examples/time_series/cases/14.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "467482e4",
   "metadata": {},
   "source": [
    "### Example 14\n",
    "\n",
    "Refer to [/examples/time_series/generate_data.ipynb](https://github.com/serhatsoyer/py4ML/blob/main/examples/time_series/generate_data.ipynb) for the artificial time-series dataset (used in this example) generation  \n",
    "\n",
    "|Case|Not Stateful LSTM|Stateful LSTM|\n",
    "|-|-|-|\n",
    "|Example Number|[13](https://github.com/serhatsoyer/py4ML/blob/main/examples/time_series/cases/13.ipynb)|14|\n",
    "|ID|0_0_1_3_128_1_0_0_0_0|0_0_1_3_128_1_0_0_1_0|\n",
    "|Total params|17,185|17,185|\n",
    "|Total time (min)|1.1|0.9|\n",
    "|Num of epochs|34|27|\n",
    "|Time per epoch (s)|1.89|1.94|\n",
    "|Train RMSE|1.48|1.54|\n",
    "|Test RMSE|1.65|1.67|\n",
    "\n",
    "- To be able to learn inter-window relations, LSTM should be stateful\n",
    "- This is implemented but nothing changed\n",
    "\n",
    "Previous example: [/examples/time_series/cases/13.ipynb](https://github.com/serhatsoyer/py4ML/blob/main/examples/time_series/cases/13.ipynb)  \n",
    "Next example: [/examples/time_series/cases/15.ipynb](https://github.com/serhatsoyer/py4ML/blob/main/examples/time_series/cases/15.ipynb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6a748aa8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.insert(0, '../') # To be able to reach 'functions.py'\n",
    "sys.path.insert(0, '../../../') # To be able to reach 'datasets' folder\n",
    "import functions as fn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f978cd32",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Do the settings in this cell and run all\n",
    "class Train_Sets:\n",
    "    vers = 0 # 0 or 1\n",
    "    within = 0 # 0 or 1\n",
    "    inter = 1 # 0 or 1\n",
    "    numOfSens = 3 # 1 or 2 or 3\n",
    "    batchSize = 128 # (2 ** 0) or ... or (2 ** 8)\n",
    "    useHalf = True # False or True\n",
    "    dense = False # False or True\n",
    "    shuffle = False # False or True\n",
    "    stateful = True # False or True\n",
    "    resetStates = False # False or True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2fb5b2fc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ID: 0_0_1_3_128_1_0_0_1_0\n"
     ]
    }
   ],
   "source": [
    "Train_Sets, Paths = fn.init(Train_Sets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e3350324",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train shape:      (9216, 32, 3)\n",
      "y_train shape:      (9216,)\n",
      "X_val shape:        (3072, 32, 3)\n",
      "y_val shape:        (3072,)\n",
      "X_test shape:       (3072, 32, 3)\n",
      "y_test shape:       (3072,)\n",
      "Common data type:   <class 'numpy.float32'>\n"
     ]
    }
   ],
   "source": [
    "X_train, X_val, X_test, y_train, y_val, y_test = fn.split_data(Train_Sets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6ad3409e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M2\n",
      "\n",
      "systemMemory: 8.00 GB\n",
      "maxCacheSize: 2.67 GB\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-12-11 23:21:41.766071: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2022-12-11 23:21:41.766578: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/128\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-12-11 23:21:42.729626: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n",
      "2022-12-11 23:21:43.725306: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:21:44.123853: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:21:45.205212: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:21:46.803556: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:21:47.422477: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "71/72 [============================>.] - ETA: 0s - loss: 7.9803 - root_mean_squared_error: 2.8249 - mean_absolute_error: 2.1920"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-12-11 23:21:49.636151: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:21:49.706281: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:21:49.768436: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "72/72 [==============================] - 7s 35ms/step - loss: 7.9468 - root_mean_squared_error: 2.8190 - mean_absolute_error: 2.1872 - val_loss: 4.9314 - val_root_mean_squared_error: 2.2207 - val_mean_absolute_error: 1.7675\n",
      "Epoch 2/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 4.4034 - root_mean_squared_error: 2.0984 - mean_absolute_error: 1.6688 - val_loss: 4.0360 - val_root_mean_squared_error: 2.0090 - val_mean_absolute_error: 1.5954\n",
      "Epoch 3/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 3.7970 - root_mean_squared_error: 1.9486 - mean_absolute_error: 1.5551 - val_loss: 3.5851 - val_root_mean_squared_error: 1.8934 - val_mean_absolute_error: 1.5002\n",
      "Epoch 4/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 3.4638 - root_mean_squared_error: 1.8611 - mean_absolute_error: 1.4852 - val_loss: 3.3328 - val_root_mean_squared_error: 1.8256 - val_mean_absolute_error: 1.4428\n",
      "Epoch 5/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 3.2262 - root_mean_squared_error: 1.7962 - mean_absolute_error: 1.4325 - val_loss: 3.1742 - val_root_mean_squared_error: 1.7816 - val_mean_absolute_error: 1.4124\n",
      "Epoch 6/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 3.0675 - root_mean_squared_error: 1.7514 - mean_absolute_error: 1.3970 - val_loss: 3.0579 - val_root_mean_squared_error: 1.7487 - val_mean_absolute_error: 1.3884\n",
      "Epoch 7/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.9579 - root_mean_squared_error: 1.7199 - mean_absolute_error: 1.3722 - val_loss: 2.9752 - val_root_mean_squared_error: 1.7249 - val_mean_absolute_error: 1.3718\n",
      "Epoch 8/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.8712 - root_mean_squared_error: 1.6944 - mean_absolute_error: 1.3514 - val_loss: 2.9238 - val_root_mean_squared_error: 1.7099 - val_mean_absolute_error: 1.3618\n",
      "Epoch 9/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.8052 - root_mean_squared_error: 1.6749 - mean_absolute_error: 1.3358 - val_loss: 2.8849 - val_root_mean_squared_error: 1.6985 - val_mean_absolute_error: 1.3516\n",
      "Epoch 10/128\n",
      "72/72 [==============================] - 2s 23ms/step - loss: 2.7551 - root_mean_squared_error: 1.6598 - mean_absolute_error: 1.3239 - val_loss: 2.8497 - val_root_mean_squared_error: 1.6881 - val_mean_absolute_error: 1.3425\n",
      "Epoch 11/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.7104 - root_mean_squared_error: 1.6463 - mean_absolute_error: 1.3132 - val_loss: 2.8317 - val_root_mean_squared_error: 1.6828 - val_mean_absolute_error: 1.3368\n",
      "Epoch 12/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.6742 - root_mean_squared_error: 1.6353 - mean_absolute_error: 1.3048 - val_loss: 2.8117 - val_root_mean_squared_error: 1.6768 - val_mean_absolute_error: 1.3323\n",
      "Epoch 13/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.6407 - root_mean_squared_error: 1.6250 - mean_absolute_error: 1.2970 - val_loss: 2.7895 - val_root_mean_squared_error: 1.6702 - val_mean_absolute_error: 1.3282\n",
      "Epoch 14/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.6108 - root_mean_squared_error: 1.6158 - mean_absolute_error: 1.2896 - val_loss: 2.7755 - val_root_mean_squared_error: 1.6660 - val_mean_absolute_error: 1.3252\n",
      "Epoch 15/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.5770 - root_mean_squared_error: 1.6053 - mean_absolute_error: 1.2812 - val_loss: 2.7636 - val_root_mean_squared_error: 1.6624 - val_mean_absolute_error: 1.3223\n",
      "Epoch 16/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.5475 - root_mean_squared_error: 1.5961 - mean_absolute_error: 1.2740 - val_loss: 2.7500 - val_root_mean_squared_error: 1.6583 - val_mean_absolute_error: 1.3174\n",
      "Epoch 17/128\n",
      "72/72 [==============================] - 2s 25ms/step - loss: 2.5185 - root_mean_squared_error: 1.5870 - mean_absolute_error: 1.2673 - val_loss: 2.7419 - val_root_mean_squared_error: 1.6559 - val_mean_absolute_error: 1.3153\n",
      "Epoch 18/128\n",
      "72/72 [==============================] - 2s 25ms/step - loss: 2.4905 - root_mean_squared_error: 1.5781 - mean_absolute_error: 1.2604 - val_loss: 2.7388 - val_root_mean_squared_error: 1.6549 - val_mean_absolute_error: 1.3148\n",
      "Epoch 19/128\n",
      "72/72 [==============================] - 2s 25ms/step - loss: 2.4659 - root_mean_squared_error: 1.5703 - mean_absolute_error: 1.2545 - val_loss: 2.7299 - val_root_mean_squared_error: 1.6522 - val_mean_absolute_error: 1.3136\n",
      "Epoch 20/128\n",
      "72/72 [==============================] - 2s 25ms/step - loss: 2.4413 - root_mean_squared_error: 1.5625 - mean_absolute_error: 1.2483 - val_loss: 2.7337 - val_root_mean_squared_error: 1.6534 - val_mean_absolute_error: 1.3138\n",
      "Epoch 21/128\n",
      "72/72 [==============================] - 2s 25ms/step - loss: 2.4156 - root_mean_squared_error: 1.5542 - mean_absolute_error: 1.2410 - val_loss: 2.7364 - val_root_mean_squared_error: 1.6542 - val_mean_absolute_error: 1.3158\n",
      "Epoch 22/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.3950 - root_mean_squared_error: 1.5476 - mean_absolute_error: 1.2357 - val_loss: 2.7430 - val_root_mean_squared_error: 1.6562 - val_mean_absolute_error: 1.3185\n",
      "Epoch 23/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.3681 - root_mean_squared_error: 1.5389 - mean_absolute_error: 1.2293 - val_loss: 2.7529 - val_root_mean_squared_error: 1.6592 - val_mean_absolute_error: 1.3207\n",
      "Epoch 24/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.3385 - root_mean_squared_error: 1.5292 - mean_absolute_error: 1.2220 - val_loss: 2.7664 - val_root_mean_squared_error: 1.6632 - val_mean_absolute_error: 1.3229\n",
      "Epoch 25/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.3210 - root_mean_squared_error: 1.5235 - mean_absolute_error: 1.2180 - val_loss: 2.7605 - val_root_mean_squared_error: 1.6615 - val_mean_absolute_error: 1.3212\n",
      "Epoch 26/128\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.2971 - root_mean_squared_error: 1.5156 - mean_absolute_error: 1.2121 - val_loss: 2.7730 - val_root_mean_squared_error: 1.6652 - val_mean_absolute_error: 1.3239\n",
      "Epoch 27/128\n",
      "70/72 [============================>.] - ETA: 0s - loss: 2.2694 - root_mean_squared_error: 1.5065 - mean_absolute_error: 1.2048Restoring model weights from the end of the best epoch: 19.\n",
      "72/72 [==============================] - 2s 24ms/step - loss: 2.2725 - root_mean_squared_error: 1.5075 - mean_absolute_error: 1.2054 - val_loss: 2.7819 - val_root_mean_squared_error: 1.6679 - val_mean_absolute_error: 1.3243\n",
      "Epoch 27: early stopping\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as lstm_cell_layer_call_fn, lstm_cell_layer_call_and_return_conditional_losses, lstm_cell_1_layer_call_fn, lstm_cell_1_layer_call_and_return_conditional_losses while saving (showing 4 of 4). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /Users/serhatsoyer/Repos/py4ML/datasets/time_series/cases/0_0_1_3_128_1_0_0_1_0/model/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /Users/serhatsoyer/Repos/py4ML/datasets/time_series/cases/0_0_1_3_128_1_0_0_1_0/model/assets\n"
     ]
    }
   ],
   "source": [
    "model, df, train_time = fn.get_model(Train_Sets, X_train, X_val, y_train, y_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "1c649667",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"0_0_1_3_128_1_0_0_1_0\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " lstm1 (LSTM)                (128, 32, 32)             4608      \n",
      "                                                                 \n",
      " lstm2 (LSTM)                (128, 32)                 8320      \n",
      "                                                                 \n",
      " dense3 (Dense)              (128, 32)                 1056      \n",
      "                                                                 \n",
      " dense4 (Dense)              (128, 32)                 1056      \n",
      "                                                                 \n",
      " dense5 (Dense)              (128, 32)                 1056      \n",
      "                                                                 \n",
      " drop0.0625 (Dropout)        (128, 32)                 0         \n",
      "                                                                 \n",
      " dense6 (Dense)              (128, 32)                 1056      \n",
      "                                                                 \n",
      " out (Dense)                 (128, 1)                  33        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 17,185\n",
      "Trainable params: 17,185\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6a3689ef",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total training time:          0.9 min\n",
      "Num of epochs:                27\n",
      "Best val RMSE epoch:          19\n",
      "Training time per epoch:      1.94 sec\n"
     ]
    }
   ],
   "source": [
    "fn.train_info(df, train_time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9939f067",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fn.plot_learn_curve(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fc576e3a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-12-11 23:22:42.245398: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:22:42.303000: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2022-12-11 23:22:42.367263: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "y_train_pred shape: (9216,)\n",
      "y_val_pred shape:   (3072,)\n",
      "y_test_pred shape:  (3072,)\n",
      "Common data type:   <class 'numpy.float32'>\n"
     ]
    }
   ],
   "source": [
    "y_train_pred, y_val_pred, y_test_pred = fn.get_predictions(Train_Sets, model, X_train, X_val, X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8c28b462",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mse_train:     2.38\n",
      "mse_val:       2.73\n",
      "mse_test:      2.78\n",
      "rmse_train:    1.54\n",
      "rmse_val:      1.65\n",
      "rmse_test:     1.67\n",
      "mae_train:     1.23\n",
      "mae_val:       1.31\n",
      "mae_test:      1.32\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fn.evaluate_model(y_train, y_val, y_test, y_train_pred, y_val_pred, y_test_pred)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "1a17012b",
   "metadata": {},
   "source": [
    "Previous example: [/examples/time_series/cases/13.ipynb](https://github.com/serhatsoyer/py4ML/blob/main/examples/time_series/cases/13.ipynb)  \n",
    "Next example: [/examples/time_series/cases/15.ipynb](https://github.com/serhatsoyer/py4ML/blob/main/examples/time_series/cases/15.ipynb)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.6 ('study')",
   "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": "3.10.8 (main, Nov 24 2022, 08:08:27) [Clang 14.0.6 ]"
  },
  "vscode": {
   "interpreter": {
    "hash": "29d06b973d1ddb34db3279b24f9b5152402e688db937648b736e855ec4de60c0"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}