neuralnetwork/dnn_complex.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/1s/tyt3m2l12dz6s028_blqch7r0000gn/T/ipykernel_86843/1971013312.py:3: DeprecationWarning: \n",
      "Pyarrow will become a required dependency of pandas in the next major release of pandas (pandas 3.0),\n",
      "(to allow more performant data types, such as the Arrow string type, and better interoperability with other libraries)\n",
      "but was not found to be installed on your system.\n",
      "If this would cause problems for you,\n",
      "please provide us feedback at https://github.com/pandas-dev/pandas/issues/54466\n",
      "        \n",
      "  import pandas as pd\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import gc\n",
    "import pandas as pd\n",
    "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import Input, Dense, BatchNormalization, Dropout\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "import sys\n",
    "sys.path.append( '../util' )\n",
    "import util as util"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_multiclass_classification_model(input_shape, num_classes):\n",
    "    inputs = Input(shape=(input_shape,))\n",
    "    x = Dense(64, activation='relu')(inputs)\n",
    "    x = Dropout(0.5)(x)\n",
    "    x = Dense(64, activation='relu')(x) \n",
    "    outputs = Dense(num_classes, activation='softmax')(x)\n",
    "    model = Model(inputs=inputs, outputs=outputs)\n",
    "    model.compile(optimizer=Adam(learning_rate=0.001),\n",
    "                  loss='sparse_categorical_crossentropy',\n",
    "                  metrics=['accuracy'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Importing datasets.....\n",
      "Processing training and test data....\n",
      "Finished processing training and test data\n",
      "Mapping labels to seven classes.....\n",
      "Data import and processing complete....\n",
      " Number of classes is:\n",
      "8\n",
      "\n",
      "Epoch 1/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m206s\u001b[0m 2ms/step - accuracy: 0.9232 - loss: 0.1789 - val_accuracy: 0.9833 - val_loss: 0.0457\n",
      "Epoch 2/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m209s\u001b[0m 2ms/step - accuracy: 0.9593 - loss: 0.0949 - val_accuracy: 0.9896 - val_loss: 0.0290\n",
      "Epoch 3/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1154s\u001b[0m 10ms/step - accuracy: 0.9675 - loss: 0.0755 - val_accuracy: 0.9897 - val_loss: 0.0272\n",
      "Epoch 4/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 2ms/step - accuracy: 0.9738 - loss: 0.0634 - val_accuracy: 0.9892 - val_loss: 0.0283\n",
      "Epoch 5/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m135s\u001b[0m 1ms/step - accuracy: 0.9739 - loss: 0.0608 - val_accuracy: 0.9902 - val_loss: 0.0262\n",
      "Epoch 6/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m183s\u001b[0m 2ms/step - accuracy: 0.9758 - loss: 0.0572 - val_accuracy: 0.9902 - val_loss: 0.0260\n",
      "Epoch 7/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m182s\u001b[0m 2ms/step - accuracy: 0.9775 - loss: 0.0537 - val_accuracy: 0.9902 - val_loss: 0.0266\n",
      "Epoch 8/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m185s\u001b[0m 2ms/step - accuracy: 0.9763 - loss: 0.0547 - val_accuracy: 0.9902 - val_loss: 0.0263\n",
      "Epoch 9/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m172s\u001b[0m 2ms/step - accuracy: 0.9783 - loss: 0.0518 - val_accuracy: 0.9902 - val_loss: 0.0338\n",
      "Epoch 10/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m139s\u001b[0m 1ms/step - accuracy: 0.9819 - loss: 0.0454 - val_accuracy: 0.9901 - val_loss: 0.0268\n",
      "Epoch 11/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m141s\u001b[0m 1ms/step - accuracy: 0.9814 - loss: 0.0455 - val_accuracy: 0.9903 - val_loss: 0.0257\n",
      "Epoch 12/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1071s\u001b[0m 9ms/step - accuracy: 0.9806 - loss: 0.0470 - val_accuracy: 0.9902 - val_loss: 0.0259\n",
      "Epoch 13/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m155s\u001b[0m 1ms/step - accuracy: 0.9820 - loss: 0.0449 - val_accuracy: 0.9904 - val_loss: 0.0261\n",
      "Epoch 14/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3593s\u001b[0m 32ms/step - accuracy: 0.9829 - loss: 0.0434 - val_accuracy: 0.9903 - val_loss: 0.0254\n",
      "Epoch 15/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3062s\u001b[0m 27ms/step - accuracy: 0.9835 - loss: 0.0427 - val_accuracy: 0.9904 - val_loss: 0.0254\n",
      "Epoch 16/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1869s\u001b[0m 16ms/step - accuracy: 0.9838 - loss: 0.0417 - val_accuracy: 0.9906 - val_loss: 0.0252\n",
      "Epoch 17/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3145s\u001b[0m 28ms/step - accuracy: 0.9806 - loss: 0.0470 - val_accuracy: 0.9903 - val_loss: 0.0257\n",
      "Epoch 18/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1972s\u001b[0m 17ms/step - accuracy: 0.9810 - loss: 0.0466 - val_accuracy: 0.9905 - val_loss: 0.0254\n",
      "Epoch 19/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1114s\u001b[0m 10ms/step - accuracy: 0.9836 - loss: 0.0411 - val_accuracy: 0.9903 - val_loss: 0.0256\n",
      "Epoch 20/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m371s\u001b[0m 3ms/step - accuracy: 0.9830 - loss: 0.0427 - val_accuracy: 0.9905 - val_loss: 0.0253\n",
      "Epoch 21/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1170s\u001b[0m 10ms/step - accuracy: 0.9838 - loss: 0.0413 - val_accuracy: 0.9903 - val_loss: 0.0257\n",
      "Epoch 22/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1265s\u001b[0m 11ms/step - accuracy: 0.9827 - loss: 0.0443 - val_accuracy: 0.9905 - val_loss: 0.0260\n",
      "Epoch 23/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2188s\u001b[0m 19ms/step - accuracy: 0.9831 - loss: 0.0429 - val_accuracy: 0.9903 - val_loss: 0.0256\n",
      "Epoch 24/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2112s\u001b[0m 19ms/step - accuracy: 0.9836 - loss: 0.0412 - val_accuracy: 0.9904 - val_loss: 0.0255\n",
      "Epoch 25/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2202s\u001b[0m 19ms/step - accuracy: 0.9840 - loss: 0.0414 - val_accuracy: 0.9907 - val_loss: 0.0249\n",
      "Epoch 26/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m689s\u001b[0m 6ms/step - accuracy: 0.9840 - loss: 0.0401 - val_accuracy: 0.9911 - val_loss: 0.0241\n",
      "Epoch 27/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1184s\u001b[0m 10ms/step - accuracy: 0.9827 - loss: 0.0417 - val_accuracy: 0.9911 - val_loss: 0.0243\n",
      "Epoch 28/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2185s\u001b[0m 19ms/step - accuracy: 0.9843 - loss: 0.0396 - val_accuracy: 0.9910 - val_loss: 0.0243\n",
      "Epoch 29/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m202s\u001b[0m 2ms/step - accuracy: 0.9829 - loss: 0.0411 - val_accuracy: 0.9909 - val_loss: 0.0241\n",
      "Epoch 30/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1113s\u001b[0m 10ms/step - accuracy: 0.9841 - loss: 0.0397 - val_accuracy: 0.9911 - val_loss: 0.0244\n",
      "Epoch 31/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1126s\u001b[0m 10ms/step - accuracy: 0.9837 - loss: 0.0406 - val_accuracy: 0.9913 - val_loss: 0.0236\n",
      "Epoch 32/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m759s\u001b[0m 7ms/step - accuracy: 0.9834 - loss: 0.0405 - val_accuracy: 0.9912 - val_loss: 0.0238\n",
      "Epoch 33/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m236s\u001b[0m 2ms/step - accuracy: 0.9843 - loss: 0.0399 - val_accuracy: 0.9908 - val_loss: 0.0248\n",
      "Epoch 34/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m241s\u001b[0m 2ms/step - accuracy: 0.9849 - loss: 0.0385 - val_accuracy: 0.9911 - val_loss: 0.0242\n",
      "Epoch 35/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m213s\u001b[0m 2ms/step - accuracy: 0.9844 - loss: 0.0391 - val_accuracy: 0.9910 - val_loss: 0.0246\n",
      "Epoch 36/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1996s\u001b[0m 18ms/step - accuracy: 0.9841 - loss: 0.0402 - val_accuracy: 0.9912 - val_loss: 0.0242\n",
      "Epoch 37/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1139s\u001b[0m 10ms/step - accuracy: 0.9858 - loss: 0.0377 - val_accuracy: 0.9906 - val_loss: 0.0252\n",
      "Epoch 38/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m989s\u001b[0m 9ms/step - accuracy: 0.9849 - loss: 0.0389 - val_accuracy: 0.9909 - val_loss: 0.0247\n",
      "Epoch 39/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m210s\u001b[0m 2ms/step - accuracy: 0.9858 - loss: 0.0380 - val_accuracy: 0.9908 - val_loss: 0.0245\n",
      "Epoch 40/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m205s\u001b[0m 2ms/step - accuracy: 0.9856 - loss: 0.0379 - val_accuracy: 0.9911 - val_loss: 0.0248\n",
      "Epoch 41/100\n",
      "\u001b[1m113583/113583\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m213s\u001b[0m 2ms/step - accuracy: 0.9843 - loss: 0.0397 - val_accuracy: 0.9911 - val_loss: 0.0241\n",
      "323131/323131 - 85s - 262us/step - accuracy: 0.9913 - loss: 0.0236\n",
      "Test accuracy: 0.9913022518157959, Test loss: 0.023623324930667877\n"
     ]
    }
   ],
   "source": [
    "train, test = util.import_dataset(7,\"dnn\")\n",
    "y_train = train[util.y_column]\n",
    "y_test = test[util.y_column]\n",
    "\n",
    "X_train = train.drop(util.y_column, axis=1)\n",
    "X_test = test.drop(util.y_column, axis=1)\n",
    "\n",
    "label_encoder = LabelEncoder()\n",
    "y_train_encoded = label_encoder.fit_transform(y_train)\n",
    "y_test_encoded = label_encoder.transform(y_test)\n",
    "\n",
    "del train,test,y_train,y_test\n",
    "num_classes = len(label_encoder.classes_)\n",
    "print(\" Number of classes is:\" )\n",
    "print(num_classes)\n",
    "print(\"\")\n",
    "model = create_multiclass_classification_model(len(util.X_columns),num_classes)\n",
    "early_stopping = EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)\n",
    "#https://keras.io/api/models/model_training_apis/\n",
    "history = model.fit(x=X_train, y=y_train_encoded,\n",
    "                    validation_split=0.2, epochs=100, \n",
    "                    batch_size=256, callbacks=[early_stopping])\n",
    "\n",
    "\n",
    "test_loss, test_acc = model.evaluate(X_test, y_test_encoded, verbose=2)\n",
    "print(f'Test accuracy: {test_acc}, Test loss: {test_loss}')\n",
    "del X_train,X_test,y_train_encoded,y_test_encoded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: 0.9913022518157959, Test loss: 0.023623324930667877\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "print(f'Test accuracy: {test_acc}, Test loss: {test_loss}')\n",
    "# Plot accuracy\n",
    "plt.plot(history.history['accuracy'], label='Training Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.title('Model Accuracy for 7 classes')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "###Change this name so you don't overrwrite the one we have now\n",
    "model.save('./dnn_model.keras')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_1\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"functional_1\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">46</span>)             │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">3,008</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)             │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">4,160</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)              │           <span style=\"color: #00af00; text-decoration-color: #00af00\">520</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer (\u001b[38;5;33mInputLayer\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m46\u001b[0m)             │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)             │         \u001b[38;5;34m3,008\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (\u001b[38;5;33mDropout\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)             │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)             │         \u001b[38;5;34m4,160\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m)              │           \u001b[38;5;34m520\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">23,066</span> (90.11 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m23,066\u001b[0m (90.11 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">7,688</span> (30.03 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m7,688\u001b[0m (30.03 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">15,378</span> (60.07 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m15,378\u001b[0m (60.07 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "None\n",
      "{'name': 'functional_1', 'trainable': True, 'layers': [{'module': 'keras.layers', 'class_name': 'InputLayer', 'config': {'batch_shape': (None, 46), 'dtype': 'float32', 'sparse': False, 'name': 'input_layer'}, 'registered_name': None, 'name': 'input_layer', 'inbound_nodes': []}, {'module': 'keras.layers', 'class_name': 'Dense', 'config': {'name': 'dense', 'trainable': True, 'dtype': 'float32', 'units': 64, 'activation': 'relu', 'use_bias': True, 'kernel_initializer': {'module': 'keras.initializers', 'class_name': 'GlorotUniform', 'config': {'seed': None}, 'registered_name': None}, 'bias_initializer': {'module': 'keras.initializers', 'class_name': 'Zeros', 'config': {}, 'registered_name': None}, 'kernel_regularizer': None, 'bias_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}, 'registered_name': None, 'build_config': {'input_shape': [None, 46]}, 'name': 'dense', 'inbound_nodes': [{'args': ({'class_name': '__keras_tensor__', 'config': {'shape': (None, 46), 'dtype': 'float32', 'keras_history': ['input_layer', 0, 0]}},), 'kwargs': {}}]}, {'module': 'keras.layers', 'class_name': 'Dropout', 'config': {'name': 'dropout', 'trainable': True, 'dtype': 'float32', 'rate': 0.5, 'seed': None, 'noise_shape': None}, 'registered_name': None, 'name': 'dropout', 'inbound_nodes': [{'args': ({'class_name': '__keras_tensor__', 'config': {'shape': (None, 64), 'dtype': 'float32', 'keras_history': ['dense', 0, 0]}},), 'kwargs': {'training': False}}]}, {'module': 'keras.layers', 'class_name': 'Dense', 'config': {'name': 'dense_1', 'trainable': True, 'dtype': 'float32', 'units': 64, 'activation': 'relu', 'use_bias': True, 'kernel_initializer': {'module': 'keras.initializers', 'class_name': 'GlorotUniform', 'config': {'seed': None}, 'registered_name': None}, 'bias_initializer': {'module': 'keras.initializers', 'class_name': 'Zeros', 'config': {}, 'registered_name': None}, 'kernel_regularizer': None, 'bias_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}, 'registered_name': None, 'build_config': {'input_shape': [None, 64]}, 'name': 'dense_1', 'inbound_nodes': [{'args': ({'class_name': '__keras_tensor__', 'config': {'shape': (None, 64), 'dtype': 'float32', 'keras_history': ['dropout', 0, 0]}},), 'kwargs': {}}]}, {'module': 'keras.layers', 'class_name': 'Dense', 'config': {'name': 'dense_2', 'trainable': True, 'dtype': 'float32', 'units': 8, 'activation': 'softmax', 'use_bias': True, 'kernel_initializer': {'module': 'keras.initializers', 'class_name': 'GlorotUniform', 'config': {'seed': None}, 'registered_name': None}, 'bias_initializer': {'module': 'keras.initializers', 'class_name': 'Zeros', 'config': {}, 'registered_name': None}, 'kernel_regularizer': None, 'bias_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}, 'registered_name': None, 'build_config': {'input_shape': [None, 64]}, 'name': 'dense_2', 'inbound_nodes': [{'args': ({'class_name': '__keras_tensor__', 'config': {'shape': (None, 64), 'dtype': 'float32', 'keras_history': ['dense_1', 0, 0]}},), 'kwargs': {}}]}], 'input_layers': [['input_layer', 0, 0]], 'output_layers': [['dense_2', 0, 0]]}\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.models import load_model\n",
    "loaded_model = load_model('./dnn_model.keras')\n",
    "print(loaded_model.summary())\n",
    "print(loaded_model.get_config())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input_layer has no weights or biases.\n",
      "dense weights shape: (46, 64)\n",
      "dense biases shape: (64,)\n",
      "dropout has no weights or biases.\n",
      "dense_1 weights shape: (64, 64)\n",
      "dense_1 biases shape: (64,)\n",
      "dense_2 weights shape: (64, 8)\n",
      "dense_2 biases shape: (8,)\n"
     ]
    }
   ],
   "source": [
    "for layer in loaded_model.layers:\n",
    "    weights = layer.get_weights()  \n",
    "    #ValueError: not enough values to unpack (expected 2, got 0) <- fixing this error, not all llayers have bias or weight\n",
    "    if len(weights) > 0:\n",
    "        print(f\"{layer.name} weights shape: {weights[0].shape}\")\n",
    "        if len(weights) > 1:\n",
    "            print(f\"{layer.name} biases shape: {weights[1].shape}\")\n",
    "    else:\n",
    "        print(f\"{layer.name} has no weights or biases.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.utils import plot_model\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "You must install pydot (`pip install pydot`) for `plot_model` to work.\n"
     ]
    }
   ],
   "source": [
    "plot_model(loaded_model, to_file='./modelpics/model_plot.png', show_shapes=True, show_layer_names=True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}