diff --git a/.gitignore b/.gitignore
index 434dcb7..3e50103 100644
--- a/.gitignore
+++ b/.gitignore
@@ -35,8 +35,8 @@
**/main.pdf
## temp models
-**/best_model.h5
-**/best_transforer_model.h5
+**/best_model*.h5
+**/best_transforer_model*.h5
## others
*.ipynb_checkpoints
diff --git a/data/README.md b/data/README.md
index 91677c8..ae5057d 100644
--- a/data/README.md
+++ b/data/README.md
@@ -1,5 +1,9 @@
# Data
+Demo datasets (1.3 GB) can be downloaded from zenodo: https://doi.org/10.5281/zenodo.10775099.
+See further details [here](../docs/protocols/experiment-24-aug-2023/README.md) on its collection.
+Notes. you can host up to 50GB of data in zenodo.
+
## Demo data Thu-24-Aug-2023
Tree of demo dataset. Each pair of video and time-series were recorded for approximately 5 minutes.
```
@@ -92,3 +96,7 @@ Tree of demo dataset. Each pair of video and time-series were recorded for appro
2 directories, 20 files
```
+
+## Citations
+
+Hekim, Sujon, and Miguel Xochicale. ‘Demo Dataset for Multimodal Real-time Ai V.1.0.0’. Zenodo, 3 March 2024. https://doi.org/10.5281/zenodo.10775099.
diff --git a/rtt4ssa/models/example-models/README.md b/rtt4ssa/models/example-models/README.md
new file mode 100644
index 0000000..bdf0c12
--- /dev/null
+++ b/rtt4ssa/models/example-models/README.md
@@ -0,0 +1,96 @@
+# Example models for timeseries
+
+## `timeseries_classification_from_scratch.ipynb`
+
+epochs = 500 ## ORIGINAL
+batch_size = 32 ##ORIGINAL
+
+
+```
+=================================================================
+Total params: 25,858
+Trainable params: 25,474
+Non-trainable params: 384
+```
+
+```
+Epoch 256/500
+90/90 [==============================] - 1s 11ms/step - loss: 0.0521 - sparse_categorical_accuracy: 0.9837 - val_loss: 0.0865 - val_sparse_categorical_accuracy: 0.9722 - lr: 1.0000e-04
+Epoch 256: early stopping
+Execution time: 4.0526354789733885 minutes
+```
+
+![fig](Screenshot%20from%202023-10-10%2002-20-17.png)
+
+
+## `timeseries_transformer_classification.ipynb`
+
+### head_size=256; num_heads=4; num_transformer_blocks = 2
+```
+==================================================================================================
+Total params: 78,758
+Trainable params: 78,758
+Non-trainable params: 0
+```
+
+```
+Epoch 77/200
+360/360 [==============================] - 14s 38ms/step - loss: 0.0783 - sparse_categorical_accuracy: 0.9785 - val_loss: 0.4438 - val_sparse_categorical_accuracy: 0.8350
+Execution time: 16.701254685719807 minutes
+```
+
+![fig](Screenshot%20from%202023-10-10%2002-53-14.png)
+
+
+### head_size=256; num_heads=4; num_transformer_blocks = 4 ##ORIGINAL
+
+```
+=================================================================
+Total params: 93,130
+Trainable params: 93,130
+Non-trainable params: 0
+```
+
+```
+Epoch 106/200
+Execution time: 43.552278610070545 minutes
+```
+
+![fig](Screenshot%20from%202023-10-10%2001-54-35.png)
+
+
+### head_size=256; num_heads=4; num_transformer_blocks = 8
+
+```
+=================================================================
+Total params: 121,874
+Trainable params: 121,874
+Non-trainable params: 0
+```
+
+```
+Epoch 97/200
+360/360 [==============================] - 48s 134ms/step - loss: 0.0737 - sparse_categorical_accuracy: 0.9747 - val_loss: 0.3841 - val_sparse_categorical_accuracy: 0.8530
+Execution time: 78.16839495897293 minutes
+```
+
+![fig](Screenshot%20from%202023-10-10%2001-54-43.png)
+
+
+### head_size=512; num_heads=4; num_transformer_blocks = 4
+```
+
+=================================================================
+Total params: 121,802
+Trainable params: 121,802
+Non-trainable params: 0
+```
+
+```
+Epoch 84/200
+360/360 [==============================] - 38s 107ms/step - loss: 0.0758 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.4588 - val_sparse_categorical_accuracy: 0.8516
+Execution time: 54.10847853024801 minutes
+```
+
+![fig](Screenshot%20from%202023-10-10%2001-54-57.png)
+
diff --git a/rtt4ssa/models/example-models/timeseries_classification_from_scratch.ipynb b/rtt4ssa/models/example-models/timeseries_classification_from_scratch.ipynb
index 9deaf41..4aca048 100644
--- a/rtt4ssa/models/example-models/timeseries_classification_from_scratch.ipynb
+++ b/rtt4ssa/models/example-models/timeseries_classification_from_scratch.ipynb
@@ -11,8 +11,8 @@
"**Author:** [hfawaz](https://github.com/hfawaz/)
\n",
"**Replication:** [mxochicale](https://github.com/mxochicale/)
\n",
"**Date created:** 2020/07/21
\n",
- "**Description:** Training a timeseries classifier from scratch on the FordA dataset from the UCR/UEA archive.\n",
- "\n",
+ "**Description:** Training a timeseries classifier from scratch on the FordA dataset from the UCR/UEA archive. \n",
+ "**Reference:** https://keras.io/examples/timeseries/timeseries_classification_from_scratch/ \n",
"\n",
"**Logs:**\n",
"* MX: Fri 21 Jul 21:45:20 BST 2023 # For ploting. I tried plot_history(history) but needs some refactor and maybe add it in utils. https://gist.github.com/whyboris/91ee793ddc92cf1e824978cf31bb790c\n",
@@ -43,12 +43,23 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 1,
"metadata": {
"id": "vSMYnBXvh8zq"
},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "2024-03-03 12:44:28.501017: I tensorflow/core/util/port.cc:110] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
+ "2024-03-03 12:44:28.757168: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
+ "To enable the following instructions: SSE4.1 SSE4.2 AVX AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
+ ]
+ }
+ ],
"source": [
+ "import time\n",
"from tensorflow import keras\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
@@ -94,7 +105,7 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 2,
"metadata": {
"id": "q4WhCJC0h8zu"
},
@@ -127,7 +138,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 3,
"metadata": {},
"outputs": [
{
@@ -186,7 +197,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 4,
"metadata": {
"id": "WxQZPakth8zw"
},
@@ -315,7 +326,7 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 5,
"metadata": {
"id": "8RKhPnTmh8zy"
},
@@ -351,7 +362,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 6,
"metadata": {
"id": "0pNBqQ-Gh8zz"
},
@@ -383,7 +394,7 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 7,
"metadata": {
"id": "yq23pCEXh8zz"
},
@@ -406,7 +417,7 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 8,
"metadata": {
"id": "IP3XNJXth8z0"
},
@@ -434,7 +445,7 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 9,
"metadata": {
"id": "xqc5-9deh8z0"
},
@@ -467,7 +478,7 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 10,
"metadata": {},
"outputs": [
{
@@ -482,16 +493,16 @@
"name": "stderr",
"output_type": "stream",
"text": [
- "2023-09-30 03:15:35.125892: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.160165: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.160437: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.165051: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.165365: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.165562: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.763479: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.763786: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.763959: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
- "2023-09-30 03:15:35.764107: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1635] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 6202 MB memory: -> device: 0, name: NVIDIA RTX A2000 8GB Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 8.6\n"
+ "2024-03-03 12:45:19.209008: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:19.306442: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:19.306804: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:19.311046: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:19.311336: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:19.311522: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:20.628451: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:20.628916: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:20.629172: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:996] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
+ "2024-03-03 12:45:20.629665: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1635] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 6202 MB memory: -> device: 0, name: NVIDIA RTX A2000 8GB Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 8.6\n"
]
},
{
@@ -508,7 +519,7 @@
""
]
},
- "execution_count": 11,
+ "execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
@@ -540,143 +551,528 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "Epoch 1/50\n"
+ "Epoch 1/500\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "2023-08-09 08:00:02.831156: I tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:424] Loaded cuDNN version 8800\n",
- "2023-08-09 08:00:04.803225: I tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:637] TensorFloat-32 will be used for the matrix multiplication. This will only be logged once.\n",
- "2023-08-09 08:00:04.814706: I tensorflow/compiler/xla/service/service.cc:169] XLA service 0x7fa174f30ad0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
- "2023-08-09 08:00:04.814777: I tensorflow/compiler/xla/service/service.cc:177] StreamExecutor device (0): NVIDIA RTX A2000 8GB Laptop GPU, Compute Capability 8.6\n",
- "2023-08-09 08:00:04.822325: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
- "2023-08-09 08:00:04.986316: I ./tensorflow/compiler/jit/device_compiler.h:180] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n"
+ "2024-03-03 12:45:58.768981: I tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:424] Loaded cuDNN version 8800\n",
+ "2024-03-03 12:46:00.762874: I tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:637] TensorFloat-32 will be used for the matrix multiplication. This will only be logged once.\n",
+ "2024-03-03 12:46:00.777074: I tensorflow/compiler/xla/service/service.cc:169] XLA service 0x7f9b39104d10 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
+ "2024-03-03 12:46:00.777116: I tensorflow/compiler/xla/service/service.cc:177] StreamExecutor device (0): NVIDIA RTX A2000 8GB Laptop GPU, Compute Capability 8.6\n",
+ "2024-03-03 12:46:00.811721: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
+ "2024-03-03 12:46:01.131524: I ./tensorflow/compiler/jit/device_compiler.h:180] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "45/45 [==============================] - 9s 26ms/step - loss: 0.5800 - sparse_categorical_accuracy: 0.6917 - val_loss: 0.7506 - val_sparse_categorical_accuracy: 0.4611 - lr: 0.0010\n",
- "Epoch 2/50\n",
- "45/45 [==============================] - 1s 14ms/step - loss: 0.4828 - sparse_categorical_accuracy: 0.7667 - val_loss: 0.8486 - val_sparse_categorical_accuracy: 0.4611 - lr: 0.0010\n",
- "Epoch 3/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.4419 - sparse_categorical_accuracy: 0.7816 - val_loss: 0.8419 - val_sparse_categorical_accuracy: 0.4611 - lr: 0.0010\n",
- "Epoch 4/50\n",
- "45/45 [==============================] - 1s 14ms/step - loss: 0.4169 - sparse_categorical_accuracy: 0.7958 - val_loss: 0.8307 - val_sparse_categorical_accuracy: 0.4611 - lr: 0.0010\n",
- "Epoch 5/50\n",
- "45/45 [==============================] - 1s 14ms/step - loss: 0.4219 - sparse_categorical_accuracy: 0.7934 - val_loss: 0.8144 - val_sparse_categorical_accuracy: 0.4611 - lr: 0.0010\n",
- "Epoch 6/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3997 - sparse_categorical_accuracy: 0.8066 - val_loss: 0.8044 - val_sparse_categorical_accuracy: 0.4611 - lr: 0.0010\n",
- "Epoch 7/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.3983 - sparse_categorical_accuracy: 0.8003 - val_loss: 0.7577 - val_sparse_categorical_accuracy: 0.5319 - lr: 0.0010\n",
- "Epoch 8/50\n",
- "45/45 [==============================] - 1s 17ms/step - loss: 0.4102 - sparse_categorical_accuracy: 0.7951 - val_loss: 0.7298 - val_sparse_categorical_accuracy: 0.5208 - lr: 0.0010\n",
- "Epoch 9/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3986 - sparse_categorical_accuracy: 0.8000 - val_loss: 0.6806 - val_sparse_categorical_accuracy: 0.5500 - lr: 0.0010\n",
- "Epoch 10/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3769 - sparse_categorical_accuracy: 0.8233 - val_loss: 0.6149 - val_sparse_categorical_accuracy: 0.5750 - lr: 0.0010\n",
- "Epoch 11/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.3829 - sparse_categorical_accuracy: 0.8094 - val_loss: 0.5508 - val_sparse_categorical_accuracy: 0.7014 - lr: 0.0010\n",
- "Epoch 12/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.3748 - sparse_categorical_accuracy: 0.8201 - val_loss: 0.4605 - val_sparse_categorical_accuracy: 0.8069 - lr: 0.0010\n",
- "Epoch 13/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3618 - sparse_categorical_accuracy: 0.8278 - val_loss: 0.4394 - val_sparse_categorical_accuracy: 0.7764 - lr: 0.0010\n",
- "Epoch 14/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3580 - sparse_categorical_accuracy: 0.8323 - val_loss: 0.4879 - val_sparse_categorical_accuracy: 0.7069 - lr: 0.0010\n",
- "Epoch 15/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3497 - sparse_categorical_accuracy: 0.8417 - val_loss: 0.3961 - val_sparse_categorical_accuracy: 0.7486 - lr: 0.0010\n",
- "Epoch 16/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3506 - sparse_categorical_accuracy: 0.8375 - val_loss: 0.6908 - val_sparse_categorical_accuracy: 0.6500 - lr: 0.0010\n",
- "Epoch 17/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3347 - sparse_categorical_accuracy: 0.8465 - val_loss: 0.3565 - val_sparse_categorical_accuracy: 0.8278 - lr: 0.0010\n",
- "Epoch 18/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3310 - sparse_categorical_accuracy: 0.8559 - val_loss: 0.6947 - val_sparse_categorical_accuracy: 0.6514 - lr: 0.0010\n",
- "Epoch 19/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3318 - sparse_categorical_accuracy: 0.8542 - val_loss: 0.4005 - val_sparse_categorical_accuracy: 0.7792 - lr: 0.0010\n",
- "Epoch 20/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3321 - sparse_categorical_accuracy: 0.8562 - val_loss: 0.5324 - val_sparse_categorical_accuracy: 0.6861 - lr: 0.0010\n",
- "Epoch 21/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3242 - sparse_categorical_accuracy: 0.8594 - val_loss: 0.9910 - val_sparse_categorical_accuracy: 0.5417 - lr: 0.0010\n",
- "Epoch 22/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.3128 - sparse_categorical_accuracy: 0.8604 - val_loss: 0.3291 - val_sparse_categorical_accuracy: 0.8556 - lr: 0.0010\n",
- "Epoch 23/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.3246 - sparse_categorical_accuracy: 0.8535 - val_loss: 0.3220 - val_sparse_categorical_accuracy: 0.8486 - lr: 0.0010\n",
- "Epoch 24/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2997 - sparse_categorical_accuracy: 0.8747 - val_loss: 0.3540 - val_sparse_categorical_accuracy: 0.7931 - lr: 0.0010\n",
- "Epoch 25/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2953 - sparse_categorical_accuracy: 0.8750 - val_loss: 0.6802 - val_sparse_categorical_accuracy: 0.6903 - lr: 0.0010\n",
- "Epoch 26/50\n",
- "45/45 [==============================] - 1s 17ms/step - loss: 0.2944 - sparse_categorical_accuracy: 0.8722 - val_loss: 0.3162 - val_sparse_categorical_accuracy: 0.8667 - lr: 0.0010\n",
- "Epoch 27/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2899 - sparse_categorical_accuracy: 0.8809 - val_loss: 0.9363 - val_sparse_categorical_accuracy: 0.5583 - lr: 0.0010\n",
- "Epoch 28/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2861 - sparse_categorical_accuracy: 0.8819 - val_loss: 0.3591 - val_sparse_categorical_accuracy: 0.8153 - lr: 0.0010\n",
- "Epoch 29/50\n",
- "45/45 [==============================] - 1s 14ms/step - loss: 0.2994 - sparse_categorical_accuracy: 0.8698 - val_loss: 0.3701 - val_sparse_categorical_accuracy: 0.8125 - lr: 0.0010\n",
- "Epoch 30/50\n",
- "45/45 [==============================] - 1s 14ms/step - loss: 0.2891 - sparse_categorical_accuracy: 0.8792 - val_loss: 0.3166 - val_sparse_categorical_accuracy: 0.8472 - lr: 0.0010\n",
- "Epoch 31/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2860 - sparse_categorical_accuracy: 0.8788 - val_loss: 0.3455 - val_sparse_categorical_accuracy: 0.8653 - lr: 0.0010\n",
- "Epoch 32/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2886 - sparse_categorical_accuracy: 0.8778 - val_loss: 0.3252 - val_sparse_categorical_accuracy: 0.8472 - lr: 0.0010\n",
- "Epoch 33/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2772 - sparse_categorical_accuracy: 0.8833 - val_loss: 0.3321 - val_sparse_categorical_accuracy: 0.8583 - lr: 0.0010\n",
- "Epoch 34/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2853 - sparse_categorical_accuracy: 0.8785 - val_loss: 0.3512 - val_sparse_categorical_accuracy: 0.8153 - lr: 0.0010\n",
- "Epoch 35/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.2554 - sparse_categorical_accuracy: 0.8983 - val_loss: 0.3587 - val_sparse_categorical_accuracy: 0.8097 - lr: 0.0010\n",
- "Epoch 36/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2499 - sparse_categorical_accuracy: 0.9021 - val_loss: 0.6153 - val_sparse_categorical_accuracy: 0.6833 - lr: 0.0010\n",
- "Epoch 37/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.2621 - sparse_categorical_accuracy: 0.8938 - val_loss: 0.2777 - val_sparse_categorical_accuracy: 0.8889 - lr: 0.0010\n",
- "Epoch 38/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2486 - sparse_categorical_accuracy: 0.8962 - val_loss: 0.3165 - val_sparse_categorical_accuracy: 0.8528 - lr: 0.0010\n",
- "Epoch 39/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.2498 - sparse_categorical_accuracy: 0.9052 - val_loss: 0.2776 - val_sparse_categorical_accuracy: 0.8708 - lr: 0.0010\n",
- "Epoch 40/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2656 - sparse_categorical_accuracy: 0.8854 - val_loss: 0.2847 - val_sparse_categorical_accuracy: 0.8847 - lr: 0.0010\n",
- "Epoch 41/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2372 - sparse_categorical_accuracy: 0.9042 - val_loss: 1.1528 - val_sparse_categorical_accuracy: 0.6417 - lr: 0.0010\n",
- "Epoch 42/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2538 - sparse_categorical_accuracy: 0.8962 - val_loss: 0.6144 - val_sparse_categorical_accuracy: 0.6611 - lr: 0.0010\n",
- "Epoch 43/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2607 - sparse_categorical_accuracy: 0.8906 - val_loss: 0.2735 - val_sparse_categorical_accuracy: 0.8722 - lr: 0.0010\n",
- "Epoch 44/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2351 - sparse_categorical_accuracy: 0.9066 - val_loss: 0.3127 - val_sparse_categorical_accuracy: 0.8653 - lr: 0.0010\n",
- "Epoch 45/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2401 - sparse_categorical_accuracy: 0.9062 - val_loss: 0.7323 - val_sparse_categorical_accuracy: 0.6903 - lr: 0.0010\n",
- "Epoch 46/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2392 - sparse_categorical_accuracy: 0.9035 - val_loss: 0.3101 - val_sparse_categorical_accuracy: 0.8500 - lr: 0.0010\n",
- "Epoch 47/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2236 - sparse_categorical_accuracy: 0.9122 - val_loss: 0.3994 - val_sparse_categorical_accuracy: 0.7972 - lr: 0.0010\n",
- "Epoch 48/50\n",
- "45/45 [==============================] - 1s 15ms/step - loss: 0.2369 - sparse_categorical_accuracy: 0.9101 - val_loss: 0.4191 - val_sparse_categorical_accuracy: 0.7583 - lr: 0.0010\n",
- "Epoch 49/50\n",
- "45/45 [==============================] - 1s 14ms/step - loss: 0.2428 - sparse_categorical_accuracy: 0.8993 - val_loss: 0.4483 - val_sparse_categorical_accuracy: 0.7972 - lr: 0.0010\n",
- "Epoch 50/50\n",
- "45/45 [==============================] - 1s 16ms/step - loss: 0.2357 - sparse_categorical_accuracy: 0.8990 - val_loss: 0.4108 - val_sparse_categorical_accuracy: 0.7917 - lr: 0.0010\n"
+ "90/90 [==============================] - 10s 15ms/step - loss: 0.5378 - sparse_categorical_accuracy: 0.7212 - val_loss: 0.7485 - val_sparse_categorical_accuracy: 0.4597 - lr: 0.0010\n",
+ "Epoch 2/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.4427 - sparse_categorical_accuracy: 0.7847 - val_loss: 0.7201 - val_sparse_categorical_accuracy: 0.4597 - lr: 0.0010\n",
+ "Epoch 3/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.4536 - sparse_categorical_accuracy: 0.7726 - val_loss: 0.7478 - val_sparse_categorical_accuracy: 0.4639 - lr: 0.0010\n",
+ "Epoch 4/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.4106 - sparse_categorical_accuracy: 0.7993 - val_loss: 0.5878 - val_sparse_categorical_accuracy: 0.7875 - lr: 0.0010\n",
+ "Epoch 5/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.4022 - sparse_categorical_accuracy: 0.7976 - val_loss: 0.5151 - val_sparse_categorical_accuracy: 0.7625 - lr: 0.0010\n",
+ "Epoch 6/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.4021 - sparse_categorical_accuracy: 0.7979 - val_loss: 0.4827 - val_sparse_categorical_accuracy: 0.6847 - lr: 0.0010\n",
+ "Epoch 7/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3992 - sparse_categorical_accuracy: 0.8000 - val_loss: 0.5255 - val_sparse_categorical_accuracy: 0.7167 - lr: 0.0010\n",
+ "Epoch 8/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3810 - sparse_categorical_accuracy: 0.8122 - val_loss: 0.4677 - val_sparse_categorical_accuracy: 0.7625 - lr: 0.0010\n",
+ "Epoch 9/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3751 - sparse_categorical_accuracy: 0.8260 - val_loss: 0.4522 - val_sparse_categorical_accuracy: 0.7208 - lr: 0.0010\n",
+ "Epoch 10/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3660 - sparse_categorical_accuracy: 0.8281 - val_loss: 0.3971 - val_sparse_categorical_accuracy: 0.7625 - lr: 0.0010\n",
+ "Epoch 11/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3702 - sparse_categorical_accuracy: 0.8198 - val_loss: 0.3700 - val_sparse_categorical_accuracy: 0.8250 - lr: 0.0010\n",
+ "Epoch 12/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3579 - sparse_categorical_accuracy: 0.8358 - val_loss: 0.3669 - val_sparse_categorical_accuracy: 0.8153 - lr: 0.0010\n",
+ "Epoch 13/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.3591 - sparse_categorical_accuracy: 0.8316 - val_loss: 0.4237 - val_sparse_categorical_accuracy: 0.7903 - lr: 0.0010\n",
+ "Epoch 14/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3567 - sparse_categorical_accuracy: 0.8382 - val_loss: 0.3661 - val_sparse_categorical_accuracy: 0.8250 - lr: 0.0010\n",
+ "Epoch 15/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3438 - sparse_categorical_accuracy: 0.8438 - val_loss: 0.4456 - val_sparse_categorical_accuracy: 0.7292 - lr: 0.0010\n",
+ "Epoch 16/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3358 - sparse_categorical_accuracy: 0.8521 - val_loss: 0.7633 - val_sparse_categorical_accuracy: 0.6556 - lr: 0.0010\n",
+ "Epoch 17/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3280 - sparse_categorical_accuracy: 0.8604 - val_loss: 0.4094 - val_sparse_categorical_accuracy: 0.7375 - lr: 0.0010\n",
+ "Epoch 18/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.3208 - sparse_categorical_accuracy: 0.8486 - val_loss: 0.3508 - val_sparse_categorical_accuracy: 0.8375 - lr: 0.0010\n",
+ "Epoch 19/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3200 - sparse_categorical_accuracy: 0.8580 - val_loss: 0.3833 - val_sparse_categorical_accuracy: 0.7597 - lr: 0.0010\n",
+ "Epoch 20/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3046 - sparse_categorical_accuracy: 0.8747 - val_loss: 0.3623 - val_sparse_categorical_accuracy: 0.8306 - lr: 0.0010\n",
+ "Epoch 21/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3128 - sparse_categorical_accuracy: 0.8656 - val_loss: 0.4997 - val_sparse_categorical_accuracy: 0.7208 - lr: 0.0010\n",
+ "Epoch 22/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3009 - sparse_categorical_accuracy: 0.8656 - val_loss: 0.3019 - val_sparse_categorical_accuracy: 0.8681 - lr: 0.0010\n",
+ "Epoch 23/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.3078 - sparse_categorical_accuracy: 0.8670 - val_loss: 0.3039 - val_sparse_categorical_accuracy: 0.8653 - lr: 0.0010\n",
+ "Epoch 24/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2999 - sparse_categorical_accuracy: 0.8719 - val_loss: 0.3097 - val_sparse_categorical_accuracy: 0.8681 - lr: 0.0010\n",
+ "Epoch 25/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2926 - sparse_categorical_accuracy: 0.8753 - val_loss: 1.1677 - val_sparse_categorical_accuracy: 0.5931 - lr: 0.0010\n",
+ "Epoch 26/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2893 - sparse_categorical_accuracy: 0.8764 - val_loss: 0.5820 - val_sparse_categorical_accuracy: 0.6819 - lr: 0.0010\n",
+ "Epoch 27/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2742 - sparse_categorical_accuracy: 0.8844 - val_loss: 0.4761 - val_sparse_categorical_accuracy: 0.7681 - lr: 0.0010\n",
+ "Epoch 28/500\n",
+ "90/90 [==============================] - 1s 8ms/step - loss: 0.2775 - sparse_categorical_accuracy: 0.8819 - val_loss: 0.9245 - val_sparse_categorical_accuracy: 0.6583 - lr: 0.0010\n",
+ "Epoch 29/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2734 - sparse_categorical_accuracy: 0.8931 - val_loss: 0.3106 - val_sparse_categorical_accuracy: 0.8528 - lr: 0.0010\n",
+ "Epoch 30/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2892 - sparse_categorical_accuracy: 0.8771 - val_loss: 0.5805 - val_sparse_categorical_accuracy: 0.7389 - lr: 0.0010\n",
+ "Epoch 31/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2937 - sparse_categorical_accuracy: 0.8757 - val_loss: 0.3179 - val_sparse_categorical_accuracy: 0.8514 - lr: 0.0010\n",
+ "Epoch 32/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2716 - sparse_categorical_accuracy: 0.8906 - val_loss: 0.2595 - val_sparse_categorical_accuracy: 0.8917 - lr: 0.0010\n",
+ "Epoch 33/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2627 - sparse_categorical_accuracy: 0.8910 - val_loss: 0.3640 - val_sparse_categorical_accuracy: 0.8181 - lr: 0.0010\n",
+ "Epoch 34/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2530 - sparse_categorical_accuracy: 0.8965 - val_loss: 0.4086 - val_sparse_categorical_accuracy: 0.7500 - lr: 0.0010\n",
+ "Epoch 35/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2602 - sparse_categorical_accuracy: 0.8917 - val_loss: 0.2517 - val_sparse_categorical_accuracy: 0.8833 - lr: 0.0010\n",
+ "Epoch 36/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2526 - sparse_categorical_accuracy: 0.9017 - val_loss: 0.2885 - val_sparse_categorical_accuracy: 0.8833 - lr: 0.0010\n",
+ "Epoch 37/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2538 - sparse_categorical_accuracy: 0.8976 - val_loss: 0.3568 - val_sparse_categorical_accuracy: 0.8083 - lr: 0.0010\n",
+ "Epoch 38/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2520 - sparse_categorical_accuracy: 0.9010 - val_loss: 0.2582 - val_sparse_categorical_accuracy: 0.8889 - lr: 0.0010\n",
+ "Epoch 39/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2544 - sparse_categorical_accuracy: 0.8997 - val_loss: 0.2351 - val_sparse_categorical_accuracy: 0.9181 - lr: 0.0010\n",
+ "Epoch 40/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2476 - sparse_categorical_accuracy: 0.8983 - val_loss: 0.2353 - val_sparse_categorical_accuracy: 0.9153 - lr: 0.0010\n",
+ "Epoch 41/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2431 - sparse_categorical_accuracy: 0.8993 - val_loss: 0.2571 - val_sparse_categorical_accuracy: 0.8833 - lr: 0.0010\n",
+ "Epoch 42/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2400 - sparse_categorical_accuracy: 0.9010 - val_loss: 0.4965 - val_sparse_categorical_accuracy: 0.7125 - lr: 0.0010\n",
+ "Epoch 43/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2429 - sparse_categorical_accuracy: 0.9010 - val_loss: 0.2455 - val_sparse_categorical_accuracy: 0.9153 - lr: 0.0010\n",
+ "Epoch 44/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2454 - sparse_categorical_accuracy: 0.8986 - val_loss: 0.2936 - val_sparse_categorical_accuracy: 0.8681 - lr: 0.0010\n",
+ "Epoch 45/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2347 - sparse_categorical_accuracy: 0.9024 - val_loss: 0.4170 - val_sparse_categorical_accuracy: 0.8125 - lr: 0.0010\n",
+ "Epoch 46/500\n",
+ "90/90 [==============================] - 1s 8ms/step - loss: 0.2359 - sparse_categorical_accuracy: 0.8979 - val_loss: 0.2453 - val_sparse_categorical_accuracy: 0.8889 - lr: 0.0010\n",
+ "Epoch 47/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2456 - sparse_categorical_accuracy: 0.9045 - val_loss: 0.5845 - val_sparse_categorical_accuracy: 0.7292 - lr: 0.0010\n",
+ "Epoch 48/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2428 - sparse_categorical_accuracy: 0.9066 - val_loss: 0.2460 - val_sparse_categorical_accuracy: 0.9014 - lr: 0.0010\n",
+ "Epoch 49/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2363 - sparse_categorical_accuracy: 0.9021 - val_loss: 0.2365 - val_sparse_categorical_accuracy: 0.8986 - lr: 0.0010\n",
+ "Epoch 50/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2301 - sparse_categorical_accuracy: 0.9045 - val_loss: 0.7744 - val_sparse_categorical_accuracy: 0.6625 - lr: 0.0010\n",
+ "Epoch 51/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2297 - sparse_categorical_accuracy: 0.9007 - val_loss: 0.6043 - val_sparse_categorical_accuracy: 0.7083 - lr: 0.0010\n",
+ "Epoch 52/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2337 - sparse_categorical_accuracy: 0.9052 - val_loss: 0.2657 - val_sparse_categorical_accuracy: 0.8722 - lr: 0.0010\n",
+ "Epoch 53/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2233 - sparse_categorical_accuracy: 0.9097 - val_loss: 0.9639 - val_sparse_categorical_accuracy: 0.6208 - lr: 0.0010\n",
+ "Epoch 54/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2190 - sparse_categorical_accuracy: 0.9187 - val_loss: 0.3675 - val_sparse_categorical_accuracy: 0.8222 - lr: 0.0010\n",
+ "Epoch 55/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2261 - sparse_categorical_accuracy: 0.9118 - val_loss: 0.4138 - val_sparse_categorical_accuracy: 0.8347 - lr: 0.0010\n",
+ "Epoch 56/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2269 - sparse_categorical_accuracy: 0.9122 - val_loss: 0.2307 - val_sparse_categorical_accuracy: 0.9111 - lr: 0.0010\n",
+ "Epoch 57/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2171 - sparse_categorical_accuracy: 0.9111 - val_loss: 0.3474 - val_sparse_categorical_accuracy: 0.8347 - lr: 0.0010\n",
+ "Epoch 58/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.2095 - sparse_categorical_accuracy: 0.9139 - val_loss: 0.1973 - val_sparse_categorical_accuracy: 0.9250 - lr: 0.0010\n",
+ "Epoch 59/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2104 - sparse_categorical_accuracy: 0.9163 - val_loss: 0.2729 - val_sparse_categorical_accuracy: 0.8639 - lr: 0.0010\n",
+ "Epoch 60/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.2196 - sparse_categorical_accuracy: 0.9128 - val_loss: 2.6805 - val_sparse_categorical_accuracy: 0.4917 - lr: 0.0010\n",
+ "Epoch 61/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1975 - sparse_categorical_accuracy: 0.9229 - val_loss: 0.2385 - val_sparse_categorical_accuracy: 0.8917 - lr: 0.0010\n",
+ "Epoch 62/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1828 - sparse_categorical_accuracy: 0.9372 - val_loss: 1.2726 - val_sparse_categorical_accuracy: 0.6611 - lr: 0.0010\n",
+ "Epoch 63/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1739 - sparse_categorical_accuracy: 0.9326 - val_loss: 0.2016 - val_sparse_categorical_accuracy: 0.9361 - lr: 0.0010\n",
+ "Epoch 64/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1553 - sparse_categorical_accuracy: 0.9427 - val_loss: 0.7312 - val_sparse_categorical_accuracy: 0.6597 - lr: 0.0010\n",
+ "Epoch 65/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1491 - sparse_categorical_accuracy: 0.9490 - val_loss: 0.2427 - val_sparse_categorical_accuracy: 0.9028 - lr: 0.0010\n",
+ "Epoch 66/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.1646 - sparse_categorical_accuracy: 0.9392 - val_loss: 0.1700 - val_sparse_categorical_accuracy: 0.9389 - lr: 0.0010\n",
+ "Epoch 67/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1350 - sparse_categorical_accuracy: 0.9521 - val_loss: 1.4531 - val_sparse_categorical_accuracy: 0.5903 - lr: 0.0010\n",
+ "Epoch 68/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.1311 - sparse_categorical_accuracy: 0.9528 - val_loss: 0.4457 - val_sparse_categorical_accuracy: 0.8097 - lr: 0.0010\n",
+ "Epoch 69/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1298 - sparse_categorical_accuracy: 0.9580 - val_loss: 0.4049 - val_sparse_categorical_accuracy: 0.7722 - lr: 0.0010\n",
+ "Epoch 70/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1348 - sparse_categorical_accuracy: 0.9552 - val_loss: 0.1891 - val_sparse_categorical_accuracy: 0.9194 - lr: 0.0010\n",
+ "Epoch 71/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1356 - sparse_categorical_accuracy: 0.9500 - val_loss: 1.6211 - val_sparse_categorical_accuracy: 0.6736 - lr: 0.0010\n",
+ "Epoch 72/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1210 - sparse_categorical_accuracy: 0.9597 - val_loss: 0.8687 - val_sparse_categorical_accuracy: 0.7028 - lr: 0.0010\n",
+ "Epoch 73/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1202 - sparse_categorical_accuracy: 0.9583 - val_loss: 0.3278 - val_sparse_categorical_accuracy: 0.8528 - lr: 0.0010\n",
+ "Epoch 74/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.1165 - sparse_categorical_accuracy: 0.9615 - val_loss: 0.2109 - val_sparse_categorical_accuracy: 0.9153 - lr: 0.0010\n",
+ "Epoch 75/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1211 - sparse_categorical_accuracy: 0.9583 - val_loss: 0.8551 - val_sparse_categorical_accuracy: 0.7139 - lr: 0.0010\n",
+ "Epoch 76/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1157 - sparse_categorical_accuracy: 0.9587 - val_loss: 2.5195 - val_sparse_categorical_accuracy: 0.6431 - lr: 0.0010\n",
+ "Epoch 77/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1190 - sparse_categorical_accuracy: 0.9587 - val_loss: 0.4082 - val_sparse_categorical_accuracy: 0.8194 - lr: 0.0010\n",
+ "Epoch 78/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1290 - sparse_categorical_accuracy: 0.9531 - val_loss: 1.7131 - val_sparse_categorical_accuracy: 0.5181 - lr: 0.0010\n",
+ "Epoch 79/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1137 - sparse_categorical_accuracy: 0.9653 - val_loss: 0.4090 - val_sparse_categorical_accuracy: 0.8375 - lr: 0.0010\n",
+ "Epoch 80/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1239 - sparse_categorical_accuracy: 0.9597 - val_loss: 0.1713 - val_sparse_categorical_accuracy: 0.9375 - lr: 0.0010\n",
+ "Epoch 81/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1200 - sparse_categorical_accuracy: 0.9639 - val_loss: 1.1018 - val_sparse_categorical_accuracy: 0.6681 - lr: 0.0010\n",
+ "Epoch 82/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1107 - sparse_categorical_accuracy: 0.9632 - val_loss: 2.4481 - val_sparse_categorical_accuracy: 0.5861 - lr: 0.0010\n",
+ "Epoch 83/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1174 - sparse_categorical_accuracy: 0.9587 - val_loss: 0.1833 - val_sparse_categorical_accuracy: 0.9222 - lr: 0.0010\n",
+ "Epoch 84/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1071 - sparse_categorical_accuracy: 0.9628 - val_loss: 0.2848 - val_sparse_categorical_accuracy: 0.8708 - lr: 0.0010\n",
+ "Epoch 85/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1093 - sparse_categorical_accuracy: 0.9622 - val_loss: 0.1623 - val_sparse_categorical_accuracy: 0.9333 - lr: 0.0010\n",
+ "Epoch 86/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1027 - sparse_categorical_accuracy: 0.9674 - val_loss: 0.1517 - val_sparse_categorical_accuracy: 0.9319 - lr: 0.0010\n",
+ "Epoch 87/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1060 - sparse_categorical_accuracy: 0.9635 - val_loss: 0.1606 - val_sparse_categorical_accuracy: 0.9500 - lr: 0.0010\n",
+ "Epoch 88/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1007 - sparse_categorical_accuracy: 0.9670 - val_loss: 0.3577 - val_sparse_categorical_accuracy: 0.8222 - lr: 0.0010\n",
+ "Epoch 89/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1001 - sparse_categorical_accuracy: 0.9667 - val_loss: 0.1152 - val_sparse_categorical_accuracy: 0.9653 - lr: 0.0010\n",
+ "Epoch 90/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1254 - sparse_categorical_accuracy: 0.9552 - val_loss: 2.5269 - val_sparse_categorical_accuracy: 0.5653 - lr: 0.0010\n",
+ "Epoch 91/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1068 - sparse_categorical_accuracy: 0.9656 - val_loss: 0.6140 - val_sparse_categorical_accuracy: 0.7500 - lr: 0.0010\n",
+ "Epoch 92/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1072 - sparse_categorical_accuracy: 0.9615 - val_loss: 0.1500 - val_sparse_categorical_accuracy: 0.9458 - lr: 0.0010\n",
+ "Epoch 93/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1153 - sparse_categorical_accuracy: 0.9618 - val_loss: 0.2745 - val_sparse_categorical_accuracy: 0.8528 - lr: 0.0010\n",
+ "Epoch 94/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1101 - sparse_categorical_accuracy: 0.9611 - val_loss: 1.5194 - val_sparse_categorical_accuracy: 0.7111 - lr: 0.0010\n",
+ "Epoch 95/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1008 - sparse_categorical_accuracy: 0.9684 - val_loss: 0.1288 - val_sparse_categorical_accuracy: 0.9569 - lr: 0.0010\n",
+ "Epoch 96/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1031 - sparse_categorical_accuracy: 0.9660 - val_loss: 0.3464 - val_sparse_categorical_accuracy: 0.8347 - lr: 0.0010\n",
+ "Epoch 97/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0909 - sparse_categorical_accuracy: 0.9698 - val_loss: 0.6550 - val_sparse_categorical_accuracy: 0.7639 - lr: 0.0010\n",
+ "Epoch 98/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0986 - sparse_categorical_accuracy: 0.9684 - val_loss: 1.1452 - val_sparse_categorical_accuracy: 0.6917 - lr: 0.0010\n",
+ "Epoch 99/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1115 - sparse_categorical_accuracy: 0.9594 - val_loss: 0.2382 - val_sparse_categorical_accuracy: 0.8931 - lr: 0.0010\n",
+ "Epoch 100/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1032 - sparse_categorical_accuracy: 0.9646 - val_loss: 0.1786 - val_sparse_categorical_accuracy: 0.9333 - lr: 0.0010\n",
+ "Epoch 101/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0953 - sparse_categorical_accuracy: 0.9653 - val_loss: 0.1808 - val_sparse_categorical_accuracy: 0.9292 - lr: 0.0010\n",
+ "Epoch 102/500\n",
+ "90/90 [==============================] - 1s 8ms/step - loss: 0.1024 - sparse_categorical_accuracy: 0.9635 - val_loss: 0.1168 - val_sparse_categorical_accuracy: 0.9569 - lr: 0.0010\n",
+ "Epoch 103/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1002 - sparse_categorical_accuracy: 0.9691 - val_loss: 1.1311 - val_sparse_categorical_accuracy: 0.6903 - lr: 0.0010\n",
+ "Epoch 104/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0972 - sparse_categorical_accuracy: 0.9674 - val_loss: 0.1573 - val_sparse_categorical_accuracy: 0.9389 - lr: 0.0010\n",
+ "Epoch 105/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1007 - sparse_categorical_accuracy: 0.9642 - val_loss: 1.1396 - val_sparse_categorical_accuracy: 0.6542 - lr: 0.0010\n",
+ "Epoch 106/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1169 - sparse_categorical_accuracy: 0.9559 - val_loss: 0.1413 - val_sparse_categorical_accuracy: 0.9500 - lr: 0.0010\n",
+ "Epoch 107/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.1087 - sparse_categorical_accuracy: 0.9628 - val_loss: 0.3754 - val_sparse_categorical_accuracy: 0.8194 - lr: 0.0010\n",
+ "Epoch 108/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0905 - sparse_categorical_accuracy: 0.9694 - val_loss: 1.2341 - val_sparse_categorical_accuracy: 0.7042 - lr: 0.0010\n",
+ "Epoch 109/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0972 - sparse_categorical_accuracy: 0.9674 - val_loss: 0.1320 - val_sparse_categorical_accuracy: 0.9500 - lr: 0.0010\n",
+ "Epoch 110/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0859 - sparse_categorical_accuracy: 0.9760 - val_loss: 0.0975 - val_sparse_categorical_accuracy: 0.9708 - lr: 5.0000e-04\n",
+ "Epoch 111/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0910 - sparse_categorical_accuracy: 0.9694 - val_loss: 0.1281 - val_sparse_categorical_accuracy: 0.9569 - lr: 5.0000e-04\n",
+ "Epoch 112/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0857 - sparse_categorical_accuracy: 0.9733 - val_loss: 0.2288 - val_sparse_categorical_accuracy: 0.9056 - lr: 5.0000e-04\n",
+ "Epoch 113/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0875 - sparse_categorical_accuracy: 0.9701 - val_loss: 0.1507 - val_sparse_categorical_accuracy: 0.9528 - lr: 5.0000e-04\n",
+ "Epoch 114/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0855 - sparse_categorical_accuracy: 0.9681 - val_loss: 0.1406 - val_sparse_categorical_accuracy: 0.9458 - lr: 5.0000e-04\n",
+ "Epoch 115/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0821 - sparse_categorical_accuracy: 0.9715 - val_loss: 0.2117 - val_sparse_categorical_accuracy: 0.9181 - lr: 5.0000e-04\n",
+ "Epoch 116/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0864 - sparse_categorical_accuracy: 0.9715 - val_loss: 0.1380 - val_sparse_categorical_accuracy: 0.9556 - lr: 5.0000e-04\n",
+ "Epoch 117/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0916 - sparse_categorical_accuracy: 0.9691 - val_loss: 0.1143 - val_sparse_categorical_accuracy: 0.9569 - lr: 5.0000e-04\n",
+ "Epoch 118/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0823 - sparse_categorical_accuracy: 0.9760 - val_loss: 0.0982 - val_sparse_categorical_accuracy: 0.9625 - lr: 5.0000e-04\n",
+ "Epoch 119/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0866 - sparse_categorical_accuracy: 0.9691 - val_loss: 0.1210 - val_sparse_categorical_accuracy: 0.9556 - lr: 5.0000e-04\n",
+ "Epoch 120/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0824 - sparse_categorical_accuracy: 0.9712 - val_loss: 0.1635 - val_sparse_categorical_accuracy: 0.9361 - lr: 5.0000e-04\n",
+ "Epoch 121/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0869 - sparse_categorical_accuracy: 0.9701 - val_loss: 0.1949 - val_sparse_categorical_accuracy: 0.9194 - lr: 5.0000e-04\n",
+ "Epoch 122/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0854 - sparse_categorical_accuracy: 0.9691 - val_loss: 0.1774 - val_sparse_categorical_accuracy: 0.9278 - lr: 5.0000e-04\n",
+ "Epoch 123/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0848 - sparse_categorical_accuracy: 0.9726 - val_loss: 0.3014 - val_sparse_categorical_accuracy: 0.8806 - lr: 5.0000e-04\n",
+ "Epoch 124/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0804 - sparse_categorical_accuracy: 0.9757 - val_loss: 0.1004 - val_sparse_categorical_accuracy: 0.9639 - lr: 5.0000e-04\n",
+ "Epoch 125/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0828 - sparse_categorical_accuracy: 0.9729 - val_loss: 0.1990 - val_sparse_categorical_accuracy: 0.9139 - lr: 5.0000e-04\n",
+ "Epoch 126/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0844 - sparse_categorical_accuracy: 0.9743 - val_loss: 0.1597 - val_sparse_categorical_accuracy: 0.9389 - lr: 5.0000e-04\n",
+ "Epoch 127/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0840 - sparse_categorical_accuracy: 0.9694 - val_loss: 0.1041 - val_sparse_categorical_accuracy: 0.9667 - lr: 5.0000e-04\n",
+ "Epoch 128/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0820 - sparse_categorical_accuracy: 0.9760 - val_loss: 0.1231 - val_sparse_categorical_accuracy: 0.9611 - lr: 5.0000e-04\n",
+ "Epoch 129/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0866 - sparse_categorical_accuracy: 0.9736 - val_loss: 0.2053 - val_sparse_categorical_accuracy: 0.9153 - lr: 5.0000e-04\n",
+ "Epoch 130/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0859 - sparse_categorical_accuracy: 0.9712 - val_loss: 0.0968 - val_sparse_categorical_accuracy: 0.9708 - lr: 5.0000e-04\n",
+ "Epoch 131/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0791 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.1068 - val_sparse_categorical_accuracy: 0.9694 - lr: 5.0000e-04\n",
+ "Epoch 132/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0730 - sparse_categorical_accuracy: 0.9792 - val_loss: 0.1100 - val_sparse_categorical_accuracy: 0.9681 - lr: 5.0000e-04\n",
+ "Epoch 133/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0847 - sparse_categorical_accuracy: 0.9740 - val_loss: 0.1603 - val_sparse_categorical_accuracy: 0.9389 - lr: 5.0000e-04\n",
+ "Epoch 134/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0804 - sparse_categorical_accuracy: 0.9729 - val_loss: 0.1108 - val_sparse_categorical_accuracy: 0.9625 - lr: 5.0000e-04\n",
+ "Epoch 135/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0847 - sparse_categorical_accuracy: 0.9712 - val_loss: 0.1002 - val_sparse_categorical_accuracy: 0.9681 - lr: 5.0000e-04\n",
+ "Epoch 136/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0845 - sparse_categorical_accuracy: 0.9736 - val_loss: 0.1072 - val_sparse_categorical_accuracy: 0.9625 - lr: 5.0000e-04\n",
+ "Epoch 137/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0765 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.1031 - val_sparse_categorical_accuracy: 0.9667 - lr: 5.0000e-04\n",
+ "Epoch 138/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0839 - sparse_categorical_accuracy: 0.9715 - val_loss: 0.1286 - val_sparse_categorical_accuracy: 0.9500 - lr: 5.0000e-04\n",
+ "Epoch 139/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0802 - sparse_categorical_accuracy: 0.9705 - val_loss: 0.0995 - val_sparse_categorical_accuracy: 0.9667 - lr: 5.0000e-04\n",
+ "Epoch 140/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0872 - sparse_categorical_accuracy: 0.9733 - val_loss: 0.1201 - val_sparse_categorical_accuracy: 0.9694 - lr: 5.0000e-04\n",
+ "Epoch 141/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0813 - sparse_categorical_accuracy: 0.9740 - val_loss: 0.1185 - val_sparse_categorical_accuracy: 0.9639 - lr: 5.0000e-04\n",
+ "Epoch 142/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0727 - sparse_categorical_accuracy: 0.9774 - val_loss: 0.1258 - val_sparse_categorical_accuracy: 0.9639 - lr: 5.0000e-04\n",
+ "Epoch 143/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0720 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.1033 - val_sparse_categorical_accuracy: 0.9653 - lr: 5.0000e-04\n",
+ "Epoch 144/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0815 - sparse_categorical_accuracy: 0.9719 - val_loss: 0.1032 - val_sparse_categorical_accuracy: 0.9653 - lr: 5.0000e-04\n",
+ "Epoch 145/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0716 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.0952 - val_sparse_categorical_accuracy: 0.9694 - lr: 5.0000e-04\n",
+ "Epoch 146/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0766 - sparse_categorical_accuracy: 0.9747 - val_loss: 0.1704 - val_sparse_categorical_accuracy: 0.9389 - lr: 5.0000e-04\n",
+ "Epoch 147/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0759 - sparse_categorical_accuracy: 0.9792 - val_loss: 0.1104 - val_sparse_categorical_accuracy: 0.9722 - lr: 5.0000e-04\n",
+ "Epoch 148/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0775 - sparse_categorical_accuracy: 0.9778 - val_loss: 0.4757 - val_sparse_categorical_accuracy: 0.8000 - lr: 5.0000e-04\n",
+ "Epoch 149/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0765 - sparse_categorical_accuracy: 0.9743 - val_loss: 0.1846 - val_sparse_categorical_accuracy: 0.9236 - lr: 5.0000e-04\n",
+ "Epoch 150/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0815 - sparse_categorical_accuracy: 0.9733 - val_loss: 0.2195 - val_sparse_categorical_accuracy: 0.9083 - lr: 5.0000e-04\n",
+ "Epoch 151/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0788 - sparse_categorical_accuracy: 0.9729 - val_loss: 0.1797 - val_sparse_categorical_accuracy: 0.9278 - lr: 5.0000e-04\n",
+ "Epoch 152/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0762 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.1142 - val_sparse_categorical_accuracy: 0.9653 - lr: 5.0000e-04\n",
+ "Epoch 153/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0730 - sparse_categorical_accuracy: 0.9747 - val_loss: 0.0997 - val_sparse_categorical_accuracy: 0.9653 - lr: 5.0000e-04\n",
+ "Epoch 154/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0755 - sparse_categorical_accuracy: 0.9743 - val_loss: 0.4295 - val_sparse_categorical_accuracy: 0.8306 - lr: 5.0000e-04\n",
+ "Epoch 155/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0705 - sparse_categorical_accuracy: 0.9785 - val_loss: 0.1023 - val_sparse_categorical_accuracy: 0.9653 - lr: 5.0000e-04\n",
+ "Epoch 156/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0761 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.5406 - val_sparse_categorical_accuracy: 0.7944 - lr: 5.0000e-04\n",
+ "Epoch 157/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0742 - sparse_categorical_accuracy: 0.9753 - val_loss: 0.1685 - val_sparse_categorical_accuracy: 0.9236 - lr: 5.0000e-04\n",
+ "Epoch 158/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0800 - sparse_categorical_accuracy: 0.9726 - val_loss: 0.2949 - val_sparse_categorical_accuracy: 0.8819 - lr: 5.0000e-04\n",
+ "Epoch 159/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0771 - sparse_categorical_accuracy: 0.9747 - val_loss: 0.1445 - val_sparse_categorical_accuracy: 0.9486 - lr: 5.0000e-04\n",
+ "Epoch 160/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0720 - sparse_categorical_accuracy: 0.9743 - val_loss: 0.3962 - val_sparse_categorical_accuracy: 0.8306 - lr: 5.0000e-04\n",
+ "Epoch 161/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0738 - sparse_categorical_accuracy: 0.9753 - val_loss: 0.1011 - val_sparse_categorical_accuracy: 0.9653 - lr: 5.0000e-04\n",
+ "Epoch 162/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0724 - sparse_categorical_accuracy: 0.9767 - val_loss: 0.1021 - val_sparse_categorical_accuracy: 0.9653 - lr: 5.0000e-04\n",
+ "Epoch 163/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0748 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.1072 - val_sparse_categorical_accuracy: 0.9597 - lr: 5.0000e-04\n",
+ "Epoch 164/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0700 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.5043 - val_sparse_categorical_accuracy: 0.8153 - lr: 5.0000e-04\n",
+ "Epoch 165/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0793 - sparse_categorical_accuracy: 0.9729 - val_loss: 0.7779 - val_sparse_categorical_accuracy: 0.7417 - lr: 5.0000e-04\n",
+ "Epoch 166/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0745 - sparse_categorical_accuracy: 0.9750 - val_loss: 0.1191 - val_sparse_categorical_accuracy: 0.9611 - lr: 2.5000e-04\n",
+ "Epoch 167/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0699 - sparse_categorical_accuracy: 0.9788 - val_loss: 0.0996 - val_sparse_categorical_accuracy: 0.9694 - lr: 2.5000e-04\n",
+ "Epoch 168/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0646 - sparse_categorical_accuracy: 0.9747 - val_loss: 0.0933 - val_sparse_categorical_accuracy: 0.9722 - lr: 2.5000e-04\n",
+ "Epoch 169/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0714 - sparse_categorical_accuracy: 0.9774 - val_loss: 0.1372 - val_sparse_categorical_accuracy: 0.9500 - lr: 2.5000e-04\n",
+ "Epoch 170/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0644 - sparse_categorical_accuracy: 0.9785 - val_loss: 0.1005 - val_sparse_categorical_accuracy: 0.9694 - lr: 2.5000e-04\n",
+ "Epoch 171/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0693 - sparse_categorical_accuracy: 0.9753 - val_loss: 0.1933 - val_sparse_categorical_accuracy: 0.9194 - lr: 2.5000e-04\n",
+ "Epoch 172/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0674 - sparse_categorical_accuracy: 0.9785 - val_loss: 0.1112 - val_sparse_categorical_accuracy: 0.9597 - lr: 2.5000e-04\n",
+ "Epoch 173/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0700 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.0966 - val_sparse_categorical_accuracy: 0.9722 - lr: 2.5000e-04\n",
+ "Epoch 174/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0672 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.1051 - val_sparse_categorical_accuracy: 0.9611 - lr: 2.5000e-04\n",
+ "Epoch 175/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0711 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.1052 - val_sparse_categorical_accuracy: 0.9653 - lr: 2.5000e-04\n",
+ "Epoch 176/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0654 - sparse_categorical_accuracy: 0.9778 - val_loss: 0.1132 - val_sparse_categorical_accuracy: 0.9639 - lr: 2.5000e-04\n",
+ "Epoch 177/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0647 - sparse_categorical_accuracy: 0.9792 - val_loss: 0.1120 - val_sparse_categorical_accuracy: 0.9597 - lr: 2.5000e-04\n",
+ "Epoch 178/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0721 - sparse_categorical_accuracy: 0.9747 - val_loss: 0.1130 - val_sparse_categorical_accuracy: 0.9625 - lr: 2.5000e-04\n",
+ "Epoch 179/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0654 - sparse_categorical_accuracy: 0.9795 - val_loss: 0.1343 - val_sparse_categorical_accuracy: 0.9556 - lr: 2.5000e-04\n",
+ "Epoch 180/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0632 - sparse_categorical_accuracy: 0.9785 - val_loss: 0.1326 - val_sparse_categorical_accuracy: 0.9528 - lr: 2.5000e-04\n",
+ "Epoch 181/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0686 - sparse_categorical_accuracy: 0.9781 - val_loss: 0.1100 - val_sparse_categorical_accuracy: 0.9625 - lr: 2.5000e-04\n",
+ "Epoch 182/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0620 - sparse_categorical_accuracy: 0.9847 - val_loss: 0.0952 - val_sparse_categorical_accuracy: 0.9694 - lr: 2.5000e-04\n",
+ "Epoch 183/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0651 - sparse_categorical_accuracy: 0.9753 - val_loss: 0.0989 - val_sparse_categorical_accuracy: 0.9681 - lr: 2.5000e-04\n",
+ "Epoch 184/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0614 - sparse_categorical_accuracy: 0.9816 - val_loss: 0.1014 - val_sparse_categorical_accuracy: 0.9597 - lr: 2.5000e-04\n",
+ "Epoch 185/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0650 - sparse_categorical_accuracy: 0.9792 - val_loss: 0.1087 - val_sparse_categorical_accuracy: 0.9639 - lr: 2.5000e-04\n",
+ "Epoch 186/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0710 - sparse_categorical_accuracy: 0.9733 - val_loss: 0.1145 - val_sparse_categorical_accuracy: 0.9556 - lr: 2.5000e-04\n",
+ "Epoch 187/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0703 - sparse_categorical_accuracy: 0.9743 - val_loss: 0.1049 - val_sparse_categorical_accuracy: 0.9597 - lr: 2.5000e-04\n",
+ "Epoch 188/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0637 - sparse_categorical_accuracy: 0.9792 - val_loss: 0.1004 - val_sparse_categorical_accuracy: 0.9736 - lr: 2.5000e-04\n",
+ "Epoch 189/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0649 - sparse_categorical_accuracy: 0.9788 - val_loss: 0.0915 - val_sparse_categorical_accuracy: 0.9736 - lr: 1.2500e-04\n",
+ "Epoch 190/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0607 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.0912 - val_sparse_categorical_accuracy: 0.9708 - lr: 1.2500e-04\n",
+ "Epoch 191/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0648 - sparse_categorical_accuracy: 0.9795 - val_loss: 0.0941 - val_sparse_categorical_accuracy: 0.9708 - lr: 1.2500e-04\n",
+ "Epoch 192/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0616 - sparse_categorical_accuracy: 0.9809 - val_loss: 0.1024 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.2500e-04\n",
+ "Epoch 193/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0615 - sparse_categorical_accuracy: 0.9823 - val_loss: 0.1049 - val_sparse_categorical_accuracy: 0.9639 - lr: 1.2500e-04\n",
+ "Epoch 194/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0662 - sparse_categorical_accuracy: 0.9792 - val_loss: 0.0940 - val_sparse_categorical_accuracy: 0.9708 - lr: 1.2500e-04\n",
+ "Epoch 195/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0607 - sparse_categorical_accuracy: 0.9802 - val_loss: 0.0940 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.2500e-04\n",
+ "Epoch 196/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0659 - sparse_categorical_accuracy: 0.9781 - val_loss: 0.0942 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.2500e-04\n",
+ "Epoch 197/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0589 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.0982 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.2500e-04\n",
+ "Epoch 198/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0627 - sparse_categorical_accuracy: 0.9809 - val_loss: 0.0935 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.2500e-04\n",
+ "Epoch 199/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0636 - sparse_categorical_accuracy: 0.9795 - val_loss: 0.1007 - val_sparse_categorical_accuracy: 0.9653 - lr: 1.2500e-04\n",
+ "Epoch 200/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0625 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.1034 - val_sparse_categorical_accuracy: 0.9611 - lr: 1.2500e-04\n",
+ "Epoch 201/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0622 - sparse_categorical_accuracy: 0.9778 - val_loss: 0.0936 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.2500e-04\n",
+ "Epoch 202/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0614 - sparse_categorical_accuracy: 0.9809 - val_loss: 0.0942 - val_sparse_categorical_accuracy: 0.9681 - lr: 1.2500e-04\n",
+ "Epoch 203/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0610 - sparse_categorical_accuracy: 0.9795 - val_loss: 0.0933 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.2500e-04\n",
+ "Epoch 204/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0586 - sparse_categorical_accuracy: 0.9802 - val_loss: 0.1058 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.2500e-04\n",
+ "Epoch 205/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0619 - sparse_categorical_accuracy: 0.9795 - val_loss: 0.1000 - val_sparse_categorical_accuracy: 0.9653 - lr: 1.2500e-04\n",
+ "Epoch 206/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0609 - sparse_categorical_accuracy: 0.9812 - val_loss: 0.0978 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.2500e-04\n",
+ "Epoch 207/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0631 - sparse_categorical_accuracy: 0.9774 - val_loss: 0.1192 - val_sparse_categorical_accuracy: 0.9569 - lr: 1.2500e-04\n",
+ "Epoch 208/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0622 - sparse_categorical_accuracy: 0.9774 - val_loss: 0.0941 - val_sparse_categorical_accuracy: 0.9653 - lr: 1.2500e-04\n",
+ "Epoch 209/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0604 - sparse_categorical_accuracy: 0.9819 - val_loss: 0.0943 - val_sparse_categorical_accuracy: 0.9736 - lr: 1.2500e-04\n",
+ "Epoch 210/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0607 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.0944 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.2500e-04\n",
+ "Epoch 211/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0549 - sparse_categorical_accuracy: 0.9844 - val_loss: 0.1044 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.0000e-04\n",
+ "Epoch 212/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0612 - sparse_categorical_accuracy: 0.9816 - val_loss: 0.0948 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.0000e-04\n",
+ "Epoch 213/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0579 - sparse_categorical_accuracy: 0.9823 - val_loss: 0.1059 - val_sparse_categorical_accuracy: 0.9653 - lr: 1.0000e-04\n",
+ "Epoch 214/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0608 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.0991 - val_sparse_categorical_accuracy: 0.9653 - lr: 1.0000e-04\n",
+ "Epoch 215/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0596 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.0917 - val_sparse_categorical_accuracy: 0.9722 - lr: 1.0000e-04\n",
+ "Epoch 216/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0610 - sparse_categorical_accuracy: 0.9809 - val_loss: 0.0954 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.0000e-04\n",
+ "Epoch 217/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0580 - sparse_categorical_accuracy: 0.9840 - val_loss: 0.0927 - val_sparse_categorical_accuracy: 0.9722 - lr: 1.0000e-04\n",
+ "Epoch 218/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0560 - sparse_categorical_accuracy: 0.9823 - val_loss: 0.0960 - val_sparse_categorical_accuracy: 0.9681 - lr: 1.0000e-04\n",
+ "Epoch 219/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0569 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.0942 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.0000e-04\n",
+ "Epoch 220/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0602 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.0933 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.0000e-04\n",
+ "Epoch 221/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0560 - sparse_categorical_accuracy: 0.9854 - val_loss: 0.0953 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.0000e-04\n",
+ "Epoch 222/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0578 - sparse_categorical_accuracy: 0.9826 - val_loss: 0.1002 - val_sparse_categorical_accuracy: 0.9708 - lr: 1.0000e-04\n",
+ "Epoch 223/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0605 - sparse_categorical_accuracy: 0.9785 - val_loss: 0.0973 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.0000e-04\n",
+ "Epoch 224/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0560 - sparse_categorical_accuracy: 0.9819 - val_loss: 0.0927 - val_sparse_categorical_accuracy: 0.9708 - lr: 1.0000e-04\n",
+ "Epoch 225/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0616 - sparse_categorical_accuracy: 0.9792 - val_loss: 0.0987 - val_sparse_categorical_accuracy: 0.9681 - lr: 1.0000e-04\n",
+ "Epoch 226/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0631 - sparse_categorical_accuracy: 0.9799 - val_loss: 0.1054 - val_sparse_categorical_accuracy: 0.9625 - lr: 1.0000e-04\n",
+ "Epoch 227/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0564 - sparse_categorical_accuracy: 0.9851 - val_loss: 0.0989 - val_sparse_categorical_accuracy: 0.9653 - lr: 1.0000e-04\n",
+ "Epoch 228/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0592 - sparse_categorical_accuracy: 0.9809 - val_loss: 0.0969 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.0000e-04\n",
+ "Epoch 229/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0596 - sparse_categorical_accuracy: 0.9816 - val_loss: 0.0926 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.0000e-04\n",
+ "Epoch 230/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0551 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.0934 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.0000e-04\n",
+ "Epoch 231/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0634 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.1122 - val_sparse_categorical_accuracy: 0.9639 - lr: 1.0000e-04\n",
+ "Epoch 232/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0608 - sparse_categorical_accuracy: 0.9812 - val_loss: 0.0929 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.0000e-04\n",
+ "Epoch 233/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0579 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.0942 - val_sparse_categorical_accuracy: 0.9722 - lr: 1.0000e-04\n",
+ "Epoch 234/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0573 - sparse_categorical_accuracy: 0.9809 - val_loss: 0.0958 - val_sparse_categorical_accuracy: 0.9694 - lr: 1.0000e-04\n",
+ "Epoch 235/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0600 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.1053 - val_sparse_categorical_accuracy: 0.9611 - lr: 1.0000e-04\n",
+ "Epoch 236/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0578 - sparse_categorical_accuracy: 0.9802 - val_loss: 0.0938 - val_sparse_categorical_accuracy: 0.9681 - lr: 1.0000e-04\n",
+ "Epoch 237/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0606 - sparse_categorical_accuracy: 0.9816 - val_loss: 0.1063 - val_sparse_categorical_accuracy: 0.9681 - lr: 1.0000e-04\n",
+ "Epoch 238/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0584 - sparse_categorical_accuracy: 0.9812 - val_loss: 0.0918 - val_sparse_categorical_accuracy: 0.9736 - lr: 1.0000e-04\n",
+ "Epoch 239/500\n",
+ "90/90 [==============================] - 1s 9ms/step - loss: 0.0575 - sparse_categorical_accuracy: 0.9823 - val_loss: 0.0955 - val_sparse_categorical_accuracy: 0.9667 - lr: 1.0000e-04\n",
+ "Epoch 240/500\n",
+ "90/90 [==============================] - 1s 10ms/step - loss: 0.0534 - sparse_categorical_accuracy: 0.9858 - val_loss: 0.0955 - val_sparse_categorical_accuracy: 0.9708 - lr: 1.0000e-04\n",
+ "Epoch 240: early stopping\n",
+ "Execution time: 3.3767328818639117 minutes \n"
]
}
],
"source": [
+ "start_time = time.time()\n",
+ "\n",
+ "\n",
"# epochs = 2\n",
"# epochs = 10\n",
- "epochs = 50\n",
+ "# epochs = 50\n",
"# epochs = 100\n",
"# epochs = 300\n",
- "# epochs = 500\n",
+ "epochs = 500 ## ORIGINAL\n",
"\n",
"# batch_size = 16\n",
- "# batch_size = 32\n",
- "batch_size = 64\n",
+ "batch_size = 32 ##ORIGINAL\n",
+ "# batch_size = 64\n",
"\n",
"callbacks = [\n",
" keras.callbacks.ModelCheckpoint(\"best_model.h5\", save_best_only=True, monitor=\"val_loss\"),\n",
" keras.callbacks.ReduceLROnPlateau(monitor=\"val_loss\", factor=0.5, patience=20, min_lr=0.0001),\n",
- " #keras.callbacks.EarlyStopping(monitor=\"val_loss\", patience=50, verbose=1),\n",
+ " keras.callbacks.EarlyStopping(monitor=\"val_loss\", patience=50, verbose=1),\n",
"]\n",
"\n",
"model.compile(\n",
@@ -697,7 +1093,15 @@
")\n",
"\n",
"# Model training APIs (methods for compile, fit, evaluate)\n",
- "# https://keras.io/api/models/model_training_apis/"
+ "# https://keras.io/api/models/model_training_apis/\n",
+ "\n",
+ "\n",
+ "\n",
+ "end_time = time.time()\n",
+ "execution_time = end_time - start_time\n",
+ "print(f\"Execution time: {execution_time/60} minutes \")\n",
+ "# Epoch 256: early stopping\n",
+ "# Execution time: 4.0526354789733885 minutes \n"
]
},
{
@@ -720,13 +1124,13 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "history.params: {'verbose': 1, 'epochs': 50, 'steps': 45}\n",
+ "history.params: {'verbose': 1, 'epochs': 500, 'steps': 90}\n",
"history.history.keys(): dict_keys(['loss', 'sparse_categorical_accuracy', 'val_loss', 'val_sparse_categorical_accuracy', 'lr'])\n"
]
},
{
"data": {
- "image/png": "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",
+ "image/png": "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",
"text/plain": [
"