8IdISEiAKIoQRRFWqxWvvvoqzjvvPBQVFeFPf/qTu5skkb+snJ6GFt0U2OSeDWCyu3fiw/1NEkUlLZPN4U7Ao5wWLK38KwBA7TBiecVjuPTIwx7z4zRRfkvAgYE/xH+8vGj4SaIIiANbZZwubkkhotDzbUWH13Em4OTLhLxHMn36dPztb39DY2MjXn/9dZx//vmQyWTuhLyyshK/+tWvkJ2djSuuuAL/+c9/4HJ536sazB599FH3x3Q6/8j/ZDIBoqBAnWEeWrRTsDPjVqyb9iTKE5ZJHZpkjjQZ3bfn161GTL9njfCjiSs87p9a3cQfBEHArKzYoRdEETld32LVgdswrXWgic+7e9nAh4hCj93Jn1s0NhN6XDcqKgrXXXcdrrvuOtTV1WH16tVYu3YtampqAAAOhwMffvghPvzwQ6SlpeH222/HnXfeifz8/IkMi8JcUowK/5n8pyEl9iKRyyVi8/GTb4/uS78BelszJrd/BgCoMczHsUFJ+I+XF03YAaJlk5MhlwnYXTNQOSCltxRLqv+GDOM+AICm7jkcSboIdZ1As9GKNANLehFRaLA5xtdpmCJTwE4LZGVl4Xe/+x2qqqrw2Wef4frrr4dKpXKvCjc2NuKxxx5DcXExli9fjnXr1sHp5Dc1jd2KqSk+E/DG7sgpnwkA6/bUe9w3KxPxyaT/xrppT6JNU4gNBT8Hvku6r5ydMeEn+BcUJLhvqxy97gQcAPS2ZkxveQ8AsL+uB0REoeJ4C0us0thJcmT37LPPxkUXXYQpU6YAGHirWhAEiKIIl8uFjRs34rrrrkNxcTFeffVVKUKkEJaiV/u89sbOugBGIq3qdjPqu7z/0VEbeyZenvUquqNzAABLJyUhN1Hrda4/RclluHRmmjuGev1sj+tF7QOdNcuajHBxbzgRhQiXKGJx9T9w9aH7cU7V45jS+jH01kYsncRqT+RbQJPwXbt24f7770daWhruuOMO7N+/3518R0VF4dxzz4VOp3OvjldVVeGWW27BqlWruEeUaAz6+p14d2/D8JMGrXp73a89QQqTY3BxSRogCNiafR8AoEeVhs8Kf4N10//pnvdZKQ9uE1FoMPY5kNmzB9k9uzCn8VVcePx3KGr/EpNSI7MiF43OhCfhXV1deOKJJzBz5kyceeaZePbZZ9HT0+NOtAsLC/HnP/8Z9fX1WL9+PZqbm/Hcc89h1qxZAAbKm7355pusqU1jMjXdd3WPbkt/ACORxtNfVYx67r1L8gPeSKI4JQY/OLcQDYYz8MHkv2DtGetwOOUyiMLJYypHmnvhcIbegW0iijy7qtqQOKRdfRE0SnbKJN8mLAn/4osvcMMNNyA9PR0PPfQQDh065E68VSoVVq1ahY0bN+Lo0aP46U9/isTERACARqPBnXfeiT179uDvf/+7+/lWr149UaFSGBrS8EUUEdtXA01/O9ZsqZYkpkDpsdjdt9X27oHyfz6smp8t2S8JhXzgx09FwlK4ZFFe55S3cZ8lEQW/2L46RLlsHmPt2hFKs1LE8+tv39raWqxZswZr165FbW0tAM+GNJMnT8bdd9+N2267DfHx8b6exu0HP/gB3nzzTWzZsgXHjh3zZ6gU5lSKgQQvt3MLijq+RHb3Duj7W/BNzgPYmXmHxNFNrNVbqgAAgujA1aUPwKowYEP+w0O6VWbHa5Bq8L1/PhBi1Ar0Wh0+r39ysNmvNcuJiCaCWZmIDyb/GUnm40gyH4O2vx0WZcLID6SI5pck/I033sDq1avx5ZdfupPuE/9Vq9W45pprcM8992Dx4sVjfu6SkhJs2bIFVqvVH6FShDixvSK3eyumf1d/GgCyu3eEdRK++Xib+/aspjeRbB54e/SWfauwK+MWbM+6C06ZCgBw1RkZksQ42EUz0vDmCIdl+x0uKBVs+0xEwanVaEW/QoeKhGWoGNST4rJZ6RJGRaHAL7/ZVq1ahfXr18Plcrm3nEybNs3dsOfFF188rQQcAJRKpT9CpAhVa5jvcT/deABypxVWe/iVv+yx2LGreqAGt9bWioU1z7ivyUUH8js3Q8RA6cZbF+QEfB+4N+mjWIn/+ljbiHOIiKTyyvZar+PJMaoAR0Khxm/bUURRRHR0NK677jrcc889WLBggV+e98Ybb3Qf0iQai2vmZOKD7XPgggwyDBzwsyr0iLU14MP9cbh2bpbEEfrXiW0oABDT3wKrIgbKfot7bEPBI3DJBv6XT9AFxy8HQRCwuCgR3wxqKBTltGBS2+eY3vI+Ppr8GA42AMunpkgYJRHR2MWovZ91ITrBL0l4SUkJ7r77btx8880wGAz+eEq3efPmYd68eX59TooMWfEa9Ct02JN+I3pVKaiNnY/O6LyB0nxdfbA7XYiSh8c2h06zZ8WX5pgZeOGMt3BW3XM4o/EVlCVdjEb9TADADfOD64+Pebnx7iR8cfU/UNK8DiqnGQAwreUDbMu+Bwfqu1GSGSthlERERP7llyR83759/ngaIr8rTonBZvzI67X1h1tw0Yy0AEfkf6Io4oWt1UPGHfJofJP7A5Qlr4Q5auCA0NzcuKBsB3/F7Ay8t7cBSqfZnYADwPSW97E96058WdbKJJyIiMJKeCwDEvkwLy/O57Ujzb0BjGTifFPePuz1Dk0BrFGxAICzi4Kze1tuggYAcDDlSo/xmP5W5HZtAwDYHOG3j5+IQpvTJUIQfVd4IhoOq8hTWEsaYe+z0yVCLpP+gOJ4nDiMOZIrZktfDcWXE4dE23ST0KSbhhRTGarjFuJA6lWojhs4X/LRgSZcdUamlGESEXmo67TggmOPIqN3P9q0xWjTFqM8finkGTOlDo1CAJNwCmsjVQCxOZwh3dGsYpTNbFZMTUFeonaCoxmf5VNSsL6sBRsKHoElKhYmVarH9ZoOi49HEhFJ41BjDxaYj0Fva4be1oyCzq9hjkqEMu4sqUOjEMDtKBT2JqXGDBnT2lqh6e/AM19VShCRf4iiiA/2NQIACjo2YV79Wshcdq9zp2f498D0RJiRaYA+OgqtuslDEvATwrG0JBGFrsrGdsT11XiMteqKsSCfjXpoZEzCKewtLkoEAMT21WJe/Rqs2n8b7tl1MWY2vy1xZONT2mgEAEQ5zFhW+Rcsrvknbt53EzJ6dnvMi9OETpmsK0fYMtNl6R/2OhFRoPQ7XEi0VLhL4AKACAHtmkIowqTyFk0sfpdQ2ItRDWw3KWleh8U1/0Kq6TAAoKDjKwBAizH0urGKoogvDrcAABbUPYuY/lYAQEJfFa47dB9Sew+55149J3T2Ucdrh2/OteFIa4AiISIanqXfgVbdJLw46zV8UvR77Eq/CYeTV8IhD74KVBScQnczLNEondgXXhG/FHMaX3WPJ1mOw2Ctx7ZKLS6fFbyHFr359+aBbTTa/nbManrT41p17Flo1k0DAKTHqkOuYcS0dL17lf9UrUZbgKMhIvKuobsPoqBAh7YQHdpCHMFKAEBRik7iyChUcCWcIkJ+khaN+hJYok6WLLTKdYi3VKOyzTzMI4OPyyXCbBvYG21WJuKNGc+hRTsFAOAQlNiY/7OBhkQArp+XLVmcp2tyqn7ImM7WgsmtnwAAHE7XkOtERIH2eWmL1/GFBYkBjoRCFVfCKSJcPCMN/9hgxqGUy6F0mFCRsBT1+jnuNu6h5NvKDo/7LTHT8NrMNShpXgeFy4bu6IHE+8YzQy8BB4Cs+IG3cuUuGya3fYbJbZ8gq2c3BIhoMMzGxqMGrGAbeyIKUiNtqyM6IfQyEKLTcOKQzJacB71e77b0I1YT/D84azrM2FHVOWRcFOTYn3ad+/6srFik6NWBDM1vBEFAVrwGje39OKfqb1A7TzZVmtT2GXapUjE/Lx6G6NDaZkNERDQYt6NQxLhgmveydwCwZks1XC4xgNGMndXuxDt7GkY1d+mk4OyMOVrLJiXBKVPieOJ5HuNT2ga2pGzkAU0iIgpxTMIpYkxNP2WvsShC098OiAPJ96ZjwZ3YvbK9dlTzZmfHjtikKNideDu3LOlC95hNrkWzbhrkLhuq2kNrHz8RhR9tfzvkztCrrkXBg9tRKKLcc04+nv26EnKnFVce/hGyjHvQrc7AmzOew/46YNmk5KBNYI19A414BNEBQYTP/eznFIX2KjgwsCXlpjOz8co2Fw4lX4bquIWojFsEp/zkFhur3Ql1lFzCKIkoUomiiOXlf0Bu11Z0ReegTVuMgylXIG/ehSM/mOg7XAmniKL9rmb4pPYvkGXcAwCItTZgaeVfAQAHG3oki204g7dfzG58HTfuvwWpvQeHzLt0ZhpksuD8I2KskvVqQJDhi6Lf4HjieR4JOICQq2pDROGjttOCJPMxyOBCQl8VJrd/Bl1/G87Ijhv5wUTfYRJOEcklKCDiZLJa3LEemv52fFnWClEMvr3h++q6AQB6ayMW1j6DJEs5bjhwF5ZV/A+UDtPAtegoFCbHSBhlYH1T3iZ1CEQUoT7eXupuknZCm644bBZBKDCYhFNEqopfDIfMsxrKtNYPAQD1XX1ShOTT418cG7ghilhW+WdEuQb2IAoQMbN5HRIsA417rjkjdDpjjtas7Fif107USiciCrS4vhqPhRyHoERndI6EEVEoYhJOEefcycmwKWJwNPF891i3OgOWqHgAwMcHm6QKbYhXBx3GVDl7oev3XP09kHoVmvQlAACDJvxK9i0ZYX97XaclQJEQEQ0QRRFN+pl48qyv8FrJanyZ/wi2Z90FUeAxOxobfsdQxCnJNGDDkVYcSL0aKqcZB1KvQq1hHiAM/E1q6XdCFEXJD2iKoogW48mT9zaFHq+VvIDZja9hQd0z6Jfr3HXPw7V5jUwmoChFh+MtJq/XPz3UjLvPyQ9wVEQUycpbB34eOeTRaI6ZgeaYGRJHRKGKSThFnBPJdUvMNPxn8v94nbO3rlvyAzZ/W398yJhLpsDuzFtQnrgMOlsLbIqBPeDTMwyBDi9gLpiWiuMt5dBbG5FkPga9tREGWyNatcU4nHKZ1OERUYT5yMe7pReXpAU4Egp1TMIpIt11dh6e31zl8/pXR6U95e6tK+ZgPepM9KgH9oBfNis9ECFJJuq7bqczmt/B/IYX3OPVsWfhcMplQfGuBRFFDl9n9wuTdIENhEIe94RTRNKrR94/PXgrSKBtKW8f9dyCCPnB36ab5HE/s2c3ohxmbDwa3E2WiCh82By+D4SzMgqNFZNwIh/e3l0vyeu+tatu1HPvWJQ7cYEEkaWTklAduwDOQQefFKIdud3fYn9dcNZ2J6Lw02t1SB0ChREm4RSxlk/xfphR5rJD29+OfocLdqcroDFZ+h3uEokJlgosrfyLuw74qS6blY5YjdLrtXAzKysW/Qod6gxzYVImYX/q1Xhn6t9RGX+O1KERUQQpbTQi2XQE8ZYqCOLJVXF9dPhVp6KJxz3hFLFmZBqwvqzFfd9grceM5ncxtfU/aNFNxftTH8eTG8rxwLICqBQT3x7daLWf3KcuurC8/I9I7z2Awo5N2JD/c1QmLPGYHynbUICTh2k/Kf4vWBV6dyWbE1wukW8FE9GEM9scWFL1f8g07oVDpkK7pgDbsr6HgkXXSB0ahSCuhFNEu3B6KgAgr/Mb3Ln7SsxreBFaeydyu7YixtYMYKAM3kRzuUSPg6Ilze8gvfcAACCmvxWXH/kpprR+7L5+56K8CY8p2MzPi4c1KnZIAg5It3WIiCJLbrwGieaBylUKlw2ppsMQIGJqul7iyCgUMQmniFacMlDir95wBmxyrXtcBhfm1a8FAFS2mSc8jsEr8nJXP86qe87jepc6C8cSznXfD8fGPCMZrlpNQ3dwdTklovD07Z69UDs9twi2aosh5ztxdBqYhFNEO/GD0y7XoCzpIve4U1DALte4a1Htq+uesBgq2kwobTSefG2ZEm+UPIea2DPdY18W/AJOuRoAcP28rAmLJZhFK4ffEuRy+agbRkTkB3anC9r+dvQqk91jfQoDTMrwbJZGE497wini3XhmNl7dXoudmbdjWut/UBG/BFty7odRneGes/FIK6an66GQ+/fv1naTDR/saxwy3qPOxDtT/4HJbZ8g2XwUdbHzAABxmiikx0b7NYZQkhEXjYYu76veDd19yIrXBDgiIooUR5t70aQvwXPzPkK0vQtJ5mNQ23sA9img08QknCJein5ghdmkSsHqOe/BokzwOu9wkxElmbF+e12TzYGXvq3xPUEQcCR5JY5gpXvo5rNy/Pb6oWjFlBSs3VrtMRbbV4toezfe3g08tKJYmsCIKOx1mPvdt/ui4lD73buVMWqmUnR6uB2FCMCq+dkA4DMBB4Avy/zbFGbjkZPPF2NrhsxlH3b+pTPT/L4SH2ritAMlGQ199VhU/SRu3XMd7thzNZZV/gUAIPpqZUdENE6HB20bHGxmVmxgA6GwEdm/0Ym+k6JXjWre418cQ7vJNu7X67b0o7x14HCPzOXA5Ycfws37bkJGz26fjylMjhn364aLGFsz5je8gIS+gYoyKeYj0NmaUdtpkTgyIgpXhcney8KWZBoCHAmFCybhRBioQ33NnMxRzX3p2xrsre06rdcRRRHtJhvWbKl2j81qegNJlnIk9FXhukP34YJjjyLK4VmR5a6zI68koS+3LMhBg2HWQL3wQQo6v8aW8g6JoiKicHeowXt33kD0kaDwxCSc6DvDHerL7NmNgo5N7vubjrZh/eEWn/O9sdqd+Nv64x77wHW2FiyofdZjXpL5GBzykyvziTol9OrIK0noS6JOBVFQoCpukXvMHBUPmehEi9EqYWREFK6aelgGlfyPpwmIBsmIjfaoOZ1gLsfimieR37UFvcpkVMSf424Wc7ChBwcbenDjmdnuw53eOF0ijjQb8Xnp0KRd6TSjKzobKeYj7rEvCx6GKJz8X/OG7/ark6fDyRcPfE0SlqBZN839dRFF0d1hk4jIH17fUYf8jq8AQUCrtnigLKEg4Iwc3/0LiEbCJJxokKvnZOKJLwe6ocVbKnHrvlXuazH9rcjv3Dykffyr22uxfEoKpqXrh7ROr2o34729DT5fr1OTj9dmrsXMprexsPYpHEtcjib9TPf1e5fkIyrCD2N6U5wSg2M4012dYLC6zj5kJ7BUIRH511l1/0aK+SgAwKrQY33BLzD77LsljopCGZNwokHkMgH3LsnHM19VojM6Dw5BCYV4siyVwuX9UOb6shaPrpdjIQpy7Eu/HscTz4NDOLnt5EfnFQ1J6mnA8qnJONbS6/Val6WfSTgR+ZXM5UCCpdJ9X+0woi8qjlsFaVy4xEZ0Co1SgXOKkwBBQL3hDPf4vtRrcCzp/Al7XbMyEbaogVP2CpnABHwYKoUcOpX3NYQNR/xbSpKIKK6vGgrRs4xsm7ZIomgoXDAJJ/Ji5nclp74o/DVatJPRGZ2Dzbk/Ov0nFEXkd3yFi488ghXH/wsxtuZhp999Tv7pv1aEmJzmu2Qj64UTkb9Y+h1QuGyo18+GVT5QprBHlQbbKRWaiMaK21GIvFDIZd8d0kzB6yWroe1vh0Pu+/ClIDohCt7LVKnt3bjy8A+Raipzj+ltjVg3/Smv8+9dkg91FEtejSR1mMOw++q6MTubB6aIaPz6+p1oiZmGt2Y8C4gi9LYmaOyd7FZP48aVcCIfrp07UDfcJYtCrzrN57x4SyVu33MNJrV9OqS+NwBYFQYoTxnvi4qD3Mv+8stnpUOj5N/Go1GQdLJxhiA6kNp7CPPq12Jy6yfYdLRNwsiIKJw4XIPeWRMEGNXpaI6ZjjP4hz6NE5NwIh8EQcD9SwuGnSNz2XHJkUcQa63HymO/wcLaZ7w9ETbneW5libE1Q4TnardGKUd+kveObDTUiT3z05vfw/3bl2PVgTuwuOafmNHyLgBuSSEi/yhr8t6uPlbDQ5k0PkzCiYahjpIj1eB728PC2qfdrdMBwCFTep1XGXe2x/1DKZfDJTu54h2vVeLeJcMn/DTU5bPSYVYmQOU8+U5Dau8hyJ1WtPV6r2RDRDQWe2u7vY5PSeOecBofvu9NNIJV87Px+BfHhowLohOJlnKPscElrDwnC3h27sewKOMgQuZuLAMAqQY1rp2T6deYI0Veohaf6GfDBRlkcAEAFKId6b0HYLUXShwdEYUz9nCg8eJ3ENEofP/cQkzPMHiMiYIc7015HBvzfgqHbKDNfEHXZigdJq/PYVYlDXTCHJSApxnUWDU/Gwr+MD8tgiCgX6FDq24yAKAjOg/7Uq+FJSoeIrgdhYjGp9dqH3kS0WniSjjRKETJZVgxNQWWfgcq2wYdshRk2Jd+PY4kXYAUUxladZPRrxjdvu45OXE4uyhxgiKOHBlx0fii8NewRMXBojz5+dxe1YmcBK2EkRFRqPu2ogMzm96EzOVAu7YIbdoiWKNiMTnVd4lUotFiEk40BpfPysDz31TB2Oe5OmKNikVN3IJRPceF01NRnBIDOZvx+MWVszPwZFffkPGGrj50mfsRp/W+T5+IaCRV7WZc3/gaYq317rFPiv4/LFz0gIRRUbjge+BEY3TX4jxcOD31tB6bl6jFlDQ9E3A/Gm5f5ro99T6vERGNxG4xeiTgANCpyYOBlVHID7gSTnQapqTp0Wt1YEt5+6jmx2qicOXsDMRquCobSL1Wh9QhEFGIEkURSaccvncKcnRo8iSKiMINV8KJTtP8vHj84NxCnD8txV0vNlopx8wszwOcM7MMuG1BLhPwCXTu5GSf15wuHtAkorGr6+yDTa7FweTL0aKdAoegRGd0HqDwXbaWaCy4Ek40Dgq5DNPSDZiW7pl4nzs5RaKIItOMDAM2HGkdekEU0dprRZohOvBBEVFI217VgQ5tIdYX/RrAQGdeTX8XipLZVI38g0k4EYW8E90zo+1dyOzZjayeXcjq2YWvc3+Mb45rcO3cLIkjJKJQc+p2NlFQwKxKwpn5CRJFROGGSTgRhY2LjzyCLOMe9/25DS/irfjFEkZERKGqp897jfB4VlwiP+GecCIKG00xMzzuZxr3AqILln4e0KSJUd1uxof7G1HZZuL5AyIaE66EE1FYmJsbh7rueZjf8ILHuNJpwe6aLpxdlCRRZBROeix2NHT34bPSZo/x8taBTrnT0vU4f9rplTAlosjCJJyIwsK83Hg8XTXXY6ws6SL0K3Q40tTLJJzGRRRFrNvTgLpOy7DzShuNKG004sYzs5GiZxWNUCWK3t/VuGZOZoAjoXDGJJyIwoI6Sg5RkONf89djessHsMs1OJRyGQDAZON2FDo9NocT/9pYMebHvbq9FjkJGszJiUNOgnYCIqOJVNtpwTlVj0PhsqFNW4xW7SS0awqQFa+ROjQKI0zCiShspOjVaDECuzNvkToUCgMbj7RiX133aT++psOCmg4LrpuXhYxYlskMJd+Wt+Oytk+gsXe5x/4z6TEAJdIFRWGHSTgRhY25uXH46ECT12sOpwuKYVrcEw1msjlGTMAF0Ymz6p6D2t4NlxCFg6lXoFOTP2TemzvrAAAPLiuEUsHvwVBgbG/wSMABoE1bLFE0FK7404CIwkZhku8mGrtqunxeIxqs3WTDv7+uHHGeKMgxp+ElzGp+G2c0vYbb9l6P3K4tPuf/c2O5z2sUXOIs1R73+2XR6FZzPzj5F1fCiShsnGja4823FR04i002aAT1XRa8tat+1PP7ouIQZTtZKUXu8l5b+oROcz/rTAc5o9UOa5QBB1Muh9phhNreg36FDkvYCZn8jEk4EYWlaHsX0owHkN57AFHOPmws+LnUIVEIGJyA62wtOLv6CWzK/yn6ouK8ztcPSsCbdVNREb9k2Od/YWs17liUi1gNE/Fgta+2G+3aIqwv/LXH+IPpBokionDF7ShEFFbuXJyHzO5duG/H+bj8yE8xr+FFTG95H3JXv9ShURATRRGPf3HMfV8Qnbjm0P2Y3P455tWv9fm40uRL3Ld3ZtwGCN7fjUk2lblvr9lSDRcb+wSt3T62rnE/P/kbv6OIKKwYoqPQppvkMaYQ+5FkPipRRBQKak+p/13c/gXirAMHKmc2vQ29tcHr4z4v+h2eWLAFnxT9f6hI8L4KPqntU9y0/1YsqfwrZK6Bcpl///K4z1rURBQZmIQTUdixKWLQfkqVinTjgXGVm6Pw9s6ek0m2IDqwuOaf7vsKsR937L7K52OdMiWOJF8EUZAPuZZkOooV5f8NADij6XVcVfoglI6B7pp1nX3+Cp/8hO9QUCAxCfcDp9OJAwcO4Pnnn8f999+PuXPnQqlUQhAECIKApUuXSh0iUcRpiilBvywatYa52JZ5J+oNZ2DjkVapw6IgdOqKtCgo8NGkx9AvO9mYZV/adWN+XoWzD5ce+TmiXDb3WJZxDxbV/AsAsG5PPTpMNl8PJwk4+e4EBRAPZo7Te++9h5tuugkWy/CtjIkocG46MxtvO36ALwsehijwxxz51tjdhze+q+M9WHPMdLw39XFcefiHaNFNxdac+8f83A55NLZl3YULyv/LYzzBcrID54vf1uDHy4sg+NhLToFV3moCRBcgeK5RslMmTQT+dhqn7u5uJuBEQSZZr4ZNoZc6DAoB68tafF5rMJyBt6c/jXZNIRxytc95iTolZmbF4suyoe+0HE65DH1RcZCLdrRpimFSJsJ5ynNtrejAosLE0/8gyG8+PdSM6w7dC7mrHw36WWiMmYm62LlYOilH6tAoDDEJ95OUlBTMmzfP/e+zzz7D3//+d6nDIqJTiKLIVUcCAPQ7XOgwDV81pzlmusf9xBgVbpqf7bUmfUlmLF7YWo1Os+dzVsWfPexr7KjqZBIeJOROK1J7D0EuOpBqOow5eBXrpj2JRN1cqUOjMMQkfJwuvPBC1NTUIDs722N8+/btEkVERMP5/HALLpiWKnUYJDGH0zXmDpZXzs5AbqJ22Dm3LcyF1e7EU5sqhp13qse/OIYfnVc0bMMpmlgOpwuppsOQiw73mAsyNOmmD/MootPHg5njlJqaOiQBJyLp5flIlg43GmG2Obxeo8jx2ol94KILZ1c/gdi+Wp9zZ2QY8NCK4hET8BPUUXLcsSh3zDFtq+oY82PIf3ZUdyLduN9jrE1bhFmFWRJFROGOSTgRhaWFBV5a1IsiIIpYu7U64PFQ8Oix2NHeO1CVZErbJ5jb8BJu2XsDFlU/iSin5xmfqel6LJ869nblsRolHlpRjJUz0nzOyeregeXlf8Dk1o8BANsrO+Fwusb8WuQf2ys7sS/9erw97Z/YmnUPqmPPQk3cAizw9rOEyA+4HYWIwlJSjAoAkNW9EwWdXyHFVIZkUxk+nvRHVCQslTY4ktTqLVUAAKXDhLOrnwAAKEQ75je8gLTeg3h7xjPuuePdujQpNQb6aAVe3+FZgSWreweuKX0QADCj5T1YFQZUxy/CPzaU46EVxeN6TTp9drkGdbHzURc73z02n1uEaIJwJZyIwtKJw5fXlD6A2U1vIL33ABSiHZcd+ZnEkZGUBh+anNPwMrT2To/ruzNucd/+8fIiv7xmql6NaOXJRj6C6HAn4CdcWfbjgdJ4JAl2LyUpMAknorB1x6LcIZ0zzVHxAPhLN1JtKW93396dcTN2pd8M53edLivjFqMqfjEA4P6lBX6roiMIAm6Yd3JfsdI5tFNmvX62+/bqb6r88ro0eoebjFKHQBGISTgRha1YjRKfF/7WY2xf2vUAgPoutgyPROWtJvftfoUOm/N+hJdmvY7KuMXYlPcT9zV11NAW9OMRq1FidnYsAMCmiMFrJavd17rUWfhw8p/dDWJ6+uxwsn16QH1e6rtePNFE4Z7wIGaz2WCznWxpbDTyL3WisWqJmYZPin6PKW2foFU3Gbu+226wr66bXfAiTIvR6nW8S5OL96c+7r5/xeyMCXn9pZOS0Wq0oaG7D80xM/DU/C9gsDagXVMwpIHP3touzM2Nn5A4aPTSDL6bNBGNF1fCg9hjjz0Gg8Hg/peVxTJJRKfjSPJKvDvtH9iS8yBcsigAniuiFBle3e67DOEJxSkxyJnAP86uG7QtxRoVi5aYaUMScADYfLx9SNMfmjhKhwmxfTUDFZQGuWZOpkQRUSRgEh7EfvGLX6Cnp8f9r66ubuQHEZEHj1KFoguxfbUobvsccqf3VVEKT65Rbu+4uCRtwhvm3H1O/siTALzAUpoB4XSJyO3+FnfsuQb37rwAl5b9DCVN6wAACjnTJJo43I4SxFQqFVQqldRhEIW0Wdmx2FrehqsO/xApvaVQOwdWwF+f8TxEcQZb2EeIF7+tHnHO4qLAtI7XqUb/q9fudCGKieCEajZa3U16NPYuFHZuglzsx4G0qyWOjMId/88morCmUsgBQQZtf5s7AQeAZPMRVkSIENXtZnRZ7Ijtq8X1B+5ETte3Q7YdAMDsrNiAxXSPj9VwpcOE+XWrkd/xFQDg5W01AYspUhn77EM6ZTbGzJQoGookTMKJKOxdOTsDLbopHmMppsOsiBAh3t3bAABYUPss0nsP4qrDP8S1h+71SLwuKUkL6NYD7Smr4SeS77t2X45FtU9hUe2/ANGFbos9YDFFqq3HmpFkPu4x1qifiaWTkiSKiCIFk3AiCns5CRq06Ka675uj4mGXsTJKJOj67nBjovk4Jrd/5h7PNO5FfufX7vuFybqAx3b7wlz37ezuHVhU+xTUjoF3ZxItlSjq2AAAqOu0BDy2SGK0iXh6/md4eebLeH/y/2JD/s/Qqp2EKWl6qUOjMMc94UQU9gRBQGXc2TBNTkaLbgpMymSAe8EjwontHCXN6zzGbXItdmbcCgD43tl5kpwNiNMqsWp+Nl7bUYvyhKVo0xQiyVLuvn5W3XM4nnAe3t5dj++fW8i94RNAFEVAEGCLMqAtyoA23ST3NX/Xiic6Ff+PJqKI0KtOQ0XCUphUKR4JeL+DrcLDmeO7qigb83+Kzwt/A6MqFcBAt0xblAEapRwx6ijJ4ks9UYdakGFb1t3u8ePxS/Fp0e/d36uljTy/MBHYFImkxJVwIopoR5t7MSPTIHUYNAEG/4ElCgqUplyGI0kXYlrL+yhLWgkAuPmsHKnCc7t2bibe2lWP8oSl2JO2CoeTL/ZYkQWAjUdaMSuAB0cjRR0755KEmIQTUUSIVsrR1+8cMm5zDB2j8PDPjeVDxpwyJQ6kXQsAiFErhhyQlEJm3HfnEwQZvsr/ic959V2Wk3PJL74+1uZ1/IJpqQGOhCKR9D99wsDKlSvR2NjoMdbc3Oy+vWvXLsyaNWvI4z7++GOkp6dPdHhEBODqMzK9lnvbW9vN9uBhqLln5GZMl5QEz8/fB5cVev2jYbC3dtXjoRXFAYooMvjqSlqcEviDuhR5mIT7weHDh1FT47uWq9lsxv79+4eM9/ezJTFRoCRolV7HTTYHrHYnD2GFmdd2jNyi3r0fOwgoFaM7omWyOcbU7IeGp7c2IMbWjDbtJPQrTibe7JRJgcD/k4koIgxuRR5t70KyqQwppjJUxC/Bnpp4LCwMTLdECg4XTg++7QbLJidj45HWYefsqu7E0knJAYoovHWa+zGp/XMsrvkXAKBLnY0jSRdgW/Y9EkdGkYJJuB9UV1dLHQIRjdLlh3+M/K4t7vsOmRrbqwqZhIeRE/v8J7d+DLtcg+q4BXDKVB5zJqfGSBHasGZmGrwm4fGWKiysfRpVcQuxF5czCfeThq4+JJuOuu/HWWsRbe9GemzwvENC4Y1JOBFFjKx4DcxKz2Q72VQGYKBesBS1osn/Pj3UDIgunF3zJHT9bbDJtShPWIqdGbejS5OLC6alBuXXWhAE3DA/C6/vqAMA6GzNWFD7LKa2fgQZXIi3VKE05XIcaujB9AxW9Bkvo9WO6d/9/39Cq24yzilmp0wKDG56IqKIsXxKskfnTABI+e6XsN3JesHhwOZworLNjAzjPuj6BypfqJxmTGv9CAIGShYG80pnmiHafTvFdATTWz+E7Lu4E/qqIHf144vDLVKFF1Z2VzTDqoiBUzh5HqRFNxmp+uD9/qDwwpVwIooYsRolWnRTAAB2mRqt2kkDSbnIhj3hYntlJwCguP0Lj/F2TQE6NfkABr4PgtmJuuH1hjOGXEs2laFJPxMul+hxzoHGxmxzwClT4tVZL0Pu6keCpQIppjJ0RucH5bskFJ6YhBNRRGnXFOHFWa+hU5MLUTj5I/BAPUsVhoPdNV0AgKaYEiRYKpFp3AsBIo4mng9gIMENdhmxA6vhNnkMelRpMNiaYFSm4EDaNeiMzgUAfFrajJUz0iSMMrRtPt7uvu2UKdGqm4LW7/5AJwoUJuFEFFFcMgU6tIVDxjcfb2cSHuIGN146knwRjiRfBK2tFcUd61ERvwQAQqLZjSAIuH9pAZ7aVIEtOQ/AKShRkXCOxx+NR5t7sXxKyqhLG5Knsiaj1CEQcU84EUWWZZNPVpaItnchr3MzFtY8Bb21Ab1Wu4SR0Xjtq+0eMmZWJWNv+o0wqjOCsiyhLyfq1h9NuhDlied6JOAnfFnGveH+duOZ2VKHQBGEK+FEFFFOlIG79uA9yDTudY93qzNQ2liCs/ITJIyOxmNrRYfPa4IATEnTBzCa8TuxN9yXI829uIhbUvwqhYcyKYC4Ek5EEeXEoStLlOfWk/Teg/h2mCSOgttI72KEYm3t0WydcblY1YcoVDEJJ6KI1BQz3eN+au8hiSIhf/hm0EE7b6alh9Yq+GgdbemVOoSQU9lmQmb3Lkxr+QBJpqOQubgNjaTB7ShEFHGWTkrCUeMMAAOlCpt1U9GonwmIIhxOFxRyrk+EmiPNvZC7+iFz2WFXaIdcjwrRr+nUdD0ON3oeIpS5HCjs2IB2bSE+PTTQ/ZNl9Ubv/X2NuKD1Q0xt+xgA4BCisCPzDpRPfVDiyCjSMAknoogzPcOAzWVT8PLMl9GuLfA49GZ1uKAL0YQtUpltDgBAfudmXHj8d6iKW4SjieejMm4RnHI1Lp2ZLnGEp+/8qSnuJDy6vxMzWt7FzOZ10PW34VDypfii6LfYXdPFyj5jlDKoU6ZCtKMvKtZdGpIoUPibhogiTpRcBqdMiTbdpCFVJ6x2p49HUbD68kgrAGBS++dQuGwo6tiAS44+gkuO/gIAkKAN7uY8wxEEASWZAy3qS1rewaLap92dQCe3fYZoe5dHzWsantMlQuHsQ1xfjcd4q24y5uTESRQVRSom4UQUkWI1UV7HT33rn4JfRasJUQ4z8rq2eI5/Vxs8LoSTcABYVJgIADiYciUcwsnvW4XYj2ktHwAA+h3s+joaVe1mqJxmlCeci251BgDABRnaNUVB30mVwg+3oxBRRErVq9FtGXoga3dNF84pTpIgIjodJ965SDUdhiA63ONOQY7jCctwzzn5UoXmN+ooORYVJmJLOXAsccXJvcwyFWKtdQCA4629mJZukDLMkPDh/kZAmYiPJj8GAFA5jIi3VMMhZ2lCCjwm4UQUkc7MT8CRZu+VJfr6nYhWygMcEZ2OA/U9AIC62Hl4dt6nKOzYhEntn8MuU8MWZYBWFR6/5oqSddhS3o5tWXfjcPIl6IrOgkmZDAgDb2h/XtrCJPw02BR6NOlLsGJqitShUAQKj59ORERjFD/MFoXdNV1YXJQYwGjodG0pP7kf2hoVi0OpV+BQ6hUQxPDa239iS01PdCZ6ojO9zumx2GHwsc2KhheuJSwpuHFPOBFFNJnLjpTeUsxqfB0XHf0Vzqr9N9pMVqnDolEQRd+NakRBHnarmxdOTx32+gcHGgMUSWgSRREZPXtQ0rQOyaYjkLlObl9iiUeSAlfCiShi3bYwF6Vv/n84u+Yf7rEW7SRsa79bwqhotPZ/txXFl3Bb3ZycGoNPDzX7vN7ea4MoikwofTD2OTCl7WPMaHkfAGCXqbA7/WZ8m3OfxJFRpOJKOBFFrHitckjnzCRzORTOPokiorHY+F1pQl/CLRkVBAFn5g1fD/zzwy0Biib0rN5ShdTew+77US4bbIoYzB/hc0o0UZiEE1FEa9FNhQsnD2HK4ESK6bC7AQwFp+G2ogCALMwS8BNmZcd6HY9ymAFRZIlNH1wuEVFOCxIsFR7jzTHTcEY264OTNJiEE1FEc8jVaNMWoyM6b6ADYcEv0RWdgx1VnVKHRsPo6RsoL7m4+h+Y0voRlA6Tx/W7z8mTIqwJp1EO2kUqupDVvQMXHvst7t15AVJMA6u8Ltfwf6BEotZeG5ROC44mrkC3euBgqwtytGonsxISSYZ7wokoohWl6PB6yfNwyTyrSuyr68ayyckSRUUjWbOlGjpbC+Y1vAgAcAhKVMctxIb8n0MVn+GZrIapaw49gCzjbvf9aa0foiVmGrZXdWJBQYKEkQWfvbVdMCsT8emk/wYAqO3dSLBUsj44SYor4UQU0dIM6iEJOIWGovYv3bcVYj8ye3bBGmXAkjBvtnTD/CwAQJ1hjsf4pPYvIHfZsK2yQ4qwgtqpPQGsUbFoMJwhUTREA5iEE1FEm57BBiehprF74ODspPbPPcYrEpbCKVMiO14jRVgBk2aIBgCUJV/sMR7lNLsPHnaZ+wMeVyi6Zo73mutEgcAknIgimkrhez9ou8kWwEhotN7YWQeIIvam34CKuLPhFAa2nhxNPB9RcgEyWXgeyhxsdnYsjOp01Brmok1ThE15D+Hfcz9Gg2E2AGDTseErx9CAFD23o5B0wn/THBHRafrqaBuu5kpZUHFXRREEHE26EEeTLoTKYURB59eoM8zDmbmRUW5uSXES9tZ248PJf0G/QjfkenW7RYKogtNwlXQi4O81CmJcCSci8kIQHajtZCITbJp6hnYztSn0OJx8CVwyBWZlxQY+KAmcqIHuLQE/obaD378A0Nbr+x0thZxpEEmHK+FEFPEKk3Woa2hAbvc2pBv3I713PwDglVmvsANhkDkwTJfMWdmxUEdFTrm52xfmYu3Wap/X1+2px0MrigMXUJB6c1cdllT+H1yCHM0x09EUMw0mZQrA/69JYkzCiSjinTs5GR+Wb8XKY792j4kQoHL0ot3Uj6QYlYTR0WBlTb6b0SybFFklJeO0yqGDogsZxv1QOYyoTFgCs80BrSqyf9U7HA5Mb3kfStfJdwbem/I4Zi+/XsKoiJiEExFBq1KgVTcZDkEJhThQVUKAiLTeA3C4pkocHZ1gczilDiHo3DA/C6/vqIPe2oiZTW+huP0L6Ptb0K3OQGX8Odha0YEVU1OkDlMy3ZZ+xPdVeyTgANCqm4ysuPCuokPBj5uhiIgAOGVKtMR4JtzpxgM84BZE1h9uhaa/A9H2LqlDCRonyhVG27sxt/Fl6PtbAACx1gakmkpxqMH39p1IUNtpQWpvqceYUZkCszIxIqroUHDjSjgREYCCZB2qYhfCJtehUV+CxpiZaNZNhbOyg90Hg8Sxll4sbnwVcxpeQW3sPBxNXIGKhGWwKWKkDk1S3zs7D899LaJLnYU4a517fFLbZ2iOmQ6TzQFdhG5J2VHVCZVuCrZl3oVUUylSe0vRHDNd6rCIADAJJyICAJw/NQVPtd7h9ZrTJULOVTNJ2RxOQBRR3P4FZHAit3sbcru3YWdfDb7J/QFuOitb6hAlo1MpBko2Jp6Ps+qfBwB0ROehK3rgc/Lvrysj9oBmr9WBXm0R2rVFAwOiC0qnhU16KCgwCSciAoatqnG0uRdT0/UBjIZOdbS5F6mmQzDYmjzHE88HACTHRG7TlRPVe8qSL4JMdOBo0gVo1xRGfPWPYy29QwcFGfoVOmSFeVdVCg3cE05ENILyNpPUIUS8FqMN8X01sMtOJtud0Tlo0xbjgWUFEkYWHC6Yloru6Bxsyf3+wKrvKQl4JLax/+hA08iTiCTEJJyIaAQVrUzCpXaooQeHky/B0/M/x0fFf0B5/FIcTr4YEASoFJFTG9yXKWnD74sfrp44EUmD21GIiL5z64IcvPhtjdRh0ClMNof7tkMejWNJ5+NY0vkSRhR8RtNQqq/fiWgl/2DJTeRWFAoOXAknIvpOgu6UpjyiCwmWCqgcXvaWUsC8s6fe57W8RG0AIwluty3MHfb6umE+j+Gm1WhFbF8NLj/8EEqa3oLBevJjv3BamoSREZ3ElXAiokFkgoAZjW8gt2sr0nsPQu0w4uPi/0JTz1R3TWYKrA6T7/3MS4qTAhhJcIv31kETgMw18E5CWwT9LfnK9lrM7tqK/K5vkN/1DQCgTj8Hb894mu8GUNDgSjgR0SBpBjVyurYhv2sL1I6BFunpxv14f1+jxJGRNxoVEyqvRBcyevbg3Io/4e6dF6Go40upIwqYE51Vc7u2eoz3qNOlCIfIJybhRESDLChIQKN+psdYeu9+9PWzZboURFEc9joPZXrKSRjY73zRsd/iukP3YmbzOmgc3Zje8r7EkQXOF4dbIHdakWnc6zFeHbcQN8zPkigqoqGYhBMRDZIRGz0kCVc5zJC57BJFFNk2H29HgrkcsxrfQGxfrcc1djId6vJZGQCAyvizPcaze3bCYK3HntouKcIKqIauPkAQ8Hnhr1GWdBEsUXFwQY7a2DO5pYyCCveEExENIpMJaNFNwd6069EYU4JGfQlMqlQAA6uyo6lCQf6zu6YLC9u/wJn1q4EqoFudiYMpV2JX5q2YlBLZ7eq9OdHZtTxhKawKvXtLlQsyZBj34aujmchP1CJW433/eDiw9DsBmQpHky7E0aQLAdGF+L4a2BT8fqHgwpVwIqJTOGUqbMr/KY4lne9OwIGBt7kp8Abv7Y211kPt6AEAxPk4iBjpZmfHwilToSzpIvSo0rEl+348N/dDHE6+BADw8rYIK8MpyNCpyZM6CqIhmIQTEY1SaaNR6hAiisslQtPfgRTzEY/x6riFEkUUGs4uGqgYsyXnAaye8y52ZN0JsyrZfd3uFOF0Db/XPlQNd4ZgcVFiACMhGhmTcCKiU9y3hG3Qg8GR5l4oXDaUJl8Cc9TA/m+bXIvGmBKsmp8tcXTBSy4TMCVND7tcAwjef80//01lgKMKjL+tP+7z2owMQwAjof+/vfuOkrK+9wf+fqZvne29Fzos0osUQSy0IOC9igpBsZBorsm9SW4Sf2pMTrzx3Jyr92KLQSGaWFGQQECKCLoUqYsLu8CyvffZOrMz8/z+GJhl3J0t7MzzzMy+X+fsOU/d+ZCMu+/9zvf5fKl/nBNORPQ97CMsP6tVxN7cKkAXhy8ynwNEKyLbLiOkswxWhRoxep3cJXq0+SMjcbHS+Sc3bcbh1+1Hp+Z/1+RZOBJOREQep6yxw/GAoEBt4EhcjlgoT0FeRqdWIkjX9zhbjaFTomqko7B2wa/L9zvAkG9gCCci6oe2qxmpDUcwovYL1LUa5S5nWNh/0flDsGPjgiWsxHutnpzQ5/nPzpRLVIk0aluMiDecweMn7sT959ZhZsmbiGn5DgAwNSVM5uqIeuJ0FCKiXkxKDkV1zgEsLPgvhHcUAgAM2hhsO7cE62ez04K7NXc478u+cHS0hJV4r++3IVRYzUht/BrjqnfgZPxDKNdPgtUqQqHwjbab7x0rxpzGbAgQEdN6ATGtF5DceBQfZG3B7Az2lCfPwxBORNSLuZkRePei3h7AASDYWAVzYxlEMYX9wt3o+rLjzih9JDRKaWLFh5hW9jYCuhoAAEZVEMr1k3C5phUjY3ynf3ZK01GH/eLQmQDA/17JI3E6ChFRLwRBQL1fKjqVjgElznDOZ9u7eYptp5xPk1g6IVbCSrzfmGtTdxSi2R7AASCz/iC0ZgN2n6+UqzSXulTdggBjDSLaHbu+FIbOwu385IQ8FEM4EZEzggIVwRMAAF0KLUqDJ6FL6YfihnaZC/Nt1YZOjKzdizVnH8Ks4tcQZzgLQTQDADKiAmWuzrssGGXrD34xajEsQveH3yqrEaNq9wDou7e2t9iVU4lOtR5/y/orPh/1Er5M/Xd8G78W1YFj+AwBeSxORyEiciJIp8KxxEdxLPFR1AaMhFVh+5FZeLYCP100QubqfNP1TxlSG79GdFseotvyML3sHeSH347do17ktIJBUisVGBEdhEvVQEHYfIyo34+KoAk4H/0DXI64HQBgslihVXlv+z6T2QrAttJtTeBo1ASOdjjvK3PeyfcwhBMRObFoTDQ+PT1W7jKGlfo2IyBakdx4zOF4mX6yTBV5v4lJIbhU3YKjSY/iaNKjaPBPczj/0ckyPDQjWabqhu5kcYPTc2M4Ck4ejCGciMiJ5PAAuUsYdg7l1yK69SL8zU0Ox4tCZ2LVpL5b7lHv4kP8MG9kJL7K7/18XYsRnV0Wr13M5vhV5yF8cnKohJUQDQ7nhBMR3QSzxSp3CT6pvLEDtQEj8NG4N/Bt/FrU+mei3i8FBl08YkO4SubN6m/J9tcPFUhUiWv1N589IlArUSVEg8eRcCKim/CPnEqsuCVe7jJ8SqvR9vClVaFGuX4yyvWT8XXKU1BZOqBRKaBWctzoZikGMJe+uaMLej+1BNW4zoVKg9wlEN00/kQjIupDfIhfr8cL69okrsT3HSuo7/W4WemHlZP4B89QKBUCUiP6nl719teFfZ73RF/kVkPX1YTZxa8iuuUC4AOdXmj4YAgnIurDqmtLfwuiGVGteZhY8SEW5/8aYe3eF1g83fnyZqfnwgM4rWColvTSY11rbsGEyk8Q0lEiQ0Wukd5wGNPKtmBNzjo8cmo5ZhW/BsD2YDWRJ+N0FCKiPigVAqKDdbjjqxWIaO+eN1sWPBlmyyKoOEXCJa63mXNGxTZzQ3bjdJ6YlvPIqtyGzPr9UFuNOBn3II6k/hsa2kwIC9D08V08R0tnFwAgo/6g/ViwsQphHcUAgFE+tBIo+Sb+9iAi6kdYgAa1AZkOx+JazuHL/FqZKvI9Z0ub+jzPXs+ucX10OLPuIMbU7oLaagQAjK7dDYW1C1uzi2SsbnCyC+qhMbciqemEw/HL4bchMzqQfyCTx+M7lIioXyIqgrIcjsQZzuG7PqZP0OB8c6UOAcZahLdd6TGvd8OcVJmq8j0jr40Ofxf9A4fjAV0NSGs8AgAoa/SOFWELaluhEM04E3c/GnWJAACLoEJh6K24a2yMzNUR9Y/TUYiI+pEaEYgTwbYQ3qhLQkXwBFsoF9mm0JXG1uzE7JLX0aKJQlHITFyKXISSkOkI0nlXxw5Pdn1KSqN/CsqDJyLecBYA0KqOQGTbZVwJX4CPT5Z5xYqwxi4roA7B1ylP4evkJxHeXoDItkswqTgKTt6BIZyIqB8ZUYHY7Z+ON6buRYcmzOFcq9GMQC1/lLpCSmM2ACDIVIPxNTtgVupQEjJd5qp8z6Ix0dh3oRr/zHwBCtGCVm0ULArHeeBmi9Wjg2yP7kSCgPqADNQHZCArse+e6ESewnP/CyMi8hBKhYCM6OAeARwAuvp5oJD6124yQ2s2ILblvMPxotCZmDcyUqaqfNe4awv3tOhi0eyX0COAA4DRg9/XVquI7WfKnZ6flhouYTVEN48hnIhoAGal9/6L/SIXCxmy4vp26Dsr0KLtbilnVmhRGjwZWQkh8hXmw+7sZ870nw9flaiSwTvTz0O8/GSKvAVDOBHRADibl3y8sEHiSnzPnu+qUBM4Cm9P3oGtt3yEQyk/xam4B2BR6qBkVxS3GBMX3O81/7PvkgSVDN7hS+xKRL6Bfy4SEQ0Aw6B7lDbc0IlDENDgn4oGf3ZD8RQWq+hR733xWuccf1MdBNGKNm2Uw3lv6XFOBHAknIhoQJwGEVGE2eK582c93Senypyem+lkChC5xm2jonocE0QLxlV9hillWwEAFU0dUpfVp4uVLQCA6aVv45FTP8Adl5+3tbW8ZvW1FW6JvAFHwomIBkHX1YTE5lOIM5xDbEsOWjWROJ62GbMzIuQuzedMT+35ICy5zsTEEHyZV2PfT2o8hrlFryCy/QrMghqXIm5HbWskEsP8ZayymyiK2JtbBV1XE8bWfA6laMbYml0YW7ML+9J/je9i7kEA54OTF+FIOBHRAD04IxlJTSewNP8/ManyfcS25iLRcBonrtbJXZpXslrFPs8LgudMg/BVyeG2gB1grMGKi08jst02qqwSu3Br8av4yoNWhS1rtI3KT6z82L7SJwBYBCWKQmfaFyIi8hYM4UREAxQZpEVF8ASHYzqzAWEdRTCaLTJV5b0u17Q6PTeQBwdp6K4vY9+mjcK5mH9xODeybh8i2i7jQoVndADadb4SAKAQzTArtPbj+RF3oVUbgynJoXKVRnRTGMKJiAahVRsDgyba4Vic4Rws/YzqUk9HLtcisekE5hb+D5KajkNpNdnPTUpioJLCjV1/jiU+gnZVCFrVEaj3S0VF0HjoO8uxN7cKXTI/92C1iugw2f7QzU7eiL9M/hzHEh5Bh0qPk/EPAgAiArV9fQsij8PJU0REg1QUOgvBxkpUBE1ARXAWqoLGIbTCgCkpnMM8GC2dZkyv+wLjq3dgcsXf0aXQ4VT8gzia9Di7XEjoiXnpeOOrAhjVevxl6j9gUfQMszvPVWDlJPkeerw+Cn5dhyYMR5OfwInEh2FRaBCoVUHhQV1ciAaCIdyFTCYTPvzwQ7z//vvIzc1FdXU1QkNDkZqaipUrV+KHP/whIiL48BaRtzuQ8esex45crmMIH4T6ViMgikhpPGo/prZ2olNpm9frSW3xfJ2fRokArRJtRkuvARywLagkpytOpi5dX+1zyYRYKcshcglOR3GRvLw8TJ8+HWvXrsU///lPlJSUwGg0oqqqCkePHsXPf/5zjB07Frt375a7VCIagr46drQZzRJW4t1OFTcivL0AQaYah+PFoTOxoJfWeeReSyfE9XvNt0XyLExV2dx/m8RYvU6CSohciyHcBcrKyrBw4UKcPXsWgO2J/nnz5uHhhx/GsmXL4OfnBwCoqanBihUrcPDgQRmrJaKhmNpHCG/pZAgfqOaOLhhVQTia+CgqA8dChIBmbSwa/FIwlg9lSm4gIfbry3WoNnRKUI2jD06U9nl+Vno4O+mQV+J0FBdYs2YNKioqAADJycnYsWMHsrKy7Ofr6upw33334cCBA+jq6sK9996LgoIChISEyFQxEd0stdL52MXhS7X4l6mJElbjvcoaOwBtNI4lPYZjSY9B19UEfWc5IAiciiIDQRDw8OxUvP1NYZ/X/f14CX6yMFOy/486u/rvOjSJXVHIS3EkfIh2796NI0eOAAA0Gg127tzpEMABICIiAjt27EBaWhoAoKGhAS+99JLktRKRa1zvrfx95R62uqCnqm0x9jjWqQ5BddBYAOwPLhe9v7rnQVFEev2XWH3+CSgttlHwbaedr3Lqah+dtI2Cj6v6DNNK38aEyo8xsnYv9J3dNfT1hzGRJ+M7d4heffVV+/a6deswfvz4Xq8LCAjACy+8YN9/8803YTbzo2sib9RjBUFRRHh7AZSWzn4XoCGgrNH5Q36Lx/MBOzlNvmFUObS9CCsvPIXleb9AouEUppS/CwAob5Tmj02LVUR9q61tZVbVx5hd8joWXn0Jiy89g8SmkwCAmenhktRC5A4M4UPQ2tqKAwcO2PfXr1/f5/WrVq1CYGAgANto+OHDh91aHxG5x/h4PQBgQuXHWHrx53j8xB1Ye+Y+xLecQ3GDvF0kvEGXxfkfKlz1UF7TbnjmYcHVl5DcdLz7XPlWBHfapl7euNy9u/zvgcsAgODOckS2XXE416EOQXSwDtPYkYi8GEP4EGRnZ8NotH2sGhAQgKlTp/Z5vU6nw8yZM+37fECTyDvp1EoAwOjafyKz4RD8zU0AgITm09h+plzGyryDWeaFX8g5nVqJhFBbM4GvUp+G9YaYoLIaMbZmJwDgbGnTgOZr36wbFwdKaD4F8YY6zAotyoMn4r6piewNTl6NIXwILl68aN8eP348VKr+n3OdNGlSr/cTkfcpD77FYT/BcFqmSryHKIo4WVAJiD2DOKeCe4Z7bokHANQFjEBOzCoAQKsmErtH/B5HEx+zX/f6oQK31fDGDd/7QvRybJ30ES5G3gURAs7FrEKnOoQBnLweu6MMQX5+vn07OTl5QPckJSXZt/Py8lxeExFJ4wcT43CucRKmlv/VfszfVA9BtMBqFRkQnNiaXYSsyk8wtXwrikNmoChkJopCZ6JTHYKHZgzs5yi5l+qGBx2zk59Ap1qPk/EPoUvZ84HkNqMZAVrXRol2kxnm7z1b0eSXhD0jfodvE36IdnUY54KTT+BI+BDU19fbt6Ojowd0T0xMjH27oUGehQ+IaOhSIwJQEZSF76KWY0/m89g8eQe2TP4UoqBEaR8PHg5noiiisb0LKY3Z8O9qxOjaf+Luy8/i1uJNAIDwwN5XayTpXW+1aVQF42jS470GcAD48+GrMJpdNy2l1WjGm19ddXq+3j8dHepQpEUEuOw1ieTCkfAhaG3tXkb3+oI8/bnxuhvv743RaLTPOQcAg8EwyAqJyF0EQYBJFYh9mf+vx7n6NhOSwxkSvs9iFaG2tCPecMbheFHoLD6Q6WHiQwb2Ow0AXvuyAI/PS4O/ZuiR4q3DzgP4jfgHG/kCjoQPQWdn98phGo1mQPdotd0/ODo6+m7z9OKLL0Kv19u/EhO5CAiRJ0mJ6H108GxJk7SFeIlTxY2IbTkPldhlP2YRlCjRT8MsTi/wOE/fnjnga/savR6oikH02eeCTuQLGMKHQKfrXubXZDIN6J4bR7b7Gz3/1a9+hebmZvtXaWnfS/cSkbSmJPfeHq25o6vX48NddkE9SkKmY/Pk7TiQ9gtcDb0VxSEzYVIFIsR/YAMZJB1BEPCDiXFOz/ub6hDW3h2+9+ZW3fRrdVms+PDb7t9xIR3FTq/dMCf1pl+HyJNwOsoQXO/5DfQ/qt3bdTfe3xutVuswck5EnuV6K7fedJgs8NMoJazGexh08ciJvRc5sfcCIhc38mSpvcy9VljNmFj5IWaUvoUmXQLez9oKUVDiQoUBM1LDe195sx+fn62wbyc3HsXKCz/BlbD5OJr0GOoCukfkNSoFgnSD//5Enogj4UMQHt798Wl1dfWA7qmq6h4pCAvjIgNE3qyv5dXfP1EiYSWer7HNyaeF7Evo0QRBwMSkEPt+UGclHjy7BvOKXobW0obotnyMr9puP//2N4V4ef8lhz7f/XnvWDFKri1ype8oxeJLzwAAMhoO4aGzazD/6n/br10zLanX70HkjRjCh2DkyJH27eJi5x+d3aikpPsX86hRo1xeExHJR2NuRWrD1xhXtZ1TUr7nbFmT03MTE0Mkq4MGb05GhH27TRMJ8Xt/OGVVfeSwL4rApoNXUNtihDNWq4jDl2rxP/suOVx3x5XfQWd2bELQ6NcdvEMDOG2JfAenowzB6NGj7dvnz5+H2Wzud8Ge06e7F/O48X4i8l4RbZdxx+UXENl2CQpYYVL4Izd6qdxleQxRFPt8WHX2DSGPPI9KqcCqSQnYdroMVoUK2UkbsTzv5wCARl0S9mS+AEG0QBQcp1+9d8w2OLVxfrp9lVlDZxdOXG3A+fLmXl9rb+ZzWH7xPxDZblum/nLYfJyLWQ0AeHxemlv+fURyYQgfglmzZkGr1cJoNKKtrQ0nT57EjBkznF5vNBpx7Ngx+/6CBQukKJOI3GjdrBR8dKgB0W3di29prO2Ias2H2TLKYeGT4arK0NnneY2K/xt5usSw7ucfikNmoFMZBJPSH9vGvYoWbUwfdw5uZU2DLh4fTtiMOy8/j5COUuwd8VtAsL0/XNECkciT8CffEAQGBmLhwoX2/S1btvR5/aeffoqWlhYAtvngc+fOdWd5RCSBsAANOjRhqPdLcTieYDiNXecr5SnKw3xwohThbVeQ2nAEKsvA29CR5xAEwd6y0KzU4XDqv+HTsZvsAVwQLZhT9L945NulWHbx5wgw1tz0a3Up/fGPkf+FT8a9bl8k6BF2RCEfxBA+RD/60Y/s21u2bEFubm6v17W3t+PZZ5+17z/22GP9Tl0hIu8QEahBefAkAIBZUKMs+Ba0qSNwtbZN5so8x/jqz7Di4s+w8fjtWJn7JDLrDgBguPImgiBg8fhYAEBu9A/Q6J9iP5fUdBxTyt9FsKkaGQ2H8K/nN/TZZhAAIIrQdvU+LQWCAp3qEAC2Di3B7IhCPoghfIiWLFmCOXPmALBNN1m6dClycnIcrqmvr8eKFStw5YptjltYWBh++ctfSl4rEbnH/JFROBt7Lz4a9wZem/ElPh7/Z+RF3S13WR6h+tpUlJTGowAAlWhCctNxhHTYHlJnuPIuzlY2bdVEYX/6r3A4+SeoDhgNvbES/3r+UUS15vW4NrizHDNK/oz1p++xd0Lpy4pb4odcN5En4lCsC/z973/HtGnTUFlZiaKiIkycOBHz5s1Deno6amtrsX//frS329ovqVQqfPTRRwgJCZG3aCJymcQwf9QHZPR6zmi2QKsavv3CswvqoO8oRWin42JjxaEzZaqIhuqROanYfKTQ4Vh9QAai2i5heuVmBJlsU1H8uxoxom4fagK7O4HFN5/B6u+egAK2FobBnZUIMNWhTdP7w7lrZya76V9BJD+OhLtAQkICDh48iIkTJwKwdQI4dOgQNm/ejM8//9wewCMjI7F9+3aHeeRE5NsuVBj6v8iHtRot0FpaURZ8C6yw/THSpg5DTcAIPHwrp6J4o2Cd2j4t5UZFoTPRpuleP+Ni5N34OvnHDtdUBY2FSdW9UJ0CVoys3dPr6ywYFYXwQC5YR76LI+EuMmrUKBw/fhwffPAB3n//feTm5qK6uhohISFIS0vDypUrsX79ekREsBUX0XByKL8WtySFyl2GbOpajEDgaHw8/s/QmFuR1HQCWksrICig9+NUFG81IjoQu887HutQh+KTsa9jed5/wKzQ4YuMZ+2dTa6zKDS4FH47JlR/aj8W15KD0+gpi/3jyccxhLuQRqPB2rVrsXbtWrlLISLyOCZVIK5EsDWrLxAEAU/MS8cbXzm2H+xSBWD7mJchiCKsit4jxsWouzGmZifyIu9CTsxqVAf2XDPj4dn8lIR8H6ejEBG5wN3jnfdKrumnT/ZwNCl5+H464Cv8NEr828LMHsctCi3MSp3T+yqCsvDW1N3Yl/ksqoPGAN9bgfPRuWnQ+/NTEvJ9DOFERC6QGXWta4RoRXh7ASZUfozF+b/GqJrdqOlj+W5f1tRucnpudnq403PkPRQKAY/OHeRKloJgbz/4fU8tyECglh/S0/DAdzoRkQsoFbbRvIUF/4UJ1Z85nNt9YTHGxevlKEtW758odXqOK4n6jkCtCk/fnglDpxlmixV/PdpPf3AnnlyQwfcFDSt8txMRuVB14BiH/YTm04AoylSNfPKqDOjssvR6LoRTDXyOIAjQ+6kRHqjF3BGRg7p3akoYnlyQATUDOA0zHAknInKRtMgAlHVMcjgW0FUPfWc5Wo3pw+pj9n+er8KYmn8greEICkNnoShkJtq0UQCAO8c6nz9P3m9ycigCtEr883xVn9cpFQLuuSUeiWH+ElVG5FmGz28EIiI3m5QUik9qEtHglwyTMgAGbRwM2hhYFCp8mVeDZVlxcpcoCYvVNvKfWXcAaY1fI7P+IADgRPw6fJPyJGKCnT+0R75hVEwwUsID8Pqhgh7nshL1mJUeAZ16+C5iRQQwhBMRuUximD8gCNg66ZMe567UtMpQkTx2na+E0mpCYvNJh+P1/um4Y2w0FArByZ3kS3RqJX66aITcZRB5LE7AIiJyoVl9dP3oq1uILymoaUW84QzU1u7WjCIEFIXOxKiYYBkrIyLyHAzhREQu1Fd/43e+KZKuEJmI1x5CrQkYib0Z/w+Xwm9HpzIQVYFj0KkOsXeRISIa7jgdhYjIhVLCA+QuQVZX69oAAJ3qEFyIXo4L0cshiGYEmOplroyIyLNwJJyIyIWG+8Nmhy/V9jgmCiq0aqNlqIaIyHMxhBMRuZjer+eUlABjDQTRjLpW3149s6m9y+m5f5maKGElRESejSGciMjF7ptmC5vxzWcwu2gTHjyzBo+dXILYllx8fLJM5urcR+xnUaL4ED+JKiEi8nycE05E5GL+GtuP1tuuvoTI9iv24ymN36AiOEuustzuch9tGFdPTpCwEiIiz8eRcCIiNykKneWwn9KYDQAwma1ylON2u3Iq4WdqgNbc0uMcV0UkInLEEE5E5CaF3wvhemMlNOZWfNXLw4u+YkrFe3ji+CLce/4xTC19B2HthXKXRETkkTgdhYjIDe4aF4MvcrJQHpSFcv0tKAydjcqgcRAFFb4rb8aiMb7ZLSSlMRsKWJBgOIMEwxmorJ04mrxR7rKIiDwOQzgRkRtkRAVij0KFjyb8Re5SJNHYZkKgsRoR7QUOx4tCZ2Hh6CiZqiIi8lycjkJE5AZq5fD58WqxitiSXYSItiswCxr78U5lEKqCxmJCQoh8xREReSiOhBMRucmPb8vAq19e6fXcd+XNGBevl7gi97hUbXsQsyhsNl6ffgAJzaeQ0pQNq6DCtHSOghMR9YYhnIjITTQq56Ph+y5U+0wIP1/WbN82K3UoCpuNorDZAIBHOQpORNSr4fN5KRERuZzFKqK8qcPpeaUgSFgNEZH3YAgnInKjKSmh3TuiiMjWfEwtfQeBxmo0tJnkK8xFLlQY+jzvp1FKVAkRkXfhdBQiIjeanByKk0WNmFP4CkbV7kFgVx0AoFOtx9bsaPx00QiZKxyaL/Nr5C6BiMgrcSSciMiNri9hH2iqtQdwAEht/EauklxGFEVYrCKmlm3BkrxfIaPuIJSWTvv5lZPiZayOiMizcSSciEgChaGzMKpur30/selbKK0mdHZZoFN755SNyuZOQBQxpnonwjpLMKJ+P0wKfxxI/yXyohYjiUvVExE5xZFwIiI3W5YVh+LQmRBhe0ixSZeA76JXQGXpwOmSRpmru3kffluKyLZLCOsssR/TWNvR5JcIABD4UCYRkVMcCScicrOMqEDsVIdib+ZzqAoci0a/ZOBaQD1R2IBZ6REyV3jzRtTtc9hv1saiKnAclk+Mk6kiIiLvwBBORCSRi1FLehwTRRkKcQGr1Vb46fgHYNDFYUTdPiQ0n8bliNsBQUBKeIDMFRIReTaGcCIime0+X4nF42PlLmNQ6lqNAIAOdSjOx6zE+ZiV8Dd1P3iqVHAqChFRXzgnnIhIAutmpTg9l1/VIl0hLvK34yU9jrVrItCuiUBaJEfBiYj6wxBORCSBsABNn+dFL5qXUtHHCpkAvG5Un4hIDgzhREQS2TAntedBUYQgWnC8sEH6gm7Sh9+W9nlereSvFiKi/nBOOBGRRK7Pk1ZYu5BgOI20hiNIbfwaxxIfxVFhCWakhctc4dBtnJ8udwlERF6BIZyISCKaayPEd11+DiNvaO2X1nAEF6OWQBRFj++t3dBmAkQrRtTtR1HoLJhUgQ7nvXXhISIiqfEzQyIiiaiuhfDikBkOx5ObjkFh7cIJL5iSsjW7CPGGs1hy6Td4/MSdWHrx5xhR+4X39lokIpIJQzgRkYQ2zk9HYegsh2NaSxviDOeQXVBv77/tiWoMnQC6F+hRiSZkNhzCtLJ3bL3BI7hMPRHRQHE6ChGRhHRqJdo1EagMHAcAuBp2KwpDb0VtwAgAQEFtKzKjg+Qs0am/HS+BIJqRWX/Q4filiEUAgDvHxshRFhGRV2IIJyKSWHpUID4e/yYsip5tC7+4UO2RIfz6CL2uy4DqwNFIbjoGpWgBAORfC+H+Gv5KISIaKP7EJCKS2LIJsXi5prXXcyaz1SMf0DxT2ggA6NCEYceYl6HtakZGw1eIas1Ds18i7p2SIHOFRETehXPCiYgk1l/A/uRUmUSVDNzhS3UO+0a1HrnRy/Fl+i8AAAmhnA9ORDQYDOFERDII8Vc7PVfW2AGLBz+gSUREQ8cQTkQkgzt6e4hRFKHvsK1GeeRyrcQVOffZmb5H5uePjJSoEiIi38E54UREMogP8bNvBxhrMKZ2F8ZW/wOBphq8OXUPzpQA80dGyVihTXN7F4rq2vu85pakUImqISLyHQzhREQy0phb8fCpFVCJXfZjI+oPIDd6uYxVdXvveLFtQxQBD3tYlIjIm3E6ChGRTGZnRMCkCuyxgubYmp0AgPyqFjnKcmAyWwEAi678Dityf4Ix1TuhNXfXtWCU/KP1RETeiCGciEgmExL0AIDc6GUOx/26GqG2tGP3+Uo5yupBaTUis/4gUpuO4s4rL+DxE3cgrf4rAMDYuGCZqyMi8k6cjkJEJBOdWgkAKAy9FU26BJTqJyM3ahkqgyZ4xNSPs6VNAICUxmxoLW324wrRgurAMQAAlZJjOUREN4MhnIhIRoFaFVqNwDuTtgFCz0D7XXkzxsXrJa+rsc2EL/NqAABhHcVo0UQhyGTbLwu+BW3aSKydmSx5XUREvoJDGEREMnpgRpJto5cADgD7LlSj1WiWsCKbLdlF9u0gYxXa1eEwaG1tFYtDZwIAwgO1ktdFROQrGMKJiGTkr1FhRlp4n9dIPTe8s8visF/nn4HotosINlahWRuHszH3YsUt8ZLWRETkazgdhYhIZjPSwnDsar3T8+WNHRJWA7x+qMBhPyd2Ner9UxHSWY4rYfPQpQpASjiXqSciGgqOhBMRyUzo4yFMpaVTwkqApnYTACC4sxwTKz6A2mJbqKdcPxm50cthVOsxPTWsz5qJiKh/DOFERB4osjUfi/N/jXVn/hUKaxe+K2+W5HXf+aYIADCp4n3cVvgnbPh2KW4t+j8EGGvs1/Q3fYaIiPrHEE5E5AEen5cGAFBZOnBP7lN48NyDGFm3D3pjBUbV7cW+C9XYm1vl1hqaO2yrdmrNBoyt/hwAoLO0YGr5XzG54m8AgJnp4VAoOApORDRUDOFERB7AX6OC3k8Ns9IPaqvjFJQpZX8FRCsuVBhQ2tDulte3WkW8/XUhAGB81WfQWLvnoVuhwJnY+wBwFJyIyFUYwomIPMSa6bZ2hd/Gr3M4HmSsRlhHEQDgk1Nlbnnt3AqDfbs8+BYUhM2FCNuI9+XwBWjRxbrldYmIhit2RyEi8hDdK2jORp1/Ovy7GnAm9j6ci10No6p7eXhRFF3+YOT+i9X27crgCfg8+E8I6SjGpIr3kRu1DABw/7Qkl74mEdFwxhBORORBNsxJxV+OFGLnqD+iRRMNi1LX45oDF2tw+5hol71mjaH3DixNfsk4mP6f9v3oYC7OQ0TkKgzhREQeJEinBmALwM6cL292WQh/++tC+wOZfRkXr2dbQiIiF+KccCIiDxOk63985H/2XcK50qYhvc7/Hbg8oACeGOaPRS4ceSciIoZwIiKPc/0Bzf4czKu56SButYowW0XbjihidtEmRLVe7PXae7hEPRGRyzGEExF5GH+NCsreenGLItLrv8S00s32QwfzagbdttBqFfHKgcv2/aTmE5hWvhUPnFuLpXm/QHi747L1vdZCRERDwhBOROSBHpzhOCc8vO0KVuX+CMvzfoGZJX9GRFt3iP7kVBmu1LQO6Pu2m8wOARyiiNnFr9l3M+u/xA8u/AyCaAEApEcFDuFfQUREzjCEExF5oLAAjX1bY27Fv57fgKTmkwAABayYf/VPgCjar9l5rgLHrtajy2Lt9fuJoogThQ1486urDsfTGo8gpvWCw7ETCeshCrZ2iUvGsz84EZE7MIQTEXmoH92WDgAwqQJxKu4Bh3OJhlNIaj7hcOxoQT02HbwCk7k7iFutIlo6u/Dy/sv45kpdj9co0U/FkeQn0XmtD3mjLhEXopYCAKakhHIqChGRm7BFIRGRh9KqlMhK1ONcaTNOxT+EcTWfI9hYBYMmGkdSfoIS/bRe73v1yysDfg2z0g8nE9YhJ2YVJlX8HXX+GbAqbL8aJiWFuuTfQUREPTGEExF5sLmZkThX2gyzUoevUn+KyLbL+DZ+Hcy9LOIzEBpzK0yqnvO8TapAHEt6zL5/9/gYBGj5K4KIyF04HYWIyIOplAosmWCbl30lfAGOJj1+0wE8rP0qfnh6NSZUbuv7ugANRsUE39RrEBHRwDCEExF5uBHRQQO6bkLlNiQ1HXd6fkn+rxHQVY/brr6E9PovnV5359iYQddIRESDwxBOROQF7h7fdzAOby/AvMI/YVXuk3jozH2Ibsl1vEAUEXGt/7cCVizO/w1SGr/p8X2WZcUiRn9zI+1ERDRwDOFERF5gVEww5mRG9HpOEM1YdPl3UIm2Jegj2gsQ1lHscI1SNOGLjGfs+yqxCy2ankvRZ0QNbNSdiIiGhiGciMhLTE7uvVvJlPJ3EdvqOPJdFnyLw75FoUVNwCj7fosmCvX+aQ7XPH17posqJSKi/jCEExF5CUEQ8IOJcT2On4p7EBVB4+37zdpYtOh6LrJjUvqjUxmIqsDR2D3yD4Bg+xUQpFPh4VtTIQjsCU5EJBVBFG9Yco08msFggF6vR3NzM4KD2bmAaLi6VN2CXTmVjgdFETGt3yGh+TSsggqn4x/o/ebvuWNsNMbG6d1QJRHR8DTQvMYmsEREXmZEdBDSFgTg9UMFMFuvjaMIAqqCxqPqhhHx/izLiuUccCIimXA6yhBZLBbk5ORg8+bN2LhxI6ZMmQKNRgNBECAIAubPny93iUTkg1RKBR6dm9b/hX1gACcikg9Hwodg+/bteOCBB9De3i53KUQ0DOnUSvz4toxBLVN/3frZKa4viIiIBowj4UPQ1NTEAE5EstKoFHj69swBh2p/jRLrZqUgxF/j3sKIiKhPHAl3gejoaEydOtX+tXfvXrzyyityl0VEw4QgCAjx1+DHt2XgVHEjmjtM0KqVOFvSZL9mSkooZqSFQ63k2AsRkSdgCB+Cu+66C8XFxUhKSnI4fvy482WjiYjcRaNSYGZ6uH3/tpFRMlZDRER9YQgfgpiYvpeRJiIiIiLqDT+XJCIiIiKSGEM4EREREZHEGMKJiIiIiCTGEE5EREREJDE+mOnBjEYjjEajfd9gMMhYDRERERG5CkfCPdiLL74IvV5v/0pMTJS7JCIiIiJyAYZwD/arX/0Kzc3N9q/S0lK5SyIiIiIiF/C56SibNm3Cpk2bXPo9f//732P16tUu/Z4DodVqodVqJX9dIiIiInIvnwvhdXV1yM/Pd+n3bGpqcun3IyIiIqLhjdNRiIiIiIgk5nMh/Pnnn4coii792rBhg9z/LCIiIiLyIT4XwomIiIiIPB1DOBERERGRxBjCiYiIiIgkxhBORERERCQxhnAiIiIiIon5XJ9wqS1evBgVFRUOx6qqquzbJ0+exMSJE3vct3v3bsTFxbm7PCIiIiLyQAzhQ3ThwgUUFxc7Pd/W1oZz5871OG4ymdxZFhERERF5ME5HISIiIiKSGEfCh6ioqEjuEoiIiIjIy3AknIiIiIhIYgzhREREREQSYwgnIiIiIpIY54R7EVEUAQAGg0HmSoiIiIioN9dz2vXc5gxDuBdpaWkBACQmJspcCRERERH1paWlBXq93ul5QewvppPHsFqtqKioQFBQEARBcPvrGQwGJCYmorS0FMHBwW5/PfJ+fM/QYPE9Q4PB9wsNlhzvGVEU0dLSgri4OCgUzmd+cyTciygUCiQkJEj+usHBwfxhR4PC9wwNFt8zNBh8v9BgSf2e6WsE/Do+mElEREREJDGGcCIiIiIiiTGEk1NarRbPPfcctFqt3KWQl+B7hgaL7xkaDL5faLA8+T3DBzOJiIiIiCTGkXAiIiIiIokxhBMRERERSYwhnIiIiIhIYgzhREREREQSYwinm1ZUVIS33noLDz74ILKyshAaGgq1Wo2wsDBMmDABjz/+OL766iu5yyQPYbFYkJOTg82bN2Pjxo2YMmUKNBoNBEGAIAiYP3++3CWSBEwmE959910sXrwYycnJ0Ol0iI2NxaxZs/Df//3fqKurk7tE8hD8mUGD5W25hN1RaNDOnDmDJ554AidOnBjQ9fPnz8fWrVuRlJTk5srIU23fvh0PPPAA2tvbnV4zb948HDp0SLqiSHJ5eXm4//77cfbsWafXREVF4Z133sHixYulK4w8Dn9m0GB4ay7hsvU0aPn5+T3e6CNGjMC4ceMQERGBpqYmZGdno6ysDABw6NAhzJw5E0eOHEFaWpocJZPMmpqa+vxlSr6vrKwMCxcuREVFBQBAEATMnTsX6enpqK2txf79+9HR0YGamhqsWLECe/bswYIFC2SumuTCnxk0GN6aSxjC6aZlZGRgw4YNePDBBxEfH+9wzmq1YsuWLXjqqafQ3t6OiooKPPDAA8jOzoYgCDJVTHKLjo7G1KlT7V979+7FK6+8IndZJIE1a9bYA3hycjJ27NiBrKws+/m6ujrcd999OHDgALq6unDvvfeioKAAISEhMlVMnoA/M2gwvC2XMITToMXGxuKdd97BQw89BKVS2es1CoUCDz/8MEJDQ7Fy5UoAwLFjx/DFF1/gzjvvlLJc8gB33XUXiouLe3z0d/z4cZkqIint3r0bR44cAQBoNBrs3LkT48ePd7gmIiICO3bswIQJE3D16lU0NDTgpZdewh/+8Ac5SiaZ8WcGDYa35hI+mEmDNm/ePPzwhz90+ka/0T333INp06bZ93ft2uXO0shDxcTEyD73juTz6quv2rfXrVvXI4BfFxAQgBdeeMG+/+abb8JsNru9PvI8/JlBg+GtuYQhnNxu9uzZ9u2ioiL5CiEiybW2tuLAgQP2/fXr1/d5/apVqxAYGAgAaGhowOHDh91aHxENP56SSxjCye1unGtlsVhkrISIpJadnQ2j0QjANtI9derUPq/X6XSYOXOmff/gwYNurY+Ihh9PySUM4eR258+ft28nJibKWAkRSe3ixYv27fHjx0Ol6v9RpEmTJvV6PxGRK3hKLmEIJ7cqKSlxGMm6/fbbZayGiKSWn59v305OTh7QPTfOBc7Ly3N5TUQ0fHlSLmEIJ7f62c9+Zv+oJykpCcuWLZO5IiKSUn19vX07Ojp6QPfExMTYtxsaGlxeExENX56USxjCyW22bt2Kbdu22fdffPFFaLVaGSsiIqm1trbat/38/AZ0z43X3Xg/EdFQeFouYQgntzh58iSeeOIJ+/7999+PNWvWyFgREcmhs7PTvq3RaAZ0z42/FDs6OlxeExENP56YS7hYjw/ZtGkTNm3a5NLv+fvf/x6rV68e1D2FhYVYtmyZ/ZfvhAkT8MYbb7i0Lho6T3m/kG/T6XT2bZPJNKB7rndTAQY+ek5E5Iyn5hKGcB9SV1fn8BCUKzQ1NQ3q+srKSixatAhVVVUAgLS0NOzZswfBwcEurYuGzhPeL+T7rvf8BgY+qn3jdTfeT0Q0WJ6cSzgdhVymvr4eixYtQkFBAQDbMrL79+9HbGyszJURkVzCw8Pt29XV1QO65/ovSwAICwtzeU1ENDx4ei5hCPchzz//PERRdOnXhg0bBvTaBoMBd955J3JzcwEAERER2L9/P1JTU935T6YhkPP9QsPHyJEj7dvFxcUDuqekpMS+PWrUKJfXRES+zxtyCUM4DVlbWxsWL16MU6dOAQD0ej327NmDMWPGyFwZEclt9OjR9u3z58/DbDb3e8/p06d7vZ+IaCC8JZcwhNOQdHZ2Yvny5fjmm28AAP7+/ti1axcmT54sc2VE5AlmzZpl73bS1taGkydP9nm90WjEsWPH7PsLFixwa31E5Fu8KZcwhNNN6+rqwqpVq+wrT2m1WuzYsQOzZ8+WuTIi8hSBgYFYuHChfX/Lli19Xv/pp5+ipaUFgG0++Ny5c91ZHhH5EG/LJQzhdFMsFgvWrFmD3bt3AwBUKhU++ugjLktPRD386Ec/sm9v2bLFPkfz+9rb2/Hss8/a9x977DGoVGziRUT988ZcwhBOgyaKIh555BF88sknAACFQoF3330Xy5cvl7kyIvJES5YswZw5cwDYppssXboUOTk5DtfU19djxYoVuHLlCgDbKPgvf/lLyWslIu/jrblEEEVRlLsI8i6vvfYafvzjH9v3MzMzcccddwz4flcvEEPeYfHixaioqHA4VlVVZW9bFxAQgIyMjB737d69G3FxcZLUSO5TVlaGadOmobKyEgAgCALmzZuH9PR01NbWYv/+/WhvbwdgG8Has2ePwzQWGn74M4MGyltzCUM4Ddrzzz+P3/72tzd9P99yw1NKSsqAW9TdqLCwECkpKa4viCSXl5eH+++/H2fPnnV6TWRkJN555x0sWbJEusLII/FnBg2Ut+YSTrYjIiJJjBo1CsePH8cHH3yA999/H7m5uaiurkZISAjS0tKwcuVKrF+/HhEREXKXSkTkdhwJJyIiIiKSGB/MJCIiIiKSGEM4EREREZHEGMKJiIiIiCTGEE5EREREJDGGcCIiIiIiiTGEExERERFJjCGciIiIiEhiDOFERERERBJjCCciIiIikhhDOBERERGRxBjCiYiIiIgkxhBORERERCQxhnAiIiIiIokxhBMRERERSYwhnIiIiIhIYgzhREREREQSYwgnIiIiIpIYQzgRERERkcQYwomIyK2uXr2K4OBgCIIAQRDwxz/+sd97nnzySfv1iYmJaGxslKBSIiLpCKIoinIXQUREvu3dd9/F2rVrAQBqtRpHjx7F5MmTe712165dWLp0KQBAoVDgwIEDmD9/vlSlEhFJgiPhRETkdg899BDuv/9+AEBXVxfWrFmDtra2HtdVVVVh/fr19v1f/OIXDOBE5JM4Ek5ERJJobm5GVlYWiouLAQAbNmzAW2+9ZT8viiLuvvtu7N27FwAwZcoUZGdnQ61Wy1IvEZE7cSSciIgkodfr8d5770GpVAIA/vKXv+Czzz6zn3/55ZftATwgIAB/+9vfGMCJyGdxJJyIiCT13HPP4YUXXgAAhIWFIScnB3V1dZg+fTqMRiMA4K233sKGDRvkLJOIyK0YwomISFIWiwVz585FdnY2AOC2225DdXU1Lly4AABYuXIltm3bJmeJRERuxxBORESSKywsxMSJE2EwGByOx8fHIycnB2FhYTJVRkQkDc4JJyIiyaWmpuK1115zOCYIAv76178ygBPRsMAQTkREsoiKinLYj42NxfTp02WqhohIWgzhREQkubq6Oqxbt87hWEVFBZ5++ml5CiIikhhDOBERSW7Dhg2orKwEAGRkZDi0Ldy+fbuMlRERSYMhnIiIJPXmm29ix44dAAA/Pz/s3LkTzzzzjP38hg0bUFFRIVd5RESSYHcUIiKSTH5+PiZNmoT29nYAwGuvvYaNGzfCYrFgzpw5OHr0KABg4cKF2LdvHwRBkLNcIiK34Ug4ERFJwmQyYc2aNfYAvmzZMmzcuBEAoFQq8d577yEoKAgAcODAAfzpT3+SrVYiIndjCCciIkk888wzOH36NAAgJiYGmzdvdjiflpaGTZs22fd/85vf4OzZs1KWSEQkGU5HISIitzt48CAWLVoEq9UKQRCwe/du3HXXXb1ee9999+HDDz8EAIwePRqnTp2Cn5+flOUSEbkdR8KJiMitGhoasHbtWlitVgDAU0895TSAA8Abb7yBpKQkAMDFixfx7//+75LUSUQkJY6EExGRW61evRrbtm0DAIwbNw7ffvstdDpdn/ccPnwYt912mz24f/7551i2bJnbayUikgpDOBERERGRxDgdhYiIiIhIYgzhREREREQSYwgnIiIiIpIYQzgRERERkcQYwomIiIiIJMYQTkREREQkMYZwIiIiIiKJMYQTEREREUmMIZyIiIiISGIM4UREREREEmMIJyIiIiKSGEM4EREREZHEGMKJiIiIiCTGEE5EREREJDGGcCIiIiIiiTGEExERERFJ7P8DRRxKVXGICDoAAAAASUVORK5CYII=", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#create a single plot with the original data and each neural network's predictions\n", + "fig,ax = plt.subplots(1,figsize = (8,8))\n", + "ax.plot(x,y,linewidth = 3.0,label = \"data\", alpha = 0.5)\n", + "ax.plot(x,y_predict_lt,linewidth = 3.0,linestyle=\"dotted\",label = \"linear-tree\")\n", + "plt.xlabel(\"x\")\n", + "plt.ylabel(\"y\")\n", + "plt.legend()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "d49ce38f", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "## Solving Optimization Problems with Linear Trees using OMLT\n", + "\n", + "\\begin{align*} \n", + "& \\min_x \\ \\hat{y}\\\\\n", + "& s.t. \\ \\hat{y} = ML(x) \n", + "\\end{align*}\n", + "\n", + "We instantiate a Pyomo `ConcreteModel` and create variables that represent the linear model decision tree input $x$ and output $\\hat y$. We also create an objective function that seeks to minimize the output $\\hat y$.\n", + "\n", + "The example uses the following general workflow:\n", + "- Create an OMLT `LinearTreeDefinition` object.\n", + "- Create a Pyomo model with variables `x` and `y` where we intend to minimize `y`.\n", + "- Create an `OmltBlock`.\n", + "- Create a formulation object. Note that `LinearTreeGDPFormulation` has an argument `transformation` that determines what Pyomo.GDP transformation is applied. Supported transformations are `bigm`, `hull`, and `mbigm`. If `custom` is applied, then the user must transform the Pyomo model or OmltBlock. Example shown below\n", + "- Build the formulation object on the `OmltBlock`.\n", + "- Add constraints connecting `x` to the linear model decision tree input and `y` to the linear tree output.\n", + "- Solve with an optimization solver (this example uses cbc for the MILPs and SCIP for any MIQCP/MIQPs).\n", + "- Query the solution.\n", + "\n", + "We also print model size and solution time following each cell where we optimize the Pyomo model. " + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "a78eb572", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "### Setup scaling and input bounds\n", + "We assume that our Pyomo model operates in the unscaled space with respect to our linear tree inputs and outputs. We additionally assume input bounds to our linear tree are given by the limits of our training data. \n", + "\n", + "To handle this, OMLT can be given scaling information (in the form of an OMLT scaling object) and input bounds (in the form of a dictionary where indices correspond to linear tree indices and values are 2-length tuples of lower and upper bounds). This maintains the space of the optimization problem and scaling is handled by OMLT underneath. The scaling object and input bounds are passed when creating an instance of the LinearTreeDefinition object." + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "5cf349b7", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Scaled input bounds: {0: (-1.7317910151019957, 1.7317910151019957)}\n" + ] + } + ], + "source": [ + "#create an omlt scaling object\n", + "scaler = omlt.scaling.OffsetScaling(offset_inputs=[mean_data['x']],\n", + " factor_inputs=[std_data['x']],\n", + " offset_outputs=[mean_data['y']],\n", + " factor_outputs=[std_data['y']])\n", + "\n", + "#create the input bounds. note that the key `0` corresponds to input `0` and that we also scale the input bounds\n", + "input_bounds={0:((min(df['x']) - mean_data['x'])/std_data['x'],\n", + " (max(df['x']) - mean_data['x'])/std_data['x'])};\n", + "print(scaler)\n", + "print(\"Scaled input bounds: \",input_bounds)" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "b4bd3410", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "## Linear Model Decision Tree with Big-M Transformation" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "id": "2d677fbf", + "metadata": { + "pycharm": { + "name": "#%%\n" + }, + "scrolled": false + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Welcome to the CBC MILP Solver \n", + "Version: 2.10.8 \n", + "Build Date: May 5 2022 \n", + "\n", + "command line - C:\\Users\\bammari\\cbc_solver\\bin\\cbc.exe -printingOptions all -import C:\\Users\\bammari\\AppData\\Local\\Temp\\tmprp8h3jhp.pyomo.lp -stat=1 -solve -solu C:\\Users\\bammari\\AppData\\Local\\Temp\\tmprp8h3jhp.pyomo.soln (default strategy 1)\n", + "Option for printingOptions changed from normal to all\n", + "Presolve 395 (-8) rows, 101 (-6) columns and 1085 (-13) elements\n", + "Statistics for presolved model\n", + "Original problem has 99 integers (99 of which binary)\n", + "Presolved problem has 99 integers (99 of which binary)\n", + "==== 100 zero objective 2 different\n", + "100 variables have objective of 0\n", + "1 variables have objective of 1\n", + "==== absolute objective values 2 different\n", + "100 variables have objective of 0\n", + "1 variables have objective of 1\n", + "==== for integers 99 zero objective 1 different\n", + "99 variables have objective of 0\n", + "==== for integers absolute objective values 1 different\n", + "99 variables have objective of 0\n", + "===== end objective counts\n", + "\n", + "\n", + "Problem has 395 rows, 101 columns (1 with objective) and 1085 elements\n", + "Column breakdown:\n", + "0 of type 0.0->inf, 0 of type 0.0->up, 0 of type lo->inf, \n", + "2 of type lo->up, 0 of type free, 0 of type fixed, \n", + "0 of type -inf->0.0, 0 of type -inf->up, 99 of type 0.0->1.0 \n", + "Row breakdown:\n", + "0 of type E 0.0, 1 of type E 1.0, 0 of type E -1.0, \n", + "0 of type E other, 0 of type G 0.0, 0 of type G 1.0, \n", + "197 of type G other, 0 of type L 0.0, 0 of type L 1.0, \n", + "197 of type L other, 0 of type Range 0.0->1.0, 0 of type Range other, \n", + "0 of type Free \n", + "Continuous objective value is -23.2758 - 0.00 seconds\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 296 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0005I 1 SOS with 99 members\n", + "Cgl0004I processed model has 394 rows, 101 columns (99 integer (99 of which binary)) and 3729 elements\n", + "Cbc0038I Initial state - 6 integers unsatisfied sum - 1\n", + "Cbc0038I Pass 1: suminf. 0.00000 (0) obj. 0.934131 iterations 29\n", + "Cbc0038I Solution found of 0.934131\n", + "Cbc0038I Relaxing continuous gives 0.845577\n", + "Cbc0038I Before mini branch and bound, 92 integers at bound fixed and 0 continuous\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 234 rows 9 columns - 0 fixed gives 234, 9 - still too large\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 229 rows 9 columns - too large\n", + "Cbc0038I Mini branch and bound did not improve solution (1.43 seconds)\n", + "Cbc0038I Round again with cutoff of -1.56656\n", + "Cbc0038I Pass 2: suminf. 0.11463 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 3: suminf. 0.11463 (2) obj. -1.56656 iterations 0\n", + "Cbc0038I Pass 4: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 5: suminf. 0.10349 (2) obj. -1.56656 iterations 5\n", + "Cbc0038I Pass 6: suminf. 0.10349 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 7: suminf. 0.07575 (2) obj. -1.56656 iterations 4\n", + "Cbc0038I Pass 8: suminf. 0.07575 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 9: suminf. 0.07575 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 10: suminf. 0.07575 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 11: suminf. 0.11463 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 12: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 13: suminf. 0.07660 (2) obj. -1.56656 iterations 6\n", + "Cbc0038I Pass 14: suminf. 0.07660 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 15: suminf. 0.08882 (2) obj. -1.56656 iterations 6\n", + "Cbc0038I Pass 16: suminf. 0.08882 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 17: suminf. 0.08882 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 18: suminf. 0.08882 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 19: suminf. 0.11463 (2) obj. -1.56656 iterations 4\n", + "Cbc0038I Pass 20: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 21: suminf. 0.15686 (2) obj. -1.56656 iterations 9\n", + "Cbc0038I Pass 22: suminf. 0.15686 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 23: suminf. 0.03798 (2) obj. -1.56656 iterations 4\n", + "Cbc0038I Pass 24: suminf. 0.03798 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 25: suminf. 0.03798 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 26: suminf. 0.03798 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 27: suminf. 0.11463 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 28: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 29: suminf. 0.03798 (2) obj. -1.56656 iterations 7\n", + "Cbc0038I Pass 30: suminf. 0.07575 (2) obj. -1.56656 iterations 9\n", + "Cbc0038I Pass 31: suminf. 0.07575 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I No solution found this major pass\n", + "Cbc0038I Before mini branch and bound, 82 integers at bound fixed and 0 continuous\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 254 rows 19 columns - 1 fixed gives 2, 2 - ok now\n", + "Cbc0038I Mini branch and bound did not improve solution (1.53 seconds)\n", + "Cbc0038I After 1.53 seconds - Feasibility pump exiting with objective of 0.845577 - took 0.11 seconds\n", + "Cbc0012I Integer solution of 0.84557742 found by feasibility pump after 0 iterations and 0 nodes (1.53 seconds)\n", + "Cbc0012I Integer solution of -0.18754905 found by DiveCoefficient after 0 iterations and 0 nodes (1.53 seconds)\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 232 rows 8 columns - 1 fixed gives 2, 2 - ok now\n", + "Cbc0031I 61 added rows had average density of 4.6721311\n", + "Cbc0013I At root node, 61 cuts changed objective from -23.275753 to -23.236823 in 54 passes\n", + "Cbc0014I Cut generator 0 (Probing) - 39 row cuts average 40.7 elements, 9 column cuts (9 active) in 0.037 seconds - new frequency is 1\n", + "Cbc0014I Cut generator 1 (Gomory) - 186 row cuts average 40.7 elements, 0 column cuts (0 active) in 0.008 seconds - new frequency is 1\n", + "Cbc0014I Cut generator 2 (Knapsack) - 0 row cuts average 0.0 elements, 0 column cuts (0 active) in 0.017 seconds - new frequency is -100\n", + "Cbc0014I Cut generator 3 (Clique) - 0 row cuts average 0.0 elements, 0 column cuts (0 active) in 0.004 seconds - new frequency is -100\n", + "Cbc0014I Cut generator 4 (MixedIntegerRounding2) - 159 row cuts average 6.9 elements, 0 column cuts (0 active) in 0.019 seconds - new frequency is 1\n", + "Cbc0014I Cut generator 5 (FlowCover) - 0 row cuts average 0.0 elements, 0 column cuts (0 active) in 0.016 seconds - new frequency is -100\n", + "Cbc0014I Cut generator 6 (TwoMirCuts) - 98 row cuts average 9.0 elements, 0 column cuts (0 active) in 0.013 seconds - new frequency is -100\n", + "Cbc0010I After 0 nodes, 1 on tree, -0.18754905 best solution, best possible -23.236823 (1.73 seconds)\n", + "Cbc0012I Integer solution of -0.79982231 found by DiveCoefficient after 1112 iterations and 7 nodes (1.80 seconds)\n", + "Cbc0004I Integer solution of -0.83246281 found after 1189 iterations and 7 nodes (1.80 seconds)\n", + "Cbc0012I Integer solution of -0.84944052 found by DiveCoefficient after 1528 iterations and 11 nodes (1.87 seconds)\n", + "Cbc0012I Integer solution of -0.8612633 found by DiveCoefficient after 1571 iterations and 11 nodes (1.88 seconds)\n", + "Cbc0001I Search completed - best objective -0.8612633046206106, took 1973 iterations and 18 nodes (1.93 seconds)\n", + "Cbc0032I Strong branching done 24 times (242 iterations), fathomed 3 nodes and fixed 0 variables\n", + "Cbc0035I Maximum depth 4, 0 variables fixed on reduced cost\n", + "Cuts at root node changed objective from -23.2758 to -23.2368\n", + "Probing was tried 194 times and created 1641 cuts of which 0 were active after adding rounds of cuts (0.074 seconds)\n", + "Gomory was tried 189 times and created 567 cuts of which 0 were active after adding rounds of cuts (0.022 seconds)\n", + "Knapsack was tried 54 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.017 seconds)\n", + "Clique was tried 54 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.004 seconds)\n", + "MixedIntegerRounding2 was tried 189 times and created 231 cuts of which 0 were active after adding rounds of cuts (0.063 seconds)\n", + "FlowCover was tried 54 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.016 seconds)\n", + "TwoMirCuts was tried 54 times and created 98 cuts of which 0 were active after adding rounds of cuts (0.013 seconds)\n", + "ZeroHalf was tried 1 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "2 bounds tightened after postprocessing\n", + "\n", + "\n", + "Result - Optimal solution found\n", + "\n", + "Objective value: -0.86126330\n", + "Enumerated nodes: 18\n", + "Total iterations: 1973\n", + "Time (CPU seconds): 1.96\n", + "Time (Wallclock seconds): 1.96\n", + "\n", + "Total time (CPU seconds): 1.97 (Wallclock seconds): 1.97\n", + "\n" + ] + } + ], + "source": [ + "#create a LinearTreeDefinition Object\n", + "ltmodel = LinearTreeDefinition(regr,scaler,input_bounds)\n", + "\n", + "#create a pyomo model with variables x and y\n", + "model1 = pyo.ConcreteModel()\n", + "model1.x = pyo.Var(initialize = 0)\n", + "model1.y = pyo.Var(initialize = 0)\n", + "model1.obj = pyo.Objective(expr=(model1.y))\n", + "\n", + "#create an OmltBlock\n", + "model1.lt = OmltBlock()\n", + "\n", + "#use the GDP formulation with a big-M, transformation\n", + "formulation1_lt = LinearTreeGDPFormulation(ltmodel, transformation='bigm')\n", + "model1.lt.build_formulation(formulation1_lt)\n", + "\n", + "#connect pyomo variables to the neural network\n", + "@model1.Constraint()\n", + "def connect_inputs(mdl):\n", + " return mdl.x == mdl.lt.inputs[0]\n", + "\n", + "@model1.Constraint()\n", + "def connect_outputs(mdl):\n", + " return mdl.y == mdl.lt.outputs[0]\n", + "\n", + "#solve the model and query the solution\n", + "status_1_bigm = pyo.SolverFactory('cbc').solve(model1, tee=True)\n", + "solution_1_bigm = (pyo.value(model1.x),pyo.value(model1.y))" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "id": "337dd04c", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Big-M Transformation Solution:\n", + "# of variables: 106\n", + "# of constraints: 402\n", + "x = -0.28571577\n", + "y = -0.8612633\n", + "Solve Time: 2.198737621307373\n" + ] + } + ], + "source": [ + "#print out model size and solution values\n", + "print(\"Big-M Transformation Solution:\")\n", + "print(\"# of variables: \",model1.nvariables())\n", + "print(\"# of constraints: \",model1.nconstraints())\n", + "print(\"x = \", solution_1_bigm[0])\n", + "print(\"y = \", solution_1_bigm[1])\n", + "print(\"Solve Time: \", status_1_bigm['Solver'][0]['Time'])" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "bb32ff8c", + "metadata": {}, + "source": [ + "## Linear Model Decision Tree with Convex Hull Transformation" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "c5a7aaff", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Welcome to the CBC MILP Solver \n", + "Version: 2.10.8 \n", + "Build Date: May 5 2022 \n", + "\n", + "command line - C:\\Users\\bammari\\cbc_solver\\bin\\cbc.exe -printingOptions all -import C:\\Users\\bammari\\AppData\\Local\\Temp\\tmp8stxvis7.pyomo.lp -stat=1 -solve -solu C:\\Users\\bammari\\AppData\\Local\\Temp\\tmp8stxvis7.pyomo.soln (default strategy 1)\n", + "Option for printingOptions changed from normal to all\n", + "Presolve 649 (-53) rows, 282 (-23) columns and 1689 (-106) elements\n", + "Statistics for presolved model\n", + "Original problem has 99 integers (99 of which binary)\n", + "Presolved problem has 99 integers (99 of which binary)\n", + "==== 281 zero objective 2 different\n", + "281 variables have objective of 0\n", + "1 variables have objective of 1\n", + "==== absolute objective values 2 different\n", + "281 variables have objective of 0\n", + "1 variables have objective of 1\n", + "==== for integers 99 zero objective 1 different\n", + "99 variables have objective of 0\n", + "==== for integers absolute objective values 1 different\n", + "99 variables have objective of 0\n", + "===== end objective counts\n", + "\n", + "\n", + "Problem has 649 rows, 282 columns (1 with objective) and 1689 elements\n", + "There are 1 singletons with objective \n", + "Column breakdown:\n", + "0 of type 0.0->inf, 0 of type 0.0->up, 0 of type lo->inf, \n", + "183 of type lo->up, 0 of type free, 0 of type fixed, \n", + "0 of type -inf->0.0, 0 of type -inf->up, 99 of type 0.0->1.0 \n", + "Row breakdown:\n", + "83 of type E 0.0, 1 of type E 1.0, 0 of type E -1.0, \n", + "1 of type E other, 0 of type G 0.0, 0 of type G 1.0, \n", + "0 of type G other, 563 of type L 0.0, 0 of type L 1.0, \n", + "0 of type L other, 0 of type Range 0.0->1.0, 1 of type Range other, \n", + "0 of type Free \n", + "Continuous objective value is -0.861263 - 0.00 seconds\n", + "Cgl0005I 1 SOS with 99 members\n", + "Cgl0004I processed model has 682 rows, 286 columns (99 integer (99 of which binary)) and 1755 elements\n", + "Cbc0038I Initial state - 0 integers unsatisfied sum - 0\n", + "Cbc0038I Solution found of -0.861263\n", + "Cbc0038I Relaxing continuous gives -0.861263\n", + "Cbc0038I Before mini branch and bound, 99 integers at bound fixed and 97 continuous\n", + "Cbc0038I Mini branch and bound did not improve solution (0.01 seconds)\n", + "Cbc0038I After 0.01 seconds - Feasibility pump exiting with objective of -0.861263 - took 0.00 seconds\n", + "Cbc0012I Integer solution of -0.8612633 found by feasibility pump after 0 iterations and 0 nodes (0.02 seconds)\n", + "Cbc0001I Search completed - best objective -0.8612633046206056, took 0 iterations and 0 nodes (0.02 seconds)\n", + "Cbc0035I Maximum depth 0, 0 variables fixed on reduced cost\n", + "Cuts at root node changed objective from -0.861263 to -0.861263\n", + "Probing was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "Gomory was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "Knapsack was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "Clique was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "MixedIntegerRounding2 was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "FlowCover was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "TwoMirCuts was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "ZeroHalf was tried 0 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "187 bounds tightened after postprocessing\n", + "\n", + "\n", + "Result - Optimal solution found\n", + "\n", + "Objective value: -0.86126330\n", + "Enumerated nodes: 0\n", + "Total iterations: 0\n", + "Time (CPU seconds): 0.02\n", + "Time (Wallclock seconds): 0.02\n", + "\n", + "Total time (CPU seconds): 0.04 (Wallclock seconds): 0.04\n", + "\n" + ] + } + ], + "source": [ + "#create a pyomo model with variables x and y\n", + "model2 = pyo.ConcreteModel()\n", + "model2.x = pyo.Var(initialize = 0)\n", + "model2.y = pyo.Var(initialize = 0)\n", + "model2.obj = pyo.Objective(expr=(model2.y))\n", + "\n", + "#create an OmltBlock\n", + "model2.lt = OmltBlock()\n", + "\n", + "#use the GDP formulation with a hull transformation\n", + "formulation2_lt = LinearTreeGDPFormulation(ltmodel, transformation='hull')\n", + "model2.lt.build_formulation(formulation2_lt)\n", + "\n", + "#connect pyomo variables to the neural network\n", + "@model2.Constraint()\n", + "def connect_inputs(mdl):\n", + " return mdl.x == mdl.lt.inputs[0]\n", + "\n", + "@model2.Constraint()\n", + "def connect_outputs(mdl):\n", + " return mdl.y == mdl.lt.outputs[0]\n", + "\n", + "#solve the model and query the solution\n", + "status_2_hull = pyo.SolverFactory('cbc').solve(model2, tee=True)\n", + "solution_2_hull = (pyo.value(model2.x),pyo.value(model2.y))" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "6a3e4be1", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Hull Transformation Solution:\n", + "# of variables: 304\n", + "# of constraints: 701\n", + "x = -0.28571577\n", + "y = -0.8612633\n", + "Solve Time: 0.07787513732910156\n" + ] + } + ], + "source": [ + "#print out model size and solution values\n", + "print(\"Hull Transformation Solution:\")\n", + "print(\"# of variables: \",model2.nvariables())\n", + "print(\"# of constraints: \",model2.nconstraints())\n", + "print(\"x = \", solution_2_hull[0])\n", + "print(\"y = \", solution_2_hull[1])\n", + "print(\"Solve Time: \", status_2_hull['Solver'][0]['Time'])" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "eea7c129", + "metadata": {}, + "source": [ + "## Linear Model Decision Tree with Custom Transformation\n", + "\n", + "By default, the transformation applied to the GDP formulation is the Big-M transformation. However if users would like to customize the transformation applied to the model, they can pass in `transformation='custom'` and the block added to the model will contain the untransformed disjuncts and disjunctions. This can be useful if user would like to pass in any of the transformation options (e.g. Big-M values, subsolvers to calculate M values etc...)\n", + "\n", + "NOTE: This will require the user to transform the model before passing it to the solver. See example below where we now call the bigm transformation outside of the model." + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "655a648b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Welcome to the CBC MILP Solver \n", + "Version: 2.10.8 \n", + "Build Date: May 5 2022 \n", + "\n", + "command line - C:\\Users\\bammari\\cbc_solver\\bin\\cbc.exe -printingOptions all -import C:\\Users\\bammari\\AppData\\Local\\Temp\\tmppzksju8x.pyomo.lp -stat=1 -solve -solu C:\\Users\\bammari\\AppData\\Local\\Temp\\tmppzksju8x.pyomo.soln (default strategy 1)\n", + "Option for printingOptions changed from normal to all\n", + "Presolve 395 (-8) rows, 101 (-6) columns and 1085 (-13) elements\n", + "Statistics for presolved model\n", + "Original problem has 99 integers (99 of which binary)\n", + "Presolved problem has 99 integers (99 of which binary)\n", + "==== 100 zero objective 2 different\n", + "100 variables have objective of 0\n", + "1 variables have objective of 1\n", + "==== absolute objective values 2 different\n", + "100 variables have objective of 0\n", + "1 variables have objective of 1\n", + "==== for integers 99 zero objective 1 different\n", + "99 variables have objective of 0\n", + "==== for integers absolute objective values 1 different\n", + "99 variables have objective of 0\n", + "===== end objective counts\n", + "\n", + "\n", + "Problem has 395 rows, 101 columns (1 with objective) and 1085 elements\n", + "Column breakdown:\n", + "0 of type 0.0->inf, 0 of type 0.0->up, 0 of type lo->inf, \n", + "2 of type lo->up, 0 of type free, 0 of type fixed, \n", + "0 of type -inf->0.0, 0 of type -inf->up, 99 of type 0.0->1.0 \n", + "Row breakdown:\n", + "0 of type E 0.0, 1 of type E 1.0, 0 of type E -1.0, \n", + "0 of type E other, 0 of type G 0.0, 0 of type G 1.0, \n", + "197 of type G other, 0 of type L 0.0, 0 of type L 1.0, \n", + "197 of type L other, 0 of type Range 0.0->1.0, 0 of type Range other, \n", + "0 of type Free \n", + "Continuous objective value is -23.2758 - 0.00 seconds\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 296 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 295 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0003I 0 fixed, 0 tightened bounds, 294 strengthened rows, 0 substitutions\n", + "Cgl0005I 1 SOS with 99 members\n", + "Cgl0004I processed model has 394 rows, 101 columns (99 integer (99 of which binary)) and 3729 elements\n", + "Cbc0038I Initial state - 6 integers unsatisfied sum - 1\n", + "Cbc0038I Pass 1: suminf. 0.00000 (0) obj. 0.934131 iterations 29\n", + "Cbc0038I Solution found of 0.934131\n", + "Cbc0038I Relaxing continuous gives 0.845577\n", + "Cbc0038I Before mini branch and bound, 92 integers at bound fixed and 0 continuous\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 234 rows 9 columns - 0 fixed gives 234, 9 - still too large\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 229 rows 9 columns - too large\n", + "Cbc0038I Mini branch and bound did not improve solution (1.23 seconds)\n", + "Cbc0038I Round again with cutoff of -1.56656\n", + "Cbc0038I Pass 2: suminf. 0.11463 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 3: suminf. 0.11463 (2) obj. -1.56656 iterations 0\n", + "Cbc0038I Pass 4: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 5: suminf. 0.10349 (2) obj. -1.56656 iterations 5\n", + "Cbc0038I Pass 6: suminf. 0.10349 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 7: suminf. 0.07575 (2) obj. -1.56656 iterations 4\n", + "Cbc0038I Pass 8: suminf. 0.07575 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 9: suminf. 0.07575 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 10: suminf. 0.07575 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 11: suminf. 0.11463 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 12: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 13: suminf. 0.07660 (2) obj. -1.56656 iterations 6\n", + "Cbc0038I Pass 14: suminf. 0.07660 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 15: suminf. 0.08882 (2) obj. -1.56656 iterations 6\n", + "Cbc0038I Pass 16: suminf. 0.08882 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 17: suminf. 0.08882 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 18: suminf. 0.08882 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 19: suminf. 0.11463 (2) obj. -1.56656 iterations 4\n", + "Cbc0038I Pass 20: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 21: suminf. 0.15686 (2) obj. -1.56656 iterations 9\n", + "Cbc0038I Pass 22: suminf. 0.15686 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 23: suminf. 0.03798 (2) obj. -1.56656 iterations 4\n", + "Cbc0038I Pass 24: suminf. 0.03798 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 25: suminf. 0.03798 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 26: suminf. 0.03798 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I Pass 27: suminf. 0.11463 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 28: suminf. 0.16849 (2) obj. -1.56656 iterations 3\n", + "Cbc0038I Pass 29: suminf. 0.03798 (2) obj. -1.56656 iterations 7\n", + "Cbc0038I Pass 30: suminf. 0.07575 (2) obj. -1.56656 iterations 9\n", + "Cbc0038I Pass 31: suminf. 0.07575 (2) obj. -1.56656 iterations 2\n", + "Cbc0038I No solution found this major pass\n", + "Cbc0038I Before mini branch and bound, 82 integers at bound fixed and 0 continuous\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 254 rows 19 columns - 1 fixed gives 2, 2 - ok now\n", + "Cbc0038I Mini branch and bound did not improve solution (1.33 seconds)\n", + "Cbc0038I After 1.33 seconds - Feasibility pump exiting with objective of 0.845577 - took 0.11 seconds\n", + "Cbc0012I Integer solution of 0.84557742 found by feasibility pump after 0 iterations and 0 nodes (1.33 seconds)\n", + "Cbc0012I Integer solution of -0.18754905 found by DiveCoefficient after 0 iterations and 0 nodes (1.34 seconds)\n", + "Cbc0038I Full problem 394 rows 101 columns, reduced to 232 rows 8 columns - 1 fixed gives 2, 2 - ok now\n", + "Cbc0031I 61 added rows had average density of 4.6721311\n", + "Cbc0013I At root node, 61 cuts changed objective from -23.275753 to -23.236823 in 54 passes\n", + "Cbc0014I Cut generator 0 (Probing) - 39 row cuts average 40.7 elements, 9 column cuts (9 active) in 0.036 seconds - new frequency is 1\n", + "Cbc0014I Cut generator 1 (Gomory) - 186 row cuts average 40.7 elements, 0 column cuts (0 active) in 0.005 seconds - new frequency is 1\n", + "Cbc0014I Cut generator 2 (Knapsack) - 0 row cuts average 0.0 elements, 0 column cuts (0 active) in 0.013 seconds - new frequency is -100\n", + "Cbc0014I Cut generator 3 (Clique) - 0 row cuts average 0.0 elements, 0 column cuts (0 active) in 0.003 seconds - new frequency is -100\n", + "Cbc0014I Cut generator 4 (MixedIntegerRounding2) - 159 row cuts average 6.9 elements, 0 column cuts (0 active) in 0.018 seconds - new frequency is 1\n", + "Cbc0014I Cut generator 5 (FlowCover) - 0 row cuts average 0.0 elements, 0 column cuts (0 active) in 0.020 seconds - new frequency is -100\n", + "Cbc0014I Cut generator 6 (TwoMirCuts) - 98 row cuts average 9.0 elements, 0 column cuts (0 active) in 0.015 seconds - new frequency is -100\n", + "Cbc0010I After 0 nodes, 1 on tree, -0.18754905 best solution, best possible -23.236823 (1.51 seconds)\n", + "Cbc0012I Integer solution of -0.79982231 found by DiveCoefficient after 1112 iterations and 7 nodes (1.57 seconds)\n", + "Cbc0004I Integer solution of -0.83246281 found after 1189 iterations and 7 nodes (1.58 seconds)\n", + "Cbc0012I Integer solution of -0.84944052 found by DiveCoefficient after 1528 iterations and 11 nodes (1.64 seconds)\n", + "Cbc0012I Integer solution of -0.8612633 found by DiveCoefficient after 1571 iterations and 11 nodes (1.65 seconds)\n", + "Cbc0001I Search completed - best objective -0.8612633046206106, took 1973 iterations and 18 nodes (1.70 seconds)\n", + "Cbc0032I Strong branching done 24 times (242 iterations), fathomed 3 nodes and fixed 0 variables\n", + "Cbc0035I Maximum depth 4, 0 variables fixed on reduced cost\n", + "Cuts at root node changed objective from -23.2758 to -23.2368\n", + "Probing was tried 194 times and created 1641 cuts of which 0 were active after adding rounds of cuts (0.076 seconds)\n", + "Gomory was tried 189 times and created 567 cuts of which 0 were active after adding rounds of cuts (0.011 seconds)\n", + "Knapsack was tried 54 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.013 seconds)\n", + "Clique was tried 54 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.003 seconds)\n", + "MixedIntegerRounding2 was tried 189 times and created 231 cuts of which 0 were active after adding rounds of cuts (0.047 seconds)\n", + "FlowCover was tried 54 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.020 seconds)\n", + "TwoMirCuts was tried 54 times and created 98 cuts of which 0 were active after adding rounds of cuts (0.015 seconds)\n", + "ZeroHalf was tried 1 times and created 0 cuts of which 0 were active after adding rounds of cuts (0.000 seconds)\n", + "2 bounds tightened after postprocessing\n", + "\n", + "\n", + "Result - Optimal solution found\n", + "\n", + "Objective value: -0.86126330\n", + "Enumerated nodes: 18\n", + "Total iterations: 1973\n", + "Time (CPU seconds): 1.76\n", + "Time (Wallclock seconds): 1.76\n", + "\n", + "Total time (CPU seconds): 1.77 (Wallclock seconds): 1.77\n", + "\n" + ] + } + ], + "source": [ + "#create a pyomo model with variables x and y\n", + "model_c = pyo.ConcreteModel()\n", + "model_c.x = pyo.Var(initialize = 0)\n", + "model_c.y = pyo.Var(initialize = 0)\n", + "model_c.obj = pyo.Objective(expr=(model_c.y))\n", + "\n", + "#create an OmltBlock\n", + "model_c.lt = OmltBlock()\n", + "\n", + "#use the GDP formulation with a custom transformation\n", + "formulation_c_lt = LinearTreeGDPFormulation(ltmodel, transformation='custom')\n", + "model_c.lt.build_formulation(formulation_c_lt)\n", + "\n", + "#connect pyomo variables to the neural network\n", + "@model_c.Constraint()\n", + "def connect_inputs(mdl):\n", + " return mdl.x == mdl.lt.inputs[0]\n", + "\n", + "@model_c.Constraint()\n", + "def connect_outputs(mdl):\n", + " return mdl.y == mdl.lt.outputs[0]\n", + "\n", + "# NOTE: Since we passed the 'custom' transformation option, the user must\n", + "# transform the model or the omlt block before passing the model to the solver\n", + "pyo.TransformationFactory('gdp.bigm').apply_to(model_c)\n", + "\n", + "#solve the model and query the solution\n", + "status_c_bigm = pyo.SolverFactory('cbc').solve(model_c, tee=True)\n", + "solution_c_bigm = (pyo.value(model_c.x),pyo.value(model_c.y))" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "id": "d0c9899b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "BigM Transformation Solution:\n", + "# of variables: 106\n", + "# of constraints: 402\n", + "x = -0.28571577\n", + "y = -0.8612633\n", + "Solve Time: 1.994880199432373\n" + ] + } + ], + "source": [ + "#print out model size and solution values\n", + "print(\"BigM Transformation Solution:\")\n", + "print(\"# of variables: \",model_c.nvariables())\n", + "print(\"# of constraints: \",model_c.nconstraints())\n", + "print(\"x = \", solution_c_bigm[0])\n", + "print(\"y = \", solution_c_bigm[1])\n", + "print(\"Solve Time: \", status_c_bigm['Solver'][0]['Time'])" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "a0c7843c", + "metadata": {}, + "source": [ + "## Linear Model Decision Tree with Hybrid Big-M Representation\n", + "\n", + "#### NOTE: This representation requires a solver that can handle MIQCPs (e.g., scip)" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "d56a9671", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "SCIP version 8.0.3 [precision: 8 byte] [memory: block] [mode: optimized] [LP solver: Soplex 6.0.3] [GitHash: 62fab8a2e3]\n", + "Copyright (C) 2002-2022 Konrad-Zuse-Zentrum fuer Informationstechnik Berlin (ZIB)\n", + "\n", + "External libraries: \n", + " Soplex 6.0.3 Linear Programming Solver developed at Zuse Institute Berlin (soplex.zib.de) [GitHash: f900e3d0]\n", + " CppAD 20180000.0 Algorithmic Differentiation of C++ algorithms developed by B. Bell (github.com/coin-or/CppAD)\n", + " ZLIB 1.2.13 General purpose compression library by J. Gailly and M. Adler (zlib.net)\n", + " AMPL/MP 4e2d45c4 AMPL .nl file reader library (github.com/ampl/mp)\n", + " PaPILO 2.1.2 parallel presolve for integer and linear optimization (github.com/scipopt/papilo) [GitHash: 2fe2543]\n", + " bliss 0.77 Computing Graph Automorphism Groups by T. Junttila and P. Kaski (www.tcs.hut.fi/Software/bliss/)\n", + " Ipopt 3.14.12 Interior Point Optimizer developed by A. Waechter et.al. (github.com/coin-or/Ipopt)\n", + "\n", + "user parameter file not found - using default parameters\n", + "read problem \n", + "============\n", + "\n", + "original problem has 106 variables (0 bin, 99 int, 0 impl, 7 cont) and 9 constraints\n", + "\n", + "solve problem\n", + "=============\n", + "\n", + " [linear] : [C] (+0) -1.37862124[C] (+0) == 1.38375294812141;\n", + ";\n", + "violation: left hand side is violated by 1.38375294812141\n", + "all 1 solutions given by solution candidate storage are infeasible\n", + "\n", + "presolving:\n", + "(round 1, fast) 5 del vars, 5 del conss, 0 add conss, 8 chg bounds, 0 chg sides, 0 chg coeffs, 0 upgd conss, 0 impls, 1 clqs\n", + "(round 2, exhaustive) 5 del vars, 5 del conss, 0 add conss, 8 chg bounds, 0 chg sides, 0 chg coeffs, 1 upgd conss, 0 impls, 1 clqs\n", + " (0.0s) probing cycle finished: starting next cycle\n", + " (0.0s) symmetry computation started: requiring (bin +, int -, cont -), (fixed: bin -, int +, cont +)\n", + " (0.0s) no symmetry present\n", + "presolving (3 rounds: 3 fast, 2 medium, 2 exhaustive):\n", + " 5 deleted vars, 5 deleted constraints, 0 added constraints, 8 tightened bounds, 0 added holes, 0 changed sides, 0 changed coefficients\n", + " 371 implications, 361 cliques\n", + "presolved problem has 101 variables (99 bin, 0 int, 0 impl, 2 cont) and 4 constraints\n", + " 1 constraints of type \n", + " 2 constraints of type \n", + " 1 constraints of type \n", + "Presolving Time: 0.00\n", + "\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + "t 0.0s| 1 | 0 | 357 | - | trysol| 0 | 201 | 4 | 202 | 0 | 0 | 0 | 0 |-2.327575e+01 | 4.575920e+00 | Inf | unknown\n", + " 0.0s| 1 | 0 | 379 | - | 3407k | 0 | 201 | 4 | 202 | 0 | 0 | 0 | 0 |-2.327575e+01 | 4.575920e+00 | Inf | unknown\n", + "L 0.0s| 1 | 0 | 379 | - | subnlp| 0 | 201 | 4 | 202 | 0 | 0 | 0 | 0 |-2.327575e+01 | 1.330887e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 608 | - | 3508k | 0 | 201 | 4 | 293 | 91 | 1 | 0 | 0 |-2.327575e+01 | 1.330887e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 608 | - | 3512k | 0 | 201 | 4 | 293 | 91 | 1 | 0 | 0 |-2.327575e+01 | 1.330887e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 626 | - | 3512k | 0 | 201 | 4 | 293 | 91 | 1 | 0 | 0 |-2.327575e+01 | 1.330887e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 632 | - | 3595k | 0 | 201 | 4 | 294 | 94 | 2 | 0 | 0 |-2.327575e+01 | 1.330887e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 651 | - | 3645k | 0 | 201 | 4 | 293 | 97 | 3 | 0 | 0 |-2.140031e+01 | 1.330887e-01 | Inf | unknown\n", + "t 0.0s| 1 | 0 | 651 | - | trysol| 0 | 201 | 4 | 293 | 97 | 3 | 0 | 0 |-2.140031e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 651 | - | 3652k | 0 | 201 | 4 | 293 | 97 | 3 | 0 | 0 |-2.140031e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 651 | - | 3652k | 0 | 201 | 4 | 293 | 97 | 3 | 0 | 0 |-2.140031e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 656 | - | 3770k | 0 | 201 | 4 | 292 | 100 | 4 | 0 | 0 |-1.570543e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 656 | - | 3776k | 0 | 201 | 4 | 292 | 100 | 4 | 0 | 0 |-1.570543e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 663 | - | 3804k | 0 | 201 | 4 | 290 | 102 | 5 | 0 | 0 |-1.244243e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 663 | - | 3808k | 0 | 201 | 4 | 290 | 102 | 5 | 0 | 0 |-1.244243e+01 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 0.0s| 1 | 0 | 668 | - | 3837k | 0 | 201 | 4 | 292 | 104 | 6 | 0 | 0 |-1.196055e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 671 | - | 3839k | 0 | 201 | 4 | 294 | 106 | 7 | 0 | 0 |-1.180980e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 671 | - | 3839k | 0 | 201 | 4 | 294 | 106 | 7 | 0 | 0 |-1.180980e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 678 | - | 3843k | 0 | 201 | 4 | 296 | 108 | 8 | 0 | 0 |-1.071253e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 678 | - | 3847k | 0 | 201 | 4 | 296 | 108 | 8 | 0 | 0 |-1.071253e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 679 | - | 3879k | 0 | 201 | 4 | 297 | 109 | 9 | 0 | 0 |-1.039798e+01 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 688 | - | 4043k | 0 | 201 | 4 | 299 | 111 | 10 | 0 | 0 |-9.245489e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 688 | - | 4048k | 0 | 201 | 4 | 299 | 111 | 10 | 0 | 0 |-9.245489e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 697 | - | 4100k | 0 | 201 | 4 | 302 | 114 | 11 | 0 | 0 |-8.270904e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 697 | - | 4104k | 0 | 201 | 4 | 302 | 114 | 11 | 0 | 0 |-8.270904e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 705 | - | 4133k | 0 | 201 | 4 | 305 | 117 | 12 | 0 | 0 |-7.452430e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 705 | - | 4138k | 0 | 201 | 4 | 305 | 117 | 12 | 0 | 0 |-7.452430e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 718 | - | 4168k | 0 | 201 | 4 | 165 | 120 | 13 | 0 | 0 |-6.057443e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 718 | - | 4169k | 0 | 201 | 4 | 165 | 120 | 13 | 0 | 0 |-6.057443e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 723 | - | 4185k | 0 | 201 | 4 | 166 | 121 | 14 | 0 | 0 |-5.722086e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 0.0s| 1 | 0 | 723 | - | 4186k | 0 | 201 | 4 | 166 | 121 | 14 | 0 | 0 |-5.722086e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 726 | - | 4203k | 0 | 201 | 4 | 168 | 123 | 15 | 0 | 0 |-5.301365e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 726 | - | 4204k | 0 | 201 | 4 | 168 | 123 | 15 | 0 | 0 |-5.301365e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 728 | - | 4220k | 0 | 201 | 4 | 169 | 124 | 16 | 0 | 0 |-4.984856e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 728 | - | 4222k | 0 | 201 | 4 | 169 | 124 | 16 | 0 | 0 |-4.984856e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 733 | - | 4238k | 0 | 201 | 4 | 172 | 127 | 17 | 0 | 0 |-4.654982e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 733 | - | 4238k | 0 | 201 | 4 | 172 | 127 | 17 | 0 | 0 |-4.654982e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 740 | - | 4262k | 0 | 201 | 4 | 173 | 128 | 18 | 0 | 0 |-4.547473e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 747 | - | 4288k | 0 | 201 | 4 | 153 | 129 | 19 | 0 | 0 |-4.499495e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 762 | - | 4303k | 0 | 201 | 4 | 155 | 131 | 20 | 0 | 0 |-4.204416e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 762 | - | 4304k | 0 | 201 | 4 | 155 | 131 | 20 | 0 | 0 |-4.204416e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 766 | - | 4321k | 0 | 201 | 4 | 156 | 133 | 21 | 0 | 0 |-3.978313e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 766 | - | 4321k | 0 | 201 | 4 | 156 | 133 | 21 | 0 | 0 |-3.978313e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 769 | - | 4321k | 0 | 201 | 4 | 157 | 135 | 22 | 0 | 0 |-3.845750e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 771 | - | 4322k | 0 | 201 | 4 | 158 | 136 | 23 | 0 | 0 |-3.797458e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 0.0s| 1 | 0 | 774 | - | 4324k | 0 | 201 | 4 | 159 | 137 | 24 | 0 | 0 |-3.681545e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 774 | - | 4325k | 0 | 201 | 4 | 159 | 137 | 24 | 0 | 0 |-3.681545e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 775 | - | 4325k | 0 | 201 | 4 | 136 | 138 | 25 | 0 | 0 |-3.670501e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 776 | - | 4325k | 0 | 201 | 4 | 137 | 139 | 26 | 0 | 0 |-3.633925e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 779 | - | 4325k | 0 | 201 | 4 | 139 | 141 | 27 | 0 | 0 |-3.608036e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 781 | - | 4326k | 0 | 201 | 4 | 140 | 142 | 28 | 0 | 0 |-3.596732e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 785 | - | 4326k | 0 | 201 | 4 | 141 | 143 | 29 | 0 | 0 |-3.580600e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 787 | - | 4328k | 0 | 201 | 4 | 142 | 145 | 30 | 0 | 0 |-3.577469e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 790 | - | 4329k | 0 | 201 | 4 | 127 | 147 | 31 | 0 | 0 |-3.550749e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 796 | - | 4329k | 0 | 201 | 4 | 128 | 149 | 32 | 0 | 0 |-3.495249e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 804 | - | 4329k | 0 | 201 | 4 | 130 | 152 | 33 | 0 | 0 |-3.420224e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 804 | - | 4330k | 0 | 201 | 4 | 130 | 152 | 33 | 0 | 0 |-3.420224e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 815 | - | 4330k | 0 | 201 | 4 | 131 | 154 | 34 | 0 | 0 |-3.379897e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 823 | - | 4331k | 0 | 201 | 4 | 132 | 155 | 35 | 0 | 0 |-3.370760e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 835 | - | 4336k | 0 | 201 | 4 | 139 | 162 | 36 | 0 | 0 |-3.200630e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 0.0s| 1 | 0 | 835 | - | 4336k | 0 | 201 | 4 | 139 | 162 | 36 | 0 | 0 |-3.200630e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 837 | - | 4401k | 0 | 201 | 4 | 136 | 163 | 37 | 0 | 0 |-3.186074e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 840 | - | 4401k | 0 | 201 | 4 | 137 | 164 | 38 | 0 | 0 |-3.176592e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 848 | - | 4401k | 0 | 201 | 4 | 138 | 165 | 39 | 0 | 0 |-3.148079e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 853 | - | 4417k | 0 | 201 | 4 | 139 | 166 | 40 | 0 | 0 |-3.144644e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 864 | - | 4417k | 0 | 201 | 4 | 140 | 167 | 41 | 0 | 0 |-3.137616e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 869 | - | 4425k | 0 | 201 | 4 | 141 | 169 | 42 | 0 | 0 |-3.132915e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 885 | - | 4425k | 0 | 201 | 4 | 142 | 175 | 43 | 0 | 0 |-3.050905e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 894 | - | 4499k | 0 | 201 | 4 | 145 | 179 | 44 | 0 | 0 |-3.029288e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 894 | - | 4500k | 0 | 201 | 4 | 145 | 179 | 44 | 0 | 0 |-3.029288e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 903 | - | 4500k | 0 | 201 | 4 | 146 | 180 | 45 | 0 | 0 |-2.980014e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 909 | - | 4745k | 0 | 201 | 4 | 150 | 184 | 46 | 0 | 0 |-2.945498e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 913 | - | 4745k | 0 | 201 | 4 | 151 | 186 | 47 | 0 | 0 |-2.933423e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 919 | - | 4746k | 0 | 201 | 4 | 152 | 188 | 48 | 0 | 0 |-2.908037e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 927 | - | 4767k | 0 | 201 | 4 | 143 | 192 | 49 | 0 | 0 |-2.880193e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 0.0s| 1 | 0 | 929 | - | 4767k | 0 | 201 | 4 | 144 | 193 | 50 | 0 | 0 |-2.877294e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 933 | - | 4768k | 0 | 201 | 4 | 146 | 195 | 51 | 0 | 0 |-2.867571e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 933 | - | 4768k | 0 | 201 | 4 | 146 | 195 | 51 | 0 | 0 |-2.867571e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 940 | - | 4768k | 0 | 201 | 4 | 147 | 196 | 52 | 0 | 0 |-2.854023e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 946 | - | 4769k | 0 | 201 | 4 | 148 | 197 | 53 | 0 | 0 |-2.850047e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 948 | - | 4771k | 0 | 201 | 4 | 149 | 198 | 54 | 0 | 0 |-2.847677e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 951 | - | 4772k | 0 | 201 | 4 | 139 | 199 | 55 | 0 | 0 |-2.838502e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 962 | - | 4772k | 0 | 201 | 4 | 145 | 205 | 56 | 0 | 0 |-2.792219e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 973 | - | 4903k | 0 | 201 | 4 | 152 | 212 | 57 | 0 | 0 |-2.736641e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 974 | - | 4903k | 0 | 201 | 4 | 153 | 213 | 58 | 0 | 0 |-2.731977e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 980 | - | 4921k | 0 | 201 | 4 | 154 | 214 | 60 | 0 | 0 |-2.727619e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 983 | - | 4923k | 0 | 201 | 4 | 156 | 216 | 61 | 0 | 0 |-2.724984e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 985 | - | 4923k | 0 | 201 | 4 | 143 | 217 | 62 | 0 | 0 |-2.723786e+00 | 1.015021e-01 | Inf | unknown\n", + " 0.0s| 1 | 0 | 988 | - | 4923k | 0 | 201 | 4 | 147 | 221 | 63 | 0 | 0 |-2.722242e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 991 | - | 4923k | 0 | 201 | 4 | 149 | 223 | 64 | 0 | 0 |-2.720862e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 994 | - | 4923k | 0 | 201 | 4 | 151 | 225 | 65 | 0 | 0 |-2.720253e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 995 | - | 4923k | 0 | 201 | 4 | 152 | 226 | 66 | 0 | 0 |-2.719956e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 997 | - | 4923k | 0 | 201 | 4 | 153 | 227 | 67 | 0 | 0 |-2.718858e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1001 | - | 4923k | 0 | 201 | 4 | 134 | 228 | 68 | 0 | 0 |-2.718758e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1006 | - | 4923k | 0 | 201 | 4 | 137 | 231 | 69 | 0 | 0 |-2.716826e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1006 | - | 4923k | 0 | 201 | 4 | 137 | 231 | 69 | 0 | 0 |-2.716826e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1019 | - | 4923k | 0 | 201 | 4 | 143 | 237 | 70 | 0 | 0 |-2.665895e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1025 | - | 4924k | 0 | 201 | 4 | 146 | 240 | 71 | 0 | 0 |-2.644971e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1027 | - | 4924k | 0 | 201 | 4 | 147 | 241 | 72 | 0 | 0 |-2.633351e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1029 | - | 4924k | 0 | 201 | 4 | 150 | 244 | 73 | 0 | 0 |-2.628571e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1035 | - | 4924k | 0 | 201 | 4 | 147 | 249 | 74 | 0 | 0 |-2.606847e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1037 | - | 4924k | 0 | 201 | 4 | 148 | 250 | 75 | 0 | 0 |-2.604073e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1040 | - | 4924k | 0 | 201 | 4 | 151 | 253 | 76 | 0 | 0 |-2.601238e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1043 | - | 4924k | 0 | 201 | 4 | 154 | 256 | 77 | 0 | 0 |-2.596541e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1044 | - | 4924k | 0 | 201 | 4 | 155 | 257 | 78 | 0 | 0 |-2.596030e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 1046 | - | 4924k | 0 | 201 | 4 | 157 | 259 | 79 | 0 | 0 |-2.593466e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1049 | - | 4924k | 0 | 201 | 4 | 142 | 261 | 80 | 0 | 0 |-2.591362e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1056 | - | 4924k | 0 | 201 | 4 | 145 | 264 | 81 | 0 | 0 |-2.588632e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1057 | - | 4924k | 0 | 201 | 4 | 146 | 265 | 82 | 0 | 0 |-2.588519e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1059 | - | 4924k | 0 | 201 | 4 | 147 | 266 | 83 | 0 | 0 |-2.587954e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1060 | - | 4924k | 0 | 201 | 4 | 148 | 267 | 84 | 0 | 0 |-2.587692e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1062 | - | 4924k | 0 | 201 | 4 | 150 | 269 | 85 | 0 | 0 |-2.586700e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1065 | - | 4924k | 0 | 201 | 4 | 135 | 272 | 86 | 0 | 0 |-2.585920e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1067 | - | 4924k | 0 | 201 | 4 | 136 | 273 | 87 | 0 | 0 |-2.585759e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1069 | - | 4925k | 0 | 201 | 4 | 137 | 274 | 88 | 0 | 0 |-2.585528e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1072 | - | 4925k | 0 | 201 | 4 | 138 | 275 | 89 | 0 | 0 |-2.584502e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1080 | - | 4925k | 0 | 201 | 4 | 142 | 279 | 90 | 0 | 0 |-2.583814e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1082 | - | 4925k | 0 | 201 | 4 | 143 | 280 | 91 | 0 | 0 |-2.583244e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1087 | - | 4925k | 0 | 201 | 4 | 134 | 282 | 92 | 0 | 0 |-2.583196e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1091 | - | 4925k | 0 | 201 | 4 | 136 | 284 | 94 | 0 | 0 |-2.583080e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 1094 | - | 4925k | 0 | 201 | 4 | 137 | 285 | 95 | 0 | 0 |-2.583022e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1096 | - | 4925k | 0 | 201 | 4 | 138 | 286 | 96 | 0 | 0 |-2.582959e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1101 | - | 4932k | 0 | 201 | 4 | 139 | 287 | 97 | 0 | 0 |-2.582855e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1106 | - | 4932k | 0 | 201 | 4 | 140 | 288 | 98 | 0 | 0 |-2.582814e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1174 | - | 4944k | 0 | 201 | 4 | 140 | 288 |100 | 0 | 0 |-2.582814e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1215 | - | 4945k | 0 | 201 | 4 | 155 | 303 |101 | 0 | 0 |-2.406285e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1215 | - | 4945k | 0 | 201 | 4 | 155 | 303 |101 | 0 | 0 |-2.406285e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1228 | - | 4945k | 0 | 201 | 4 | 158 | 306 |102 | 0 | 0 |-2.202288e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1228 | - | 4945k | 0 | 201 | 4 | 158 | 306 |102 | 0 | 0 |-2.202288e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1238 | - | 4945k | 0 | 201 | 4 | 161 | 309 |103 | 0 | 0 |-2.056316e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1238 | - | 4946k | 0 | 201 | 4 | 161 | 309 |103 | 0 | 0 |-2.056316e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1245 | - | 4946k | 0 | 201 | 4 | 163 | 311 |104 | 0 | 0 |-2.021493e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1247 | - | 4948k | 0 | 201 | 4 | 150 | 312 |105 | 0 | 0 |-2.012366e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1252 | - | 4948k | 0 | 201 | 4 | 151 | 313 |106 | 0 | 0 |-1.996384e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1257 | - | 4948k | 0 | 201 | 4 | 152 | 314 |107 | 0 | 0 |-1.955347e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 1265 | - | 4948k | 0 | 201 | 4 | 153 | 315 |108 | 0 | 0 |-1.921560e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1265 | - | 4948k | 0 | 201 | 4 | 153 | 315 |108 | 0 | 0 |-1.921560e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1268 | - | 4948k | 0 | 201 | 4 | 154 | 316 |109 | 0 | 0 |-1.912654e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1276 | - | 4948k | 0 | 201 | 4 | 155 | 317 |110 | 0 | 0 |-1.907029e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1287 | - | 4948k | 0 | 201 | 4 | 130 | 318 |111 | 0 | 0 |-1.865064e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1291 | - | 5079k | 0 | 201 | 4 | 131 | 319 |112 | 0 | 0 |-1.823179e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1296 | - | 5079k | 0 | 201 | 4 | 132 | 320 |113 | 0 | 0 |-1.802190e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1296 | - | 5079k | 0 | 201 | 4 | 132 | 320 |113 | 0 | 0 |-1.802190e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1300 | - | 5079k | 0 | 201 | 4 | 133 | 321 |114 | 0 | 0 |-1.788799e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1308 | - | 5080k | 0 | 201 | 4 | 134 | 322 |115 | 0 | 0 |-1.770434e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1314 | - | 5081k | 0 | 201 | 4 | 135 | 323 |116 | 0 | 0 |-1.767061e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1320 | - | 5083k | 0 | 201 | 4 | 120 | 325 |117 | 0 | 0 |-1.763275e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1324 | - | 5083k | 0 | 201 | 4 | 122 | 328 |118 | 0 | 0 |-1.756420e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1331 | - | 5084k | 0 | 201 | 4 | 124 | 330 |119 | 0 | 0 |-1.748752e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1333 | - | 5085k | 0 | 201 | 4 | 125 | 331 |120 | 0 | 0 |-1.735226e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 1340 | - | 5086k | 0 | 201 | 4 | 126 | 332 |121 | 0 | 0 |-1.705447e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1343 | - | 5087k | 0 | 201 | 4 | 127 | 333 |122 | 0 | 0 |-1.703066e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1343 | - | 5088k | 0 | 201 | 4 | 127 | 333 |122 | 0 | 0 |-1.703066e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1350 | - | 5088k | 0 | 201 | 4 | 125 | 334 |123 | 0 | 0 |-1.692534e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1354 | - | 5088k | 0 | 201 | 4 | 126 | 335 |124 | 0 | 0 |-1.689660e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1362 | - | 5088k | 0 | 201 | 4 | 127 | 336 |125 | 0 | 0 |-1.672401e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1372 | - | 5088k | 0 | 201 | 4 | 128 | 337 |126 | 0 | 0 |-1.654144e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1377 | - | 5088k | 0 | 201 | 4 | 129 | 338 |127 | 0 | 0 |-1.649238e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1381 | - | 5090k | 0 | 201 | 4 | 130 | 339 |128 | 0 | 0 |-1.645776e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1387 | - | 5090k | 0 | 201 | 4 | 126 | 340 |129 | 0 | 0 |-1.634856e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1390 | - | 5090k | 0 | 201 | 4 | 127 | 341 |130 | 0 | 0 |-1.628467e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1393 | - | 5090k | 0 | 201 | 4 | 128 | 342 |131 | 0 | 0 |-1.625388e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1394 | - | 5090k | 0 | 201 | 4 | 129 | 343 |132 | 0 | 0 |-1.619549e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1395 | - | 5090k | 0 | 201 | 4 | 130 | 344 |133 | 0 | 0 |-1.618533e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1397 | - | 5090k | 0 | 201 | 4 | 131 | 345 |134 | 0 | 0 |-1.614928e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 1399 | - | 5090k | 0 | 201 | 4 | 124 | 346 |135 | 0 | 0 |-1.613378e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1401 | - | 5090k | 0 | 201 | 4 | 125 | 347 |136 | 0 | 0 |-1.612070e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1402 | - | 5090k | 0 | 201 | 4 | 126 | 348 |137 | 0 | 0 |-1.610826e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1402 | - | 5090k | 0 | 201 | 4 | 126 | 348 |137 | 0 | 0 |-1.610826e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1411 | - | 5090k | 0 | 201 | 4 | 127 | 349 |138 | 0 | 0 |-1.597546e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1425 | - | 5090k | 0 | 201 | 4 | 128 | 350 |139 | 0 | 0 |-1.572121e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1431 | - | 5090k | 0 | 201 | 4 | 131 | 353 |140 | 0 | 0 |-1.568082e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1439 | - | 5090k | 0 | 201 | 4 | 123 | 354 |141 | 0 | 0 |-1.555021e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1444 | - | 5090k | 0 | 201 | 4 | 124 | 355 |142 | 0 | 0 |-1.545798e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1450 | - | 5090k | 0 | 201 | 4 | 125 | 356 |143 | 0 | 0 |-1.543402e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1453 | - | 5090k | 0 | 201 | 4 | 126 | 357 |144 | 0 | 0 |-1.541965e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1456 | - | 5090k | 0 | 201 | 4 | 127 | 358 |145 | 0 | 0 |-1.540646e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1457 | - | 5090k | 0 | 201 | 4 | 128 | 359 |146 | 0 | 0 |-1.538945e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1462 | - | 5090k | 0 | 201 | 4 | 126 | 360 |147 | 0 | 0 |-1.535542e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1464 | - | 5090k | 0 | 201 | 4 | 127 | 361 |148 | 0 | 0 |-1.534705e+00 | 1.015021e-01 | Inf | unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 1465 | - | 5090k | 0 | 201 | 4 | 128 | 362 |149 | 0 | 0 |-1.534559e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1466 | - | 5090k | 0 | 201 | 4 | 129 | 363 |150 | 0 | 0 |-1.533710e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1468 | - | 5090k | 0 | 201 | 4 | 130 | 364 |151 | 0 | 0 |-1.533614e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1470 | - | 5090k | 0 | 201 | 4 | 131 | 365 |152 | 0 | 0 |-1.533365e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1473 | - | 5090k | 0 | 201 | 4 | 121 | 366 |153 | 0 | 0 |-1.530984e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1480 | - | 5090k | 0 | 201 | 4 | 122 | 367 |154 | 0 | 0 |-1.529665e+00 | 1.015021e-01 | Inf | unknown\n", + " 1.0s| 1 | 0 | 1481 | - | 5090k | 0 | 201 | 4 | 123 | 368 |155 | 0 | 0 |-1.529237e+00 | 1.015021e-01 | Inf | unknown\n", + "t 1.0s| 1 | 0 | 1481 | - | trysol| 0 | 201 | 4 | 123 | 368 |155 | 0 | 0 |-1.529237e+00 |-2.230405e-01 | 585.63%| unknown\n", + " 1.0s| 1 | 0 | 1484 | - | 5091k | 0 | 201 | 4 | 124 | 369 |156 | 0 | 0 |-1.528082e+00 |-2.230405e-01 | 585.11%| unknown\n", + " 1.0s| 1 | 0 | 1484 | - | 5091k | 0 | 201 | 4 | 124 | 369 |156 | 0 | 0 |-1.528082e+00 |-2.230405e-01 | 585.11%| unknown\n", + " 1.0s| 1 | 0 | 1498 | - | 5091k | 0 | 201 | 4 | 123 | 369 |156 | 0 | 0 |-1.515148e+00 |-2.230405e-01 | 579.32%| unknown\n", + " 1.0s| 1 | 0 | 1502 | - | 5091k | 0 | 201 | 4 | 123 | 371 |157 | 0 | 0 |-1.497708e+00 |-2.230405e-01 | 571.50%| unknown\n", + " 1.0s| 1 | 0 | 1509 | - | 5091k | 0 | 201 | 4 | 119 | 373 |158 | 0 | 0 |-1.483028e+00 |-2.230405e-01 | 564.91%| unknown\n", + " 1.0s| 1 | 0 | 1513 | - | 5091k | 0 | 201 | 4 | 120 | 375 |159 | 0 | 0 |-1.475901e+00 |-2.230405e-01 | 561.72%| unknown\n", + "t 1.0s| 1 | 0 | 1513 | - | trysol| 0 | 201 | 4 | 120 | 375 |159 | 0 | 0 |-1.475901e+00 |-8.487397e-01 | 73.89%| unknown\n", + " time | node | left |LP iter|LP it/n|mem/heur|mdpt |vars |cons |rows |cuts |sepa|confs|strbr| dualbound | primalbound | gap | compl. \n", + " 1.0s| 1 | 0 | 1517 | - | 5092k | 0 | 201 | 4 | 121 | 378 |160 | 0 | 0 |-1.453804e+00 |-8.487397e-01 | 71.29%| unknown\n", + " 1.0s| 1 | 0 | 1517 | - | 5093k | 0 | 201 | 4 | 121 | 378 |160 | 0 | 0 |-1.453804e+00 |-8.487397e-01 | 71.29%| unknown\n", + "t 1.0s| 1 | 0 | 1525 | - | trysol| 0 | 201 | 4 | 109 | 378 |160 | 0 | 0 |-1.196708e+00 |-8.560812e-01 | 39.79%| unknown\n", + " 1.0s| 1 | 0 | 1525 | - | 5094k | 0 | 201 | 4 | 109 | 378 |160 | 0 | 0 |-1.196708e+00 |-8.560812e-01 | 39.79%| unknown\n", + " 1.0s| 1 | 0 | 1525 | - | 5094k | 0 | 201 | 4 | 102 | 378 |160 | 0 | 0 |-1.196708e+00 |-8.560812e-01 | 39.79%| unknown\n", + " 1.0s| 1 | 0 | 1528 | - | 5094k | 0 | 201 | 4 | 102 | 378 |160 | 0 | 0 |-8.612633e-01 |-8.560812e-01 | 0.61%| unknown\n", + " 1.0s| 1 | 0 | 1528 | - | 5094k | 0 | 201 | 4 | 100 | 378 |160 | 0 | 0 |-8.612633e-01 |-8.560812e-01 | 0.61%| unknown\n", + " 1.0s| 1 | 0 | 1530 | - | 5094k | 0 | 201 | 4 | 101 | 381 |161 | 0 | 0 |-8.612633e-01 |-8.560812e-01 | 0.61%| unknown\n", + " 1.0s| 1 | 0 | 1530 | - | 5094k | 0 | 201 | 1 | 101 | 381 |163 | 0 | 0 |-8.612633e-01 |-8.560812e-01 | 0.61%| unknown\n", + "r 1.0s| 1 | 0 | 1530 | - |randroun| 0 | 201 | 1 | 101 | 381 |165 | 0 | 0 |-8.612633e-01 |-8.612633e-01 | 0.00%| unknown\n", + "* 1.0s| 1 | 0 | 1530 | - | LP | 0 | 201 | 1 | 101 | 381 |165 | 0 | 0 |-8.612633e-01 |-8.612633e-01 | 0.00%| unknown\n", + "\n", + "SCIP Status : problem is solved [optimal solution found]\n", + "Solving Time (sec) : 1.00\n", + "Solving Nodes : 1\n", + "Primal Bound : -8.61263306337283e-01 (9 solutions)\n", + "Dual Bound : -8.61263306337283e-01\n", + "Gap : 0.00 %\n" + ] + } + ], + "source": [ + "#create a pyomo model with variables x and y\n", + "model3 = pyo.ConcreteModel()\n", + "model3.x = pyo.Var(initialize = 0)\n", + "model3.y = pyo.Var(initialize = 0)\n", + "model3.obj = pyo.Objective(expr=(model3.y))\n", + "\n", + "#create an OmltBlock\n", + "model3.lt = OmltBlock()\n", + "\n", + "#use the Hybrid Big-M formulation\n", + "formulation3_lt = LinearTreeHybridBigMFormulation(ltmodel)\n", + "model3.lt.build_formulation(formulation3_lt)\n", + "\n", + "#connect pyomo variables to the neural network\n", + "@model3.Constraint()\n", + "def connect_inputs(mdl):\n", + " return mdl.x == mdl.lt.inputs[0]\n", + "\n", + "@model3.Constraint()\n", + "def connect_outputs(mdl):\n", + " return mdl.y == mdl.lt.outputs[0]\n", + "\n", + "#solve the model and query the solution\n", + "status_3_hyb = pyo.SolverFactory('scip').solve(model3, tee=True)\n", + "solution_3_hyb = (pyo.value(model3.x),pyo.value(model3.y))" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "35e58b1d", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Hull Transformation Solution:\n", + "# of variables: 106\n", + "# of constraints: 9\n", + "x = -0.28571576298269263\n", + "y = -0.8612633063372832\n", + "Solve Time: 1.0\n" + ] + } + ], + "source": [ + "#print out model size and solution values\n", + "print(\"Hull Transformation Solution:\")\n", + "print(\"# of variables: \",model3.nvariables())\n", + "print(\"# of constraints: \",model3.nconstraints())\n", + "print(\"x = \", solution_3_hyb[0])\n", + "print(\"y = \", solution_3_hyb[1])\n", + "print(\"Solve Time: \", status_3_hyb['Solver'][0]['Time'])" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "308f6dd2", + "metadata": { + "pycharm": { + "name": "#%% md\n" + } + }, + "source": [ + "### Final Plots and Discussion" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "id": "3f23c904", + "metadata": { + "pycharm": { + "name": "#%%\n" + }, + "scrolled": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 35, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#create a plot with 3 subplots\n", + "fig,axs = plt.subplots(1,3,figsize = (24,8))\n", + "\n", + "#GDP Representation - Big-M Transformation\n", + "axs[0].plot(x,y_predict_lt,linewidth = 3.0,linestyle=\"dotted\",color = \"orange\", label='Fitted Model')\n", + "axs[0].set_title(\"Big-M\")\n", + "axs[0].scatter([solution_1_bigm[0]],[solution_1_bigm[1]],color = \"black\",s = 300, label='Optimum')\n", + "axs[0].legend()\n", + "\n", + "#GDP Representation - Hull Transformation\n", + "axs[1].plot(x,y_predict_lt,linewidth = 3.0,linestyle=\"dotted\",color = \"orange\", label='Fitted Model')\n", + "axs[1].set_title(\"Convex Hull\")\n", + "axs[1].scatter([solution_2_hull[0]],[solution_2_hull[1]],color = \"black\",s = 300, label='Optimum')\n", + "axs[1].legend()\n", + "\n", + "\n", + "#Hybrid Big-M Representation\n", + "axs[2].plot(x,y_predict_lt,linewidth = 3.0,linestyle=\"dotted\",color = \"orange\", label='Fitted Model')\n", + "axs[2].set_title(\"Hybrid Big-M\")\n", + "axs[2].scatter([solution_3_hyb[0]],[solution_3_hyb[1]],color = \"black\",s = 300, label='Optimum')\n", + "axs[2].legend()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "7f6fe8cb", + "metadata": { + "pycharm": { + "name": "#%%\n" + } + }, + "source": [ + "## References\n", + "[1] \n", + "B.L. Ammari, E.S. Johnson, G. Stinchfield, T. Kim, M. Bynum, W.E. Hart, J. Pulsipher, C.D. Laird (2023). \n", + "Linear model decision trees as surrogates in optimization of engineering applications. Computers & Chemical Engineering, 108347 " + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.16" + }, + "vscode": { + "interpreter": { + "hash": "b9796abd98c586628167f9644fb316c5d68ef1536b751187190e2dd23541a2e4" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/setup.cfg b/setup.cfg index ff8eb614..5e008d57 100644 --- a/setup.cfg +++ b/setup.cfg @@ -76,6 +76,7 @@ testing = ipywidgets jupyter lightgbm + linear-tree matplotlib pandas keras diff --git a/src/omlt/dependencies.py b/src/omlt/dependencies.py index 98042fba..a5fc41ba 100644 --- a/src/omlt/dependencies.py +++ b/src/omlt/dependencies.py @@ -3,3 +3,5 @@ # check for dependencies and create shortcut if available onnx, onnx_available = attempt_import("onnx") keras, keras_available = attempt_import("tensorflow.keras") + +lineartree, lineartree_available = attempt_import("lineartree") diff --git a/src/omlt/linear_tree/__init__.py b/src/omlt/linear_tree/__init__.py new file mode 100644 index 00000000..b08e2684 --- /dev/null +++ b/src/omlt/linear_tree/__init__.py @@ -0,0 +1,24 @@ +r""" +There are multiple formulations for representing linear model decision trees. + +Please see the following reference: + * Ammari et al. (2023) Linear Model Decision Trees as Surrogates in Optimization + of Engineering Applications. Computers & Chemical Engineering + +We utilize the following common nomenclature in the formulations: + +.. math:: + \begin{align*} + L &:= \text{Set of leaves} \\ + z_{\ell} &:= \text{Binary variable indicating which leaf is selected} \\ + x &:= \text{Vector of input variables to the decision tree} \\ + d &:= \text{Output variable from the decision tree} \\ + a_{\ell} &:= \text{Vector of slopes learned by the tree for leaf } \ell \in L\\ + b_{\ell} &:= \text{Bias term learned by the tree for leaf } \ell \in L\\ + \end{align*} +""" +from omlt.linear_tree.lt_formulation import ( + LinearTreeGDPFormulation, + LinearTreeHybridBigMFormulation, +) +from omlt.linear_tree.lt_definition import LinearTreeDefinition diff --git a/src/omlt/linear_tree/lt_definition.py b/src/omlt/linear_tree/lt_definition.py new file mode 100644 index 00000000..e45274fd --- /dev/null +++ b/src/omlt/linear_tree/lt_definition.py @@ -0,0 +1,398 @@ +import numpy as np +import lineartree + + +class LinearTreeDefinition: + """ + Class to represent a linear tree model trained in the linear-tree package + + Attributes: + __model (linear-tree model) : Linear Tree Model trained in linear-tree + __splits (dict) : Dict containing split node information + __leaves (dict) : Dict containing leaf node information + __thresholds (dict) : Dict containing splitting threshold information + __scaling_object (scaling object) : Scaling object to ensure scaled + data match units of broader optimization problem + __scaled_input_bounds (dict): Dict containing scaled input bounds + __unscaled_input_bounds (dict): Dict containing unscaled input bounds + + References: + * linear-tree : https://github.com/cerlymarco/linear-tree + """ + + def __init__( + self, + lt_regressor, + scaling_object=None, + scaled_input_bounds=None, + unscaled_input_bounds=None, + ): + """Create a LinearTreeDefinition object and define attributes based on the + trained linear model decision tree. + + Arguments: + lt_regressor -- A LinearTreeRegressor model that is trained by the + linear-tree package + + Keyword Arguments: + scaling_object -- A scaling object to specify the scaling parameters + for the linear model tree inputs and outputs. If None, then no + scaling is performed. (default: {None}) + scaled_input_bounds -- A dict that contains the bounds on the scaled + variables (the direct inputs to the tree). If None, then the + user must specify the bounds via the input_bounds argument. + (default: {None}) + unscaled_input_bounds -- A dict that contains the bounds on the + variables (the direct inputs to the tree). If None, then the + user must specify the scaled bounds via the scaled_input_bounds + argument. (default: {None}) + + Raises: + Exception: Input bounds required. If unscaled_input_bounds and + scaled_input_bounds is None, raise Exception. + """ + self.__model = lt_regressor + self.__scaling_object = scaling_object + + # Process input bounds to insure scaled input bounds exist for formulations + if scaled_input_bounds is None: + if unscaled_input_bounds is not None and scaling_object is not None: + lbs = scaling_object.get_scaled_input_expressions( + {k: t[0] for k, t in unscaled_input_bounds.items()} + ) + ubs = scaling_object.get_scaled_input_expressions( + {k: t[1] for k, t in unscaled_input_bounds.items()} + ) + + scaled_input_bounds = { + k: (lbs[k], ubs[k]) for k in unscaled_input_bounds.keys() + } + + # If unscaled input bounds provided and no scaler provided, scaled + # input bounds = unscaled input bounds + elif unscaled_input_bounds is not None and scaling_object is None: + scaled_input_bounds = unscaled_input_bounds + elif unscaled_input_bounds is None: + raise ValueError( + "Input Bounds needed to represent linear trees as MIPs" + ) + + self.__unscaled_input_bounds = unscaled_input_bounds + self.__scaled_input_bounds = scaled_input_bounds + + self.__splits, self.__leaves, self.__thresholds = _parse_tree_data( + lt_regressor, scaled_input_bounds + ) + + self.__n_inputs = _find_n_inputs(self.__leaves) + self.__n_outputs = 1 + + @property + def scaling_object(self): + """Returns scaling object""" + return self.__scaling_object + + @property + def scaled_input_bounds(self): + """Returns dict containing scaled input bounds""" + return self.__scaled_input_bounds + + @property + def splits(self): + """Returns dict containing split information""" + return self.__splits + + @property + def leaves(self): + """Returns dict containing leaf information""" + return self.__leaves + + @property + def thresholds(self): + """Returns dict containing threshold information""" + return self.__thresholds + + @property + def n_inputs(self): + """Returns number of inputs to the linear tree""" + return self.__n_inputs + + @property + def n_outputs(self): + """Returns number of outputs to the linear tree""" + return self.__n_outputs + + +def _find_all_children_splits(split, splits_dict): + """ + This helper function finds all multigeneration children splits for an + argument split. + + Arguments: + split --The split for which you are trying to find children splits + splits_dict -- A dictionary of all the splits in the tree + + Returns: + A list containing the Node IDs of all children splits + """ + all_splits = [] + + # Check if the immediate left child of the argument split is also a split. + # If so append to the list then use recursion to generate the remainder + left_child = splits_dict[split]["children"][0] + if left_child in splits_dict: + all_splits.append(left_child) + all_splits.extend(_find_all_children_splits(left_child, splits_dict)) + + # Same as above but with right child + right_child = splits_dict[split]["children"][1] + if right_child in splits_dict: + all_splits.append(right_child) + all_splits.extend(_find_all_children_splits(right_child, splits_dict)) + + return all_splits + + +def _find_all_children_leaves(split, splits_dict, leaves_dict): + """ + This helper function finds all multigeneration children leaves for an + argument split. + + Arguments: + split -- The split for which you are trying to find children leaves + splits_dict -- A dictionary of all the split info in the tree + leaves_dict -- A dictionary of all the leaf info in the tree + + Returns: + A list containing all the Node IDs of all children leaves + """ + all_leaves = [] + + # Find all the splits that are children of the relevant split + all_splits = _find_all_children_splits(split, splits_dict) + + # Ensure the current split is included + if split not in all_splits: + all_splits.append(split) + + # For each leaf, check if the parents appear in the list of children + # splits (all_splits). If so, it must be a leaf of the argument split + for leaf in leaves_dict: + if leaves_dict[leaf]["parent"] in all_splits: + all_leaves.append(leaf) + + return all_leaves + + +def _find_n_inputs(leaves): + """ + Finds the number of inputs using the length of the slope vector in the + first leaf + + Arguments: + leaves -- Dictionary of leaf information + + Returns: + Number of inputs + """ + tree_indices = np.array(list(leaves.keys())) + leaf_indices = np.array(list(leaves[tree_indices[0]].keys())) + tree_one = tree_indices[0] + leaf_one = leaf_indices[0] + n_inputs = len(np.arange(0, len(leaves[tree_one][leaf_one]["slope"]))) + return n_inputs + + +def _reassign_none_bounds(leaves, input_bounds): + """ + This helper function reassigns bounds that are None to the bounds + input by the user + + Arguments: + leaves -- The dictionary of leaf information. Attribute of the + LinearTreeDefinition object + input_bounds -- The nested dictionary + + Returns: + The modified leaves dict without any bounds that are listed as None + """ + leaf_indices = np.array(list(leaves.keys())) + leaf_one = leaf_indices[0] + features = np.arange(0, len(leaves[leaf_one]["slope"])) + + for leaf in leaf_indices: + for feat in features: + if leaves[leaf]["bounds"][feat][0] is None: + leaves[leaf]["bounds"][feat][0] = input_bounds[feat][0] + if leaves[leaf]["bounds"][feat][1] is None: + leaves[leaf]["bounds"][feat][1] = input_bounds[feat][1] + + return leaves + + +def _parse_tree_data(model, input_bounds): + """ + This function creates the data structures with the information required + for creation of the variables, sets, and constraints in the pyomo + reformulation of the linear model decision trees. Note that these data + structures are attributes of the LinearTreeDefinition Class. + + Arguments: + model -- Trained linear-tree model or dic containing linear-tree model + summary (e.g. dict = model.summary()) + + Returns: + leaves - Dict containing the following information for each leaf: + 1) 'slope' - The slope of the fitted line at that leaf + 2) 'intercept' - The intercept of the line at that lead + 3) 'parent' - The parent split or node of that leaf + splits - Dict containing the following information for each split: + 1) 'children' - The child nodes of that split + 2) 'col' - The variable(feature) to split on (beginning at 0) + 3) 'left_leaves' - All the leaves to the left of that split + 4) 'right_leaves' - All the leaves to the right of that split + 5) 'parent' - The parent node of the split. Node zero has no parent + 6) 'th' - The threshold of the split + 7) 'y_index' - Indices corresponding to Mistry et. al. y binary + variable + vars_dict - Dict of tree inputs and their respective thresholds + + Raises: + Exception: If input dict is not equal to model.summary() + Exception: If input model is not a dict or linear-tree instance + """ + # Create the initial leaves and splits dictionaries depending on the + # instance of the model (can be either a LinearTreeRegressor or dict). + # Include checks to ensure that the input dict is the model summary which + # is obtained by calling the summary() method contained within the + # linear-tree package (e.g. dict = model.summary()) + if isinstance(model, lineartree.lineartree.LinearTreeRegressor) is True: + leaves = model.summary(only_leaves=True) + splits = model.summary() + elif isinstance(model, dict) is True: + splits = model + leaves = {} + num_splits_in_model = 0 + count = 0 + # Checks to ensure that the input nested dictionary contains the + # correct information + for entry in model: + if "children" not in model[entry].keys(): + leaves[entry] = model[entry] + else: + left_child = model[entry]["children"][0] + right_child = model[entry]["children"][1] + num_splits_in_model += 1 + if left_child not in model.keys() or right_child not in model.keys(): + count += 1 + if count > 0 or num_splits_in_model == 0: + raise ValueError( + "Input dict must be the summary of the linear-tree model" + + " e.g. dict = model.summary()" + ) + else: + raise TypeError("Model entry must be dict or linear-tree instance") + + # This loop adds keys for the slopes and intercept and removes the leaf + # keys in the splits dictionary + for leaf in leaves: + del splits[leaf] + leaves[leaf]["slope"] = list(leaves[leaf]["models"].coef_) + leaves[leaf]["intercept"] = leaves[leaf]["models"].intercept_ + + # This loop creates an parent node id entry for each node in the tree + for split in splits: + left_child = splits[split]["children"][0] + right_child = splits[split]["children"][1] + + if left_child in splits: + splits[left_child]["parent"] = split + else: + leaves[left_child]["parent"] = split + + if right_child in splits: + splits[right_child]["parent"] = split + else: + leaves[right_child]["parent"] = split + + # This loop creates an entry for the all the leaves to the left and right + # of a split + for split in splits: + left_child = splits[split]["children"][0] + right_child = splits[split]["children"][1] + + if left_child in splits: + splits[split]["left_leaves"] = _find_all_children_leaves( + left_child, splits, leaves + ) + else: + splits[split]["left_leaves"] = [left_child] + + if right_child in splits: + splits[split]["right_leaves"] = _find_all_children_leaves( + right_child, splits, leaves + ) + else: + splits[split]["right_leaves"] = [right_child] + + # For each variable that appears in the tree, go through all the splits + # and record its splitting threshold + splitting_thresholds = {} + for split in splits: + var = splits[split]["col"] + splitting_thresholds[var] = {} + for split in splits: + var = splits[split]["col"] + splitting_thresholds[var][split] = splits[split]["th"] + + # Make sure every nested dictionary in the splitting_thresholds dictionary + # is sorted by value + for var in splitting_thresholds: + splitting_thresholds[var] = dict( + sorted(splitting_thresholds[var].items(), key=lambda x: x[1]) + ) + + # NOTE: Can eliminate if not implementing the Mistry et. al. formulations + # Record the ordered indices of the binary variable y. The first index + # is the splitting variable. The second index is its location in the + # ordered dictionary of thresholds for that variable. + for split in splits: + var = splits[split]["col"] + splits[split]["y_index"] = [] + splits[split]["y_index"].append(var) + splits[split]["y_index"].append(list(splitting_thresholds[var]).index(split)) + + # For each leaf, create an empty dictionary that will store the lower + # and upper bounds of each feature. + for leaf in leaves: + leaves[leaf]["bounds"] = {} + + leaf_ids = np.array(list(leaves.keys())) + features = np.arange(0, len(leaves[leaf_ids[0]]["slope"])) + + # For each feature in each leaf, initialize lower and upper bounds to None + for feat in features: + for leaf in leaves: + leaves[leaf]["bounds"][feat] = [None, None] + + # Finally, go through each split and assign it's threshold value as the + # upper bound to all the leaves descending to the left of the split and + # as the lower bound to all the leaves descending to the right. + for split in splits: + var = splits[split]["col"] + for leaf in splits[split]["left_leaves"]: + leaves[leaf]["bounds"][var][1] = splits[split]["th"] + + for leaf in splits[split]["right_leaves"]: + leaves[leaf]["bounds"][var][0] = splits[split]["th"] + + leaves = _reassign_none_bounds(leaves, input_bounds) + + # We use the same formulations developed for gradient boosted linear trees + # so we nest the leaves, splits, and thresholds attributes in a "one-tree" + # tree. + leaves = {0: leaves} + splits = {0: splits} + splitting_thresholds = {0: splitting_thresholds} + + return splits, leaves, splitting_thresholds diff --git a/src/omlt/linear_tree/lt_formulation.py b/src/omlt/linear_tree/lt_formulation.py new file mode 100644 index 00000000..4f83e7f3 --- /dev/null +++ b/src/omlt/linear_tree/lt_formulation.py @@ -0,0 +1,381 @@ +import numpy as np +import pyomo.environ as pe +from pyomo.gdp import Disjunct + +from omlt.formulation import _PyomoFormulation, _setup_scaled_inputs_outputs + + +class LinearTreeGDPFormulation(_PyomoFormulation): + r""" + Class to add a Linear Tree GDP formulation to OmltBlock. We use Pyomo.GDP + to create the disjuncts and disjunctions and then apply a transformation + to convert to a mixed-integer programming representation. + + .. math:: + \begin{align*} + & \underset{\ell \in L}{\bigvee} \left[ \begin{gathered} + Z_{\ell} \\ + \underline{x}_{\ell} \leq x \leq \overline{x}_{\ell} \\ + d = a_{\ell}^T x + b_{\ell} \end{gathered} \right] \\ + & \texttt{exactly_one} \{ Z_{\ell} : \ell \in L \} \\ + & x^L \leq x \leq x^U \\ + & x \in \mathbb{R}^n \\ + & Z_{\ell} \in \{ \texttt{True, False} \} \quad \forall \ \ell \in L + \end{align*} + + Additional nomenclature for this formulation is as follows: + + .. math:: + \begin{align*} + Z_{\ell} &:= \text{Boolean variable indicating which leaf is selected} \\ + \overline{x}_{\ell} &:= \text{Vector of upper bounds for leaf } \ell \in L \\ + \underline{x}_{\ell} &:= \text{Vector of lower bounds for leaf } \ell \in L \\ + x^U &:= \text{Vector of global upper bounds} \\ + x^L &:= \text{Vector of global lower bounds} \\ + \end{align*} + + + Attributes: + Inherited from _PyomoFormulation Class + model_definition : LinearTreeDefinition object + transformation : choose which transformation to apply. The supported + transformations are bigm, mbigm, hull, and custom. + + References: + * Ammari et al. (2023) Linear Model Decision Trees as Surrogates in + Optimization of Engineering Applications. Computers & Chemical Engineering + * Chen et al. (2022) Pyomo.GDP: An ecosystem for logic based modeling and + optimization development. Optimization and Engineering, 23:607–642 + """ + + def __init__(self, lt_definition, transformation="bigm"): + """ + Create a LinearTreeGDPFormulation object + + Arguments: + lt_definition -- LinearTreeDefintion Object + + Keyword Arguments: + transformation -- choose which Pyomo.GDP formulation to apply. + Supported transformations are bigm, hull, mbigm, and custom + (default: {'bigm'}) + + Raises: + Exception: If transformation not in supported transformations + """ + super().__init__() + self.model_definition = lt_definition + self.transformation = transformation + + # Ensure that the GDP transformation given is supported + supported_transformations = ["bigm", "hull", "mbigm", "custom"] + if transformation not in supported_transformations: + raise NotImplementedError( + "Supported transformations are: bigm, mbigm, hull, and custom" + ) + + @property + def input_indexes(self): + """The indexes of the formulation inputs.""" + return list(range(self.model_definition.n_inputs)) + + @property + def output_indexes(self): + """The indexes of the formulation output.""" + return list(range(self.model_definition.n_outputs)) + + def _build_formulation(self): + """This method is called by the OmltBlock to build the corresponding + mathematical formulation on the Pyomo block. + """ + _setup_scaled_inputs_outputs( + self.block, + self.model_definition.scaling_object, + self.model_definition.scaled_input_bounds, + ) + + _add_gdp_formulation_to_block( + block=self.block, + model_definition=self.model_definition, + input_vars=self.block.scaled_inputs, + output_vars=self.block.scaled_outputs, + transformation=self.transformation, + ) + + +class LinearTreeHybridBigMFormulation(_PyomoFormulation): + r""" + Class to add a Linear Tree Hybrid Big-M formulation to OmltBlock. + + .. math:: + \begin{align*} + & d = \sum_{\ell \in L} (a_{\ell}^T x + b_{\ell})z_{\ell} \\ + & x_i \leq \sum_{\ell \in L} \overline{x}_{i,\ell} z_{\ell} && + \forall i \in [n] \\ + & x_i \geq \sum_{\ell \in L} \underline{x}_{i,\ell} z_{\ell} && + \forall i \in [n] \\ + & \sum_{\ell \in L} z_{\ell} = 1 + \end{align*} + + Where the following additional nomenclature is defined: + + .. math:: + \begin{align*} + [n] &:= \text{the integer set of variables that the tree splits on + (e.g. [n] = {1, 2, ... , n})} \\ + \overline{x}_{\ell} &:= \text{Vector of upper bounds for leaf } \ell \in L \\ + \underline{x}_{\ell} &:= \text{Vector of lower bounds for leaf } \ell \in L \\ + \end{align*} + + Attributes: + Inherited from _PyomoFormulation Class + model_definition : LinearTreeDefinition object + + """ + + def __init__(self, lt_definition): + """ + Create a LinearTreeHybridBigMFormulation object + + Arguments: + lt_definition -- LinearTreeDefinition Object + """ + super().__init__() + self.model_definition = lt_definition + + @property + def input_indexes(self): + """The indexes of the formulation inputs.""" + return list(range(self.model_definition.n_inputs)) + + @property + def output_indexes(self): + """The indexes of the formulation output.""" + return list(range(self.model_definition.n_outputs)) + + def _build_formulation(self): + """This method is called by the OmltBlock to build the corresponding + mathematical formulation on the Pyomo block. + """ + _setup_scaled_inputs_outputs( + self.block, + self.model_definition.scaling_object, + self.model_definition.scaled_input_bounds, + ) + + _add_hybrid_formulation_to_block( + block=self.block, + model_definition=self.model_definition, + input_vars=self.block.scaled_inputs, + output_vars=self.block.scaled_outputs, + ) + + +def _build_output_bounds(model_def, input_bounds): + """ + This helper function develops bounds of the output variable based on the + values of the input_bounds and the signs of the slope + + Arguments: + model_def -- Model definition + input_bounds -- Dict of input bounds + + Returns: + List that contains the conservative lower and upper bounds of the + output variable + """ + leaves = model_def.leaves + n_inputs = model_def.n_inputs + tree_ids = np.array(list(leaves.keys())) + features = np.arange(0, n_inputs) + + # Initialize bounds and variables + bounds = [0, 0] + upper_bound = 0 + lower_bound = 0 + for tree in tree_ids: + for leaf in leaves[tree]: + slopes = leaves[tree][leaf]["slope"] + intercept = leaves[tree][leaf]["intercept"] + for k in features: + if slopes[k] <= 0: + upper_bound += slopes[k] * input_bounds[k][0] + intercept + lower_bound += slopes[k] * input_bounds[k][1] + intercept + else: + upper_bound += slopes[k] * input_bounds[k][1] + intercept + lower_bound += slopes[k] * input_bounds[k][0] + intercept + if upper_bound >= bounds[1]: + bounds[1] = upper_bound + if lower_bound <= bounds[0]: + bounds[0] = lower_bound + upper_bound = 0 + lower_bound = 0 + + return bounds + + +def _add_gdp_formulation_to_block( + block, model_definition, input_vars, output_vars, transformation +): + """ + This function adds the GDP representation to the OmltBlock using Pyomo.GDP + + Arguments: + block -- OmltBlock + model_definition -- LinearTreeDefinition Object + input_vars -- input variables to the linear tree model + output_vars -- output variable of the linear tree model + transformation -- Transformation to apply + + """ + leaves = model_definition.leaves + input_bounds = model_definition.scaled_input_bounds + n_inputs = model_definition.n_inputs + + # The set of leaves and the set of features + tree_ids = list(leaves.keys()) + t_l = [] + for tree in tree_ids: + for leaf in leaves[tree].keys(): + t_l.append((tree, leaf)) + features = np.arange(0, n_inputs) + + # Use the input_bounds and the linear models in the leaves to calculate + # the lower and upper bounds on the output variable. Required for Pyomo.GDP + output_bounds = _build_output_bounds(model_definition, input_bounds) + + # Ouptuts are automatically scaled based on whether inputs are scaled + block.outputs.setub(output_bounds[1]) + block.outputs.setlb(output_bounds[0]) + block.scaled_outputs.setub(output_bounds[1]) + block.scaled_outputs.setlb(output_bounds[0]) + + block.intermediate_output = pe.Var( + tree_ids, bounds=(output_bounds[0], output_bounds[1]) + ) + + # Create a disjunct for each leaf containing the bound constraints + # and the linear model expression. + def disjuncts_rule(dsj, tree, leaf): + def lb_rule(dsj, feat): + return input_vars[feat] >= leaves[tree][leaf]["bounds"][feat][0] + + dsj.lb_constraint = pe.Constraint(features, rule=lb_rule) + + def ub_rule(dsj, feat): + return input_vars[feat] <= leaves[tree][leaf]["bounds"][feat][1] + + dsj.ub_constraint = pe.Constraint(features, rule=ub_rule) + + slope = leaves[tree][leaf]["slope"] + intercept = leaves[tree][leaf]["intercept"] + dsj.linear_exp = pe.Constraint( + expr=sum(slope[k] * input_vars[k] for k in features) + intercept + == block.intermediate_output[tree] + ) + + block.disjunct = Disjunct(t_l, rule=disjuncts_rule) + + @block.Disjunction(tree_ids) + def disjunction_rule(b, tree): + leaf_ids = list(leaves[tree].keys()) + return [block.disjunct[tree, leaf] for leaf in leaf_ids] + + block.total_output = pe.Constraint( + expr=output_vars[0] == sum(block.intermediate_output[tree] for tree in tree_ids) + ) + + transformation_string = "gdp." + transformation + + if transformation != "custom": + pe.TransformationFactory(transformation_string).apply_to(block) + + +def _add_hybrid_formulation_to_block(block, model_definition, input_vars, output_vars): + """ + This function adds the Hybrid BigM representation to the OmltBlock + + Arguments: + block -- OmltBlock + model_definition -- LinearTreeDefinition Object + input_vars -- input variables to the linear tree model + output_vars -- output variable of the linear tree model + """ + leaves = model_definition.leaves + input_bounds = model_definition.scaled_input_bounds + n_inputs = model_definition.n_inputs + + # The set of trees + tree_ids = list(leaves.keys()) + # Create a list of tuples that contains the tree and leaf indices. Note that + # the leaf indices depend on the tree in the ensemble. + t_l = [] + for tree in tree_ids: + for leaf in leaves[tree].keys(): + t_l.append((tree, leaf)) + + features = np.arange(0, n_inputs) + + # Use the input_bounds and the linear models in the leaves to calculate + # the lower and upper bounds on the output variable. Required for Pyomo.GDP + output_bounds = _build_output_bounds(model_definition, input_bounds) + + # Ouptuts are automatically scaled based on whether inputs are scaled + block.outputs.setub(output_bounds[1]) + block.outputs.setlb(output_bounds[0]) + block.scaled_outputs.setub(output_bounds[1]) + block.scaled_outputs.setlb(output_bounds[0]) + + # Create the intermeditate variables. z is binary that indicates which leaf + # in tree t is returned. intermediate_output is the output of tree t and + # the total output of the model is the sum of the intermediate_output vars + block.z = pe.Var(t_l, within=pe.Binary) + block.intermediate_output = pe.Var(tree_ids) + + @block.Constraint(features, tree_ids) + def lower_bound_constraints(mdl, feat, tree): + leaf_ids = list(leaves[tree].keys()) + return ( + sum( + leaves[tree][leaf]["bounds"][feat][0] * mdl.z[tree, leaf] + for leaf in leaf_ids + ) + <= input_vars[feat] + ) + + @block.Constraint(features, tree_ids) + def upper_bound_constraints(mdl, feat, tree): + leaf_ids = list(leaves[tree].keys()) + return ( + sum( + leaves[tree][leaf]["bounds"][feat][1] * mdl.z[tree, leaf] + for leaf in leaf_ids + ) + >= input_vars[feat] + ) + + @block.Constraint(tree_ids) + def linear_constraint(mdl, tree): + leaf_ids = list(leaves[tree].keys()) + return block.intermediate_output[tree] == sum( + ( + sum( + leaves[tree][leaf]["slope"][feat] * input_vars[feat] + for feat in features + ) + + leaves[tree][leaf]["intercept"] + ) + * block.z[tree, leaf] + for leaf in leaf_ids + ) + + @block.Constraint(tree_ids) + def only_one_leaf_per_tree(b, tree): + leaf_ids = list(leaves[tree].keys()) + return sum(block.z[tree, leaf] for leaf in leaf_ids) == 1 + + @block.Constraint() + def output_sum_of_trees(b): + return output_vars[0] == sum( + block.intermediate_output[tree] for tree in tree_ids + ) diff --git a/tests/linear_tree/test_lt_formulation.py b/tests/linear_tree/test_lt_formulation.py new file mode 100644 index 00000000..0b1fc59d --- /dev/null +++ b/tests/linear_tree/test_lt_formulation.py @@ -0,0 +1,765 @@ +import numpy as np +import pyomo.environ as pe +import pytest + +from omlt.dependencies import lineartree_available +from pytest import approx + +if lineartree_available: + from lineartree import LinearTreeRegressor + from sklearn.linear_model import LinearRegression + from omlt.linear_tree import ( + LinearTreeGDPFormulation, + LinearTreeHybridBigMFormulation, + LinearTreeDefinition, + ) + +from omlt import OmltBlock +import omlt + +scip_available = pe.SolverFactory("scip").available() +cbc_available = pe.SolverFactory("cbc").available() +gurobi_available = pe.SolverFactory("gurobi").available() + + +def linear_model_tree(X, y): + regr = LinearTreeRegressor(LinearRegression(), criterion="mse", max_depth=5) + regr.fit(X, y) + return regr + + +### SINGLE VARIABLE INPUT TESTING #### + +X_small = np.array( + [ + [-0.68984135], + [0.91672866], + [-1.05874972], + [0.95275351], + [1.03796615], + [0.45117668], + [-0.14704376], + [1.66043409], + [-0.73972191], + [-0.8176603], + [0.96175973], + [-1.238874], + [-0.97492265], + [1.07121986], + [-0.95379269], + [-0.86546252], + [0.8277057], + [0.50486757], + [-1.38435899], + [1.54092856], + ] +) + +y_small = np.array( + [ + [0.04296633], + [-0.78349216], + [0.27114188], + [-0.58516476], + [-0.15997756], + [-0.37529212], + [-1.49249696], + [1.56412122], + [0.18697725], + [0.4035928], + [-0.53231771], + [-0.02669967], + [0.36972983], + [0.09201347], + [0.44041505], + [0.46047019], + [-1.04855941], + [-0.586915], + [0.15472157], + [1.71225268], + ] +) + + +@pytest.mark.skipif(not lineartree_available, reason="Need Linear-Tree Package") +def test_linear_tree_model_single_var(): + # construct a LinearTreeDefinition + regr_small = linear_model_tree(X=X_small, y=y_small) + input_bounds = {0: (min(X_small)[0], max(X_small)[0])} + ltmodel_small = LinearTreeDefinition(regr_small, unscaled_input_bounds=input_bounds) + + scaled_input_bounds = ltmodel_small.scaled_input_bounds + n_inputs = ltmodel_small.n_inputs + n_outputs = ltmodel_small.n_outputs + splits = ltmodel_small.splits + leaves = ltmodel_small.leaves + thresholds = ltmodel_small.thresholds + + assert scaled_input_bounds is not None + assert n_inputs == 1 + assert n_outputs == 1 + # test for splits + # assert the number of splits + assert len(splits[0].keys()) == 5 + splits_key_list = [ + "col", + "th", + "loss", + "samples", + "parent", + "children", + "models", + "left_leaves", + "right_leaves", + "y_index", + ] + # assert whether all the dicts have such keys + for i in splits[0].keys(): + for key in splits[0][i].keys(): + assert key in splits_key_list + # test for leaves + # assert the number of leaves + assert len(leaves[0].keys()) == 6 + # assert whether all the dicts have such keys + leaves_key_list = [ + "loss", + "samples", + "models", + "slope", + "intercept", + "parent", + "bounds", + ] + for j in leaves[0].keys(): + for key in leaves[0][j].keys(): + assert key in leaves_key_list + # if the key is slope, ensure slope dimension match n_inputs + if key == "slope": + assert len(leaves[0][j][key]) == n_inputs + # elif the key is bounds, test ensure lb <= ub + elif key == "bounds": + features = leaves[0][j][key].keys() + for k in range(len(features)): + lb = leaves[0][j][key][k][0] + ub = leaves[0][j][key][k][1] + # there is chance that don't have lb and ub at this step + if lb is not None and ub is not None: + assert lb <= ub + # test for thresholds + # assert whether each feature has threshold + assert len(thresholds[0].keys()) == n_inputs + # assert the number of thresholds + thresholds_count = 0 + for k in range(len(thresholds[0].keys())): + for _ in range(len(thresholds[0][k].keys())): + thresholds_count += 1 + assert thresholds_count == len(splits[0].keys()) + + +@pytest.mark.skipif( + not lineartree_available or not cbc_available, + reason="Need Linear-Tree Package and cbc", +) +def test_bigm_formulation_single_var(): + regr_small = linear_model_tree(X=X_small, y=y_small) + input_bounds = {0: (min(X_small)[0], max(X_small)[0])} + ltmodel_small = LinearTreeDefinition(regr_small, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeGDPFormulation(ltmodel_small, transformation="bigm") + + model1 = pe.ConcreteModel() + model1.x = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_inputs(mdl): + return mdl.x == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x.fix(0.5) + + status_1_bigm = pe.SolverFactory("cbc").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = (pe.value(model1.x), pe.value(model1.y)) + y_pred = regr_small.predict(np.array(solution_1_bigm[0]).reshape(1, -1)) + assert y_pred[0] == approx(solution_1_bigm[1]) + + +@pytest.mark.skipif( + not lineartree_available or not cbc_available, + reason="Need Linear-Tree Package and cbc", +) +def test_hull_formulation_single_var(): + regr_small = linear_model_tree(X=X_small, y=y_small) + input_bounds = {0: (min(X_small)[0], max(X_small)[0])} + ltmodel_small = LinearTreeDefinition(regr_small, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeGDPFormulation(ltmodel_small, transformation="hull") + + model1 = pe.ConcreteModel() + model1.x = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_inputs(mdl): + return mdl.x == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x.fix(0.5) + + status_1_bigm = pe.SolverFactory("cbc").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = (pe.value(model1.x), pe.value(model1.y)) + y_pred = regr_small.predict(np.array(solution_1_bigm[0]).reshape(1, -1)) + assert y_pred[0] == approx(solution_1_bigm[1]) + + +@pytest.mark.skipif( + not lineartree_available or not gurobi_available, + reason="Need Linear-Tree Package and gurobi", +) +def test_mbigm_formulation_single_var(): + regr_small = linear_model_tree(X=X_small, y=y_small) + input_bounds = {0: (min(X_small)[0], max(X_small)[0])} + ltmodel_small = LinearTreeDefinition(regr_small, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeGDPFormulation(ltmodel_small, transformation="mbigm") + + model1 = pe.ConcreteModel() + model1.x = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_inputs(mdl): + return mdl.x == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x.fix(0.5) + + status_1_bigm = pe.SolverFactory("gurobi").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = (pe.value(model1.x), pe.value(model1.y)) + y_pred = regr_small.predict(np.array(solution_1_bigm[0]).reshape(1, -1)) + assert y_pred[0] == approx(solution_1_bigm[1]) + + +@pytest.mark.skipif( + not lineartree_available or not scip_available, + reason="Need Linear-Tree Package and scip", +) +def test_hybrid_bigm_formulation_single_var(): + regr_small = linear_model_tree(X=X_small, y=y_small) + input_bounds = {0: (min(X_small)[0], max(X_small)[0])} + ltmodel_small = LinearTreeDefinition(regr_small, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeHybridBigMFormulation(ltmodel_small) + + model1 = pe.ConcreteModel() + model1.x = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_inputs(mdl): + return mdl.x == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x.fix(0.5) + + status_1_bigm = pe.SolverFactory("scip").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = (pe.value(model1.x), pe.value(model1.y)) + y_pred = regr_small.predict(np.array(solution_1_bigm[0]).reshape(1, -1)) + assert y_pred[0] == approx(solution_1_bigm[1]) + + +@pytest.mark.skipif(not lineartree_available, reason="Need Linear-Tree Package") +def test_scaling(): + mean_x_small = np.mean(X_small) + std_x_small = np.std(X_small) + mean_y_small = np.mean(y_small) + std_y_small = np.std(y_small) + scaled_x = (X_small - mean_x_small) / std_x_small + scaled_y = (y_small - mean_y_small) / std_y_small + scaled_input_bounds = {0: (np.min(scaled_x), np.max(scaled_x))} + unscaled_input_bounds = {0: (np.min(X_small), np.max(X_small))} + + scaler = omlt.scaling.OffsetScaling( + offset_inputs=[mean_x_small], + factor_inputs=[std_x_small], + offset_outputs=[mean_y_small], + factor_outputs=[std_y_small], + ) + + regr = linear_model_tree(scaled_x, scaled_y) + + regr.fit(np.reshape(scaled_x, (-1, 1)), scaled_y) + + lt_def2 = LinearTreeDefinition( + regr, unscaled_input_bounds=unscaled_input_bounds, scaling_object=scaler + ) + assert lt_def2.scaled_input_bounds[0][0] == approx(scaled_input_bounds[0][0]) + assert lt_def2.scaled_input_bounds[0][1] == approx(scaled_input_bounds[0][1]) + with pytest.raises( + Exception, match="Input Bounds needed to represent linear trees as MIPs" + ): + ltmodel_scaled = LinearTreeDefinition(regr) + + +#### MULTIVARIATE INPUT TESTING #### + +X = np.array( + [ + [4.98534526, 1.8977914], + [4.38751717, 4.48456528], + [2.65451539, 2.44426211], + [3.32761277, 4.58757063], + [0.36806515, 0.82428634], + [4.16036314, 1.09680059], + [2.29025371, 0.72246559], + [1.92725929, 0.34359974], + [4.02101578, 1.39448628], + [3.28019501, 1.22160752], + [2.73026047, 3.9482306], + [0.45621172, 0.56130164], + [2.64296795, 4.75411397], + [4.72526084, 3.35223772], + [2.39270941, 4.41622262], + [4.42707908, 0.35276571], + [1.58452501, 3.28957671], + [0.20009184, 2.90255483], + [4.36453075, 3.61985047], + [1.05576503, 2.57532169], + ] +) + +Y = np.array( + [ + [10.23341638], + [4.00860872], + [3.85046103], + [9.48457266], + [6.36974536], + [3.19763555], + [4.78390803], + [1.51994021], + [3.18768132], + [3.7972809], + [7.93779383], + [3.46714285], + [7.89435163], + [10.62832561], + [1.50713442], + [7.44321537], + [9.39437373], + [4.38891182], + [1.32105126], + [3.37287403], + ] +) + + +@pytest.mark.skipif(not lineartree_available, reason="Need Linear-Tree Package") +def test_linear_tree_model_multi_var(): + # construct a LinearTreeDefinition + regr = linear_model_tree(X=X, y=Y) + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + ltmodel_small = LinearTreeDefinition(regr, unscaled_input_bounds=input_bounds) + + scaled_input_bounds = ltmodel_small.scaled_input_bounds + n_inputs = ltmodel_small.n_inputs + n_outputs = ltmodel_small.n_outputs + splits = ltmodel_small.splits + leaves = ltmodel_small.leaves + thresholds = ltmodel_small.thresholds + + # assert attributes in LinearTreeDefinition + assert scaled_input_bounds is not None + assert n_inputs == 2 + assert n_outputs == 1 + + # test for splits + # assert the number of splits + assert len(splits[0].keys()) == 5 + splits_key_list = [ + "col", + "th", + "loss", + "samples", + "parent", + "children", + "models", + "left_leaves", + "right_leaves", + "y_index", + ] + # assert whether all the dicts have such keys + for i in splits[0].keys(): + for key in splits[0][i].keys(): + assert key in splits_key_list + # test for leaves + # assert the number of leaves + assert len(leaves[0].keys()) == 6 + # assert whether all the dicts have such keys + leaves_key_list = [ + "loss", + "samples", + "models", + "slope", + "intercept", + "parent", + "bounds", + ] + for j in leaves[0].keys(): + for key in leaves[0][j].keys(): + assert key in leaves_key_list + # if the key is slope, test the shape of it + if key == "slope": + assert len(leaves[0][j][key]) == n_inputs + # elif the key is bounds, test the lb <= ub + elif key == "bounds": + features = leaves[0][j][key].keys() + for k in range(len(features)): + lb = leaves[0][j][key][k][0] + ub = leaves[0][j][key][k][1] + # there is chance that don't have lb and ub at this step + if lb is not None and ub is not None: + assert lb <= ub + # test for thresholds + # assert whether each feature has threshold + assert len(thresholds[0].keys()) == n_inputs + # assert the number of thresholds + thresholds_count = 0 + for k in range(len(thresholds[0].keys())): + for _ in range(len(thresholds[0][k].keys())): + thresholds_count += 1 + assert thresholds_count == len(splits[0].keys()) + + +@pytest.mark.skipif( + not lineartree_available or not cbc_available, + reason="Need Linear-Tree Package and cbc", +) +def test_bigm_formulation_multi_var(): + regr = linear_model_tree(X=X, y=Y) + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + ltmodel_small = LinearTreeDefinition(regr, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeGDPFormulation(ltmodel_small, transformation="bigm") + + model1 = pe.ConcreteModel() + model1.x0 = pe.Var(initialize=0) + model1.x1 = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_input1(mdl): + return mdl.x0 == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_input2(mdl): + return mdl.x1 == mdl.lt.inputs[1] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x0.fix(0.5) + model1.x1.fix(0.8) + + status_1_bigm = pe.SolverFactory("cbc").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = pe.value(model1.y) + y_pred = regr.predict( + np.array([pe.value(model1.x0), pe.value(model1.x1)]).reshape(1, -1) + ) + assert y_pred[0] == approx(solution_1_bigm) + + +@pytest.mark.skipif( + not lineartree_available or not cbc_available, + reason="Need Linear-Tree Package and cbc", +) +def test_hull_formulation_multi_var(): + regr = linear_model_tree(X=X, y=Y) + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + ltmodel_small = LinearTreeDefinition(regr, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeGDPFormulation(ltmodel_small, transformation="hull") + + model1 = pe.ConcreteModel() + model1.x0 = pe.Var(initialize=0) + model1.x1 = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_input1(mdl): + return mdl.x0 == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_input2(mdl): + return mdl.x1 == mdl.lt.inputs[1] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x0.fix(0.5) + model1.x1.fix(0.8) + + status_1_bigm = pe.SolverFactory("cbc").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = pe.value(model1.y) + y_pred = regr.predict( + np.array([pe.value(model1.x0), pe.value(model1.x1)]).reshape(1, -1) + ) + assert y_pred[0] == approx(solution_1_bigm) + + +@pytest.mark.skipif( + not lineartree_available or not gurobi_available, + reason="Need Linear-Tree Package and gurobi", +) +def test_mbigm_formulation_multi_var(): + regr = linear_model_tree(X=X, y=Y) + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + ltmodel_small = LinearTreeDefinition(regr, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeGDPFormulation(ltmodel_small, transformation="mbigm") + + model1 = pe.ConcreteModel() + model1.x0 = pe.Var(initialize=0) + model1.x1 = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_input1(mdl): + return mdl.x0 == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_input2(mdl): + return mdl.x1 == mdl.lt.inputs[1] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x0.fix(0.5) + model1.x1.fix(0.8) + + status_1_bigm = pe.SolverFactory("gurobi").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = pe.value(model1.y) + y_pred = regr.predict( + np.array([pe.value(model1.x0), pe.value(model1.x1)]).reshape(1, -1) + ) + assert y_pred[0] == approx(solution_1_bigm) + + +@pytest.mark.skipif( + not lineartree_available or not scip_available, + reason="Need Linear-Tree Package and scip", +) +def test_hybrid_bigm_formulation_multi_var(): + regr = linear_model_tree(X=X, y=Y) + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + ltmodel_small = LinearTreeDefinition(regr, unscaled_input_bounds=input_bounds) + formulation1_lt = LinearTreeHybridBigMFormulation(ltmodel_small) + + model1 = pe.ConcreteModel() + model1.x0 = pe.Var(initialize=0) + model1.x1 = pe.Var(initialize=0) + model1.y = pe.Var(initialize=0) + model1.obj = pe.Objective(expr=1) + model1.lt = OmltBlock() + model1.lt.build_formulation(formulation1_lt) + + @model1.Constraint() + def connect_input1(mdl): + return mdl.x0 == mdl.lt.inputs[0] + + @model1.Constraint() + def connect_input2(mdl): + return mdl.x1 == mdl.lt.inputs[1] + + @model1.Constraint() + def connect_outputs(mdl): + return mdl.y == mdl.lt.outputs[0] + + model1.x0.fix(0.5) + model1.x1.fix(0.8) + + status_1_bigm = pe.SolverFactory("scip").solve(model1, tee=True) + pe.assert_optimal_termination(status_1_bigm) + solution_1_bigm = pe.value(model1.y) + y_pred = regr.predict( + np.array([pe.value(model1.x0), pe.value(model1.x1)]).reshape(1, -1) + ) + assert y_pred[0] == approx(solution_1_bigm) + + +@pytest.mark.skipif(not lineartree_available, reason="Need Linear-Tree Package") +def test_summary_dict_as_argument(): + # construct a LinearTreeDefinition + regr = linear_model_tree(X=X, y=Y) + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + ltmodel_small = LinearTreeDefinition( + regr.summary(), unscaled_input_bounds=input_bounds + ) + + scaled_input_bounds = ltmodel_small.scaled_input_bounds + n_inputs = ltmodel_small.n_inputs + n_outputs = ltmodel_small.n_outputs + splits = ltmodel_small.splits + leaves = ltmodel_small.leaves + thresholds = ltmodel_small.thresholds + + # assert attributes in LinearTreeDefinition + assert scaled_input_bounds is not None + assert n_inputs == 2 + assert n_outputs == 1 + # test for splits + # assert the number of splits + assert len(splits[0].keys()) == 5 + splits_key_list = [ + "col", + "th", + "loss", + "samples", + "parent", + "children", + "models", + "left_leaves", + "right_leaves", + "y_index", + ] + # assert whether all the dicts have such keys + for i in splits[0].keys(): + for key in splits[0][i].keys(): + assert key in splits_key_list + # test for leaves + # assert the number of leaves + assert len(leaves[0].keys()) == 6 + # assert whether all the dicts have such keys + leaves_key_list = [ + "loss", + "samples", + "models", + "slope", + "intercept", + "parent", + "bounds", + ] + for j in leaves[0].keys(): + for key in leaves[0][j].keys(): + assert key in leaves_key_list + # if the key is slope, test the shape of it + if key == "slope": + assert len(leaves[0][j][key]) == n_inputs + # elif the key is bounds, test the lb <= ub + elif key == "bounds": + features = leaves[0][j][key].keys() + for k in range(len(features)): + lb = leaves[0][j][key][k][0] + ub = leaves[0][j][key][k][1] + # there is chance that don't have lb and ub at this step + if lb is not None and ub is not None: + assert lb <= ub + # test for thresholds + # assert whether each feature has threshold + assert len(thresholds[0].keys()) == n_inputs + # assert the number of thresholds + thresholds_count = 0 + for k in range(len(thresholds[0].keys())): + for _ in range(len(thresholds[0][k].keys())): + thresholds_count += 1 + assert thresholds_count == len(splits[0].keys()) + + +@pytest.mark.skipif(not lineartree_available, reason="Need Linear-Tree Package") +def test_raise_exception_if_wrong_model_instance(): + regr = linear_model_tree(X=X, y=Y) + wrong_summary_dict = regr.summary() + del wrong_summary_dict[1] + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + with pytest.raises( + Exception, + match="Input dict must be the summary of the linear-tree model" + + " e.g. dict = model.summary()", + ): + ltmodel_small = LinearTreeDefinition( + regr.summary(only_leaves=True), scaled_input_bounds=input_bounds + ) + with pytest.raises( + Exception, match="Model entry must be dict or linear-tree instance" + ): + ltmodel_small = LinearTreeDefinition((0, 0), scaled_input_bounds=input_bounds) + with pytest.raises( + Exception, + match="Input dict must be the summary of the linear-tree model" + + " e.g. dict = model.summary()", + ): + ltmodel_small = LinearTreeDefinition( + wrong_summary_dict, scaled_input_bounds=input_bounds + ) + + +@pytest.mark.skipif(not lineartree_available, reason="Need Linear-Tree Package") +def test_children_node_finders(): + # Train a linear model decision tree + X = np.linspace(-4, 4).reshape((-1, 1)) + Y = np.sin(X) + regr = linear_model_tree(X=X, y=Y) + + # Create a LinearTreeDefinition Object + inBounds = {0: (-4, 4)} + model_def = LinearTreeDefinition(regr, unscaled_input_bounds=inBounds) + + # Extract leaf and split information + spts = model_def.splits + lvs = model_def.leaves + + # Ensure that at the root node, the number of left leaves and the number of + # right leaves sum to the total number of leaves in the tree + num_left_leaves_at_root = len(spts[0][0]["left_leaves"]) + num_right_leaves_at_root = len(spts[0][0]["right_leaves"]) + total_leaves = len(lvs[0]) + + assert num_left_leaves_at_root + num_right_leaves_at_root == total_leaves + + +@pytest.mark.skipif(not lineartree_available, reason="Need Linear-Tree Package") +def test_raise_exception_for_wrong_transformation(): + regr = linear_model_tree(X=X, y=Y) + input_bounds = {0: (min(X[:, 0]), max(X[:, 0])), 1: (min(X[:, 1]), max(X[:, 1]))} + model_def = LinearTreeDefinition(regr, unscaled_input_bounds=input_bounds) + with pytest.raises( + Exception, + match="Supported transformations are: bigm, mbigm, hull, and custom", + ): + formulation = LinearTreeGDPFormulation(model_def, transformation="hello") diff --git a/tox.ini b/tox.ini index 1241bc6c..e1a56c01 100644 --- a/tox.ini +++ b/tox.ini @@ -4,7 +4,7 @@ [tox] minversion = 3.15 -envlist = py36, py37, py38, py39, lint +envlist = py36, py37, py38, py39, py310, lint [gh-actions] python = @@ -12,6 +12,7 @@ python = 3.7: py37 3.8: py38, lint 3.9: py39 + 3.10: py310 [testenv] deps = pytest