cp-speech-recognition-model-training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import os\n",
    "import sys\n",
    "\n",
    "import librosa\n",
    "import librosa.display\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "import IPython.display as ipd\n",
    "from IPython.display import Audio\n",
    "import tensorflow as tf\n",
    "import keras\n",
    "from keras.preprocessing import sequence\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Embedding\n",
    "from keras.layers import LSTM,BatchNormalization , GRU\n",
    "from tensorflow.keras.preprocessing.text import Tokenizer\n",
    "from keras.preprocessing.sequence import pad_sequences\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from keras.layers import Input, Flatten, Dropout, Activation\n",
    "from keras.layers import Conv1D, MaxPooling1D, AveragePooling1D\n",
    "from keras.models import Model\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "from tensorflow.keras.optimizers import SGD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>2367</th>\n",
       "      <th>2368</th>\n",
       "      <th>2369</th>\n",
       "      <th>2370</th>\n",
       "      <th>2371</th>\n",
       "      <th>2372</th>\n",
       "      <th>2373</th>\n",
       "      <th>2374</th>\n",
       "      <th>2375</th>\n",
       "      <th>Emotions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.237305</td>\n",
       "      <td>0.352051</td>\n",
       "      <td>0.516602</td>\n",
       "      <td>0.597168</td>\n",
       "      <td>0.557129</td>\n",
       "      <td>0.614746</td>\n",
       "      <td>0.606934</td>\n",
       "      <td>0.571289</td>\n",
       "      <td>0.576660</td>\n",
       "      <td>0.545898</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.255371</td>\n",
       "      <td>0.375977</td>\n",
       "      <td>0.493164</td>\n",
       "      <td>0.486328</td>\n",
       "      <td>0.480469</td>\n",
       "      <td>0.483887</td>\n",
       "      <td>0.488281</td>\n",
       "      <td>0.497070</td>\n",
       "      <td>0.489746</td>\n",
       "      <td>0.490723</td>\n",
       "      <td>...</td>\n",
       "      <td>2.296772</td>\n",
       "      <td>2.171422</td>\n",
       "      <td>3.164991</td>\n",
       "      <td>2.652079</td>\n",
       "      <td>4.357493</td>\n",
       "      <td>1.500388</td>\n",
       "      <td>0.353610</td>\n",
       "      <td>4.728288</td>\n",
       "      <td>9.432391</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.187988</td>\n",
       "      <td>0.274902</td>\n",
       "      <td>0.364746</td>\n",
       "      <td>0.368652</td>\n",
       "      <td>0.342285</td>\n",
       "      <td>0.320801</td>\n",
       "      <td>0.294922</td>\n",
       "      <td>0.237305</td>\n",
       "      <td>0.207520</td>\n",
       "      <td>0.207031</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.241211</td>\n",
       "      <td>0.369141</td>\n",
       "      <td>0.493652</td>\n",
       "      <td>0.494141</td>\n",
       "      <td>0.511230</td>\n",
       "      <td>0.513672</td>\n",
       "      <td>0.515137</td>\n",
       "      <td>0.519043</td>\n",
       "      <td>0.506836</td>\n",
       "      <td>0.501953</td>\n",
       "      <td>...</td>\n",
       "      <td>4.394286</td>\n",
       "      <td>3.424817</td>\n",
       "      <td>3.059559</td>\n",
       "      <td>-0.118024</td>\n",
       "      <td>-0.434105</td>\n",
       "      <td>2.880042</td>\n",
       "      <td>-0.472088</td>\n",
       "      <td>3.096140</td>\n",
       "      <td>-0.649375</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.074707</td>\n",
       "      <td>0.133789</td>\n",
       "      <td>0.202148</td>\n",
       "      <td>0.227051</td>\n",
       "      <td>0.203125</td>\n",
       "      <td>0.161621</td>\n",
       "      <td>0.118164</td>\n",
       "      <td>0.082520</td>\n",
       "      <td>0.086426</td>\n",
       "      <td>0.091797</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2377 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          0         1         2         3         4         5         6  \\\n",
       "0  0.237305  0.352051  0.516602  0.597168  0.557129  0.614746  0.606934   \n",
       "1  0.255371  0.375977  0.493164  0.486328  0.480469  0.483887  0.488281   \n",
       "2  0.187988  0.274902  0.364746  0.368652  0.342285  0.320801  0.294922   \n",
       "3  0.241211  0.369141  0.493652  0.494141  0.511230  0.513672  0.515137   \n",
       "4  0.074707  0.133789  0.202148  0.227051  0.203125  0.161621  0.118164   \n",
       "\n",
       "          7         8         9  ...      2367      2368      2369      2370  \\\n",
       "0  0.571289  0.576660  0.545898  ...  0.000000  0.000000  0.000000  0.000000   \n",
       "1  0.497070  0.489746  0.490723  ...  2.296772  2.171422  3.164991  2.652079   \n",
       "2  0.237305  0.207520  0.207031  ...  0.000000  0.000000  0.000000  0.000000   \n",
       "3  0.519043  0.506836  0.501953  ...  4.394286  3.424817  3.059559 -0.118024   \n",
       "4  0.082520  0.086426  0.091797  ...       NaN       NaN       NaN       NaN   \n",
       "\n",
       "       2371      2372      2373      2374      2375  Emotions  \n",
       "0  0.000000  0.000000  0.000000  0.000000  0.000000     happy  \n",
       "1  4.357493  1.500388  0.353610  4.728288  9.432391     happy  \n",
       "2  0.000000  0.000000  0.000000  0.000000  0.000000     happy  \n",
       "3 -0.434105  2.880042 -0.472088  3.096140 -0.649375     happy  \n",
       "4       NaN       NaN       NaN       NaN       NaN      fear  \n",
       "\n",
       "[5 rows x 2377 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Emotions = pd.read_csv('./emotion.csv')\n",
    "Emotions.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0           False\n",
      "1           False\n",
      "2           False\n",
      "3           False\n",
      "4           False\n",
      "            ...  \n",
      "2372         True\n",
      "2373         True\n",
      "2374         True\n",
      "2375         True\n",
      "Emotions    False\n",
      "Length: 2377, dtype: bool\n"
     ]
    }
   ],
   "source": [
    "print(Emotions.isna().any())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0           False\n",
      "1           False\n",
      "2           False\n",
      "3           False\n",
      "4           False\n",
      "            ...  \n",
      "2372        False\n",
      "2373        False\n",
      "2374        False\n",
      "2375        False\n",
      "Emotions    False\n",
      "Length: 2377, dtype: bool\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(4746, 2377)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Emotions=Emotions.fillna(0)\n",
    "print(Emotions.isna().any())\n",
    "Emotions.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0           0\n",
       "1           0\n",
       "2           0\n",
       "3           0\n",
       "4           0\n",
       "           ..\n",
       "2372        0\n",
       "2373        0\n",
       "2374        0\n",
       "2375        0\n",
       "Emotions    0\n",
       "Length: 2377, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(Emotions.isna())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = Emotions.iloc[: ,:-1].values\n",
    "Y = Emotions['Emotions'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "encoder = OneHotEncoder()\n",
    "Y = encoder.fit_transform(np.array(Y).reshape(-1,1)).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4746, 6)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(4746, 2376)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(Y.shape)\n",
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((3796, 2376), (3796, 6), (950, 2376), (950, 6))"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(X, Y, random_state=42,test_size=0.2, shuffle=True)\n",
    "x_train.shape, y_train.shape, x_test.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = x_train.reshape(x_train.shape[0] , x_train.shape[1] , 1)\n",
    "X_test = x_test.reshape(x_test.shape[0] , x_test.shape[1] , 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((3796, 2376), (3796, 6), (950, 2376), (950, 6))"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler = StandardScaler()\n",
    "x_train = scaler.fit_transform(x_train)\n",
    "x_test = scaler.transform(x_test)\n",
    "x_train.shape, y_train.shape, x_test.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.callbacks import ModelCheckpoint, EarlyStopping,ReduceLROnPlateau\n",
    "model_checkpoint = ModelCheckpoint(filepath='best_model1_weights.keras', monitor='val_accuracy', save_best_only=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# early_stop=EarlyStopping(monitor='val_acc',mode='auto',patience=5,restore_best_weights=True)\n",
    "lr_reduction=ReduceLROnPlateau(monitor='val_acc',patience=3,verbose=1,factor=0.5,min_lr=0.00001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((3796, 2376, 1), (3796, 6), (950, 2376, 1), (950, 6))"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_traincnn =np.expand_dims(x_train, axis=2)\n",
    "x_testcnn= np.expand_dims(x_test, axis=2)\n",
    "x_traincnn.shape, y_train.shape, x_testcnn.shape, y_test.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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: \"sequential_3\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential_3\"\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",
       "│ conv1d_15 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv1D</span>)              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2376</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">3,072</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_18          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2376</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_15 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling1D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1188</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_16 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv1D</span>)              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1188</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">1,311,232</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_19          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1188</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_16 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling1D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">594</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_9 (<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\">594</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_17 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv1D</span>)              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">594</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │       <span style=\"color: #00af00; text-decoration-color: #00af00\">655,616</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_20          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">594</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_17 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling1D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">297</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_18 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv1D</span>)              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">297</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │       <span style=\"color: #00af00; text-decoration-color: #00af00\">196,864</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_21          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">297</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_18 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling1D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">149</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_10 (<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\">149</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_19 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv1D</span>)              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">149</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)       │        <span style=\"color: #00af00; text-decoration-color: #00af00\">98,432</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_22          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">149</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)       │           <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_19 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling1D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">75</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_11 (<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\">75</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">9600</span>)           │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_6 (<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\">512</span>)            │     <span style=\"color: #00af00; text-decoration-color: #00af00\">4,915,712</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_23          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)            │         <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_7 (<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\">6</span>)              │         <span style=\"color: #00af00; text-decoration-color: #00af00\">3,078</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",
       "│ conv1d_15 (\u001b[38;5;33mConv1D\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2376\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │         \u001b[38;5;34m3,072\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_18          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2376\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │         \u001b[38;5;34m2,048\u001b[0m │\n",
       "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_15 (\u001b[38;5;33mMaxPooling1D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1188\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_16 (\u001b[38;5;33mConv1D\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1188\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │     \u001b[38;5;34m1,311,232\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_19          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1188\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │         \u001b[38;5;34m2,048\u001b[0m │\n",
       "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_16 (\u001b[38;5;33mMaxPooling1D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m594\u001b[0m, \u001b[38;5;34m512\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_9 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m594\u001b[0m, \u001b[38;5;34m512\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_17 (\u001b[38;5;33mConv1D\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m594\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │       \u001b[38;5;34m655,616\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_20          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m594\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │         \u001b[38;5;34m1,024\u001b[0m │\n",
       "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_17 (\u001b[38;5;33mMaxPooling1D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m297\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_18 (\u001b[38;5;33mConv1D\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m297\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │       \u001b[38;5;34m196,864\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_21          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m297\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │         \u001b[38;5;34m1,024\u001b[0m │\n",
       "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_18 (\u001b[38;5;33mMaxPooling1D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_10 (\u001b[38;5;33mDropout\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv1d_19 (\u001b[38;5;33mConv1D\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m128\u001b[0m)       │        \u001b[38;5;34m98,432\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_22          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m128\u001b[0m)       │           \u001b[38;5;34m512\u001b[0m │\n",
       "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling1d_19 (\u001b[38;5;33mMaxPooling1D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m75\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_11 (\u001b[38;5;33mDropout\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m75\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_3 (\u001b[38;5;33mFlatten\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m9600\u001b[0m)           │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_6 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)            │     \u001b[38;5;34m4,915,712\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_23          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)            │         \u001b[38;5;34m2,048\u001b[0m │\n",
       "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_7 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m)              │         \u001b[38;5;34m3,078\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\">7,192,710</span> (27.44 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m7,192,710\u001b[0m (27.44 MB)\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\">7,188,358</span> (27.42 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m7,188,358\u001b[0m (27.42 MB)\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\">4,352</span> (17.00 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m4,352\u001b[0m (17.00 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import tensorflow.keras.layers as L\n",
    "\n",
    "model = tf.keras.Sequential([\n",
    "    L.Conv1D(512,kernel_size=5, strides=1,padding='same', activation='relu',input_shape=(X_train.shape[1],1)),\n",
    "    L.BatchNormalization(),\n",
    "    L.MaxPool1D(pool_size=5,strides=2,padding='same'),\n",
    "    \n",
    "    L.Conv1D(512,kernel_size=5,strides=1,padding='same',activation='relu'),\n",
    "    L.BatchNormalization(),\n",
    "    L.MaxPool1D(pool_size=5,strides=2,padding='same'),\n",
    "    Dropout(0.2),  # Add dropout layer after the second max pooling layer\n",
    "    \n",
    "    L.Conv1D(256,kernel_size=5,strides=1,padding='same',activation='relu'),\n",
    "    L.BatchNormalization(),\n",
    "    L.MaxPool1D(pool_size=5,strides=2,padding='same'),\n",
    "    \n",
    "    L.Conv1D(256,kernel_size=3,strides=1,padding='same',activation='relu'),\n",
    "    L.BatchNormalization(),\n",
    "    L.MaxPool1D(pool_size=5,strides=2,padding='same'),\n",
    "    Dropout(0.2),  # Add dropout layer after the fourth max pooling layer\n",
    "    \n",
    "    L.Conv1D(128,kernel_size=3,strides=1,padding='same',activation='relu'),\n",
    "    L.BatchNormalization(),\n",
    "    L.MaxPool1D(pool_size=3,strides=2,padding='same'),\n",
    "    Dropout(0.2),  # Add dropout layer after the fifth max pooling layer\n",
    "    \n",
    "    L.Flatten(),\n",
    "    L.Dense(512,activation='relu'),\n",
    "    L.BatchNormalization(),\n",
    "    L.Dense(6,activation='softmax')\n",
    "])\n",
    "model.compile(optimizer='adam',loss='categorical_crossentropy',metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5s/step - accuracy: 0.4282 - loss: 2.1329"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Asus\\AppData\\Roaming\\Python\\Python311\\site-packages\\keras\\src\\callbacks\\callback_list.py:96: UserWarning: Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: accuracy,loss,val_accuracy,val_loss,learning_rate.\n",
      "  callback.on_epoch_end(epoch, logs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m297s\u001b[0m 5s/step - accuracy: 0.4298 - loss: 2.1233 - val_accuracy: 0.4274 - val_loss: 1.8092 - learning_rate: 0.0010\n",
      "Epoch 2/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m293s\u001b[0m 5s/step - accuracy: 0.6642 - loss: 0.9735 - val_accuracy: 0.4242 - val_loss: 2.0385 - learning_rate: 0.0010\n",
      "Epoch 3/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9090s\u001b[0m 154s/step - accuracy: 0.7156 - loss: 0.7959 - val_accuracy: 0.4253 - val_loss: 1.9257 - learning_rate: 0.0010\n",
      "Epoch 4/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m290s\u001b[0m 5s/step - accuracy: 0.7712 - loss: 0.6259 - val_accuracy: 0.4263 - val_loss: 3.0327 - learning_rate: 0.0010\n",
      "Epoch 5/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m291s\u001b[0m 5s/step - accuracy: 0.8275 - loss: 0.4828 - val_accuracy: 0.4168 - val_loss: 2.1758 - learning_rate: 0.0010\n",
      "Epoch 6/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m312s\u001b[0m 5s/step - accuracy: 0.8782 - loss: 0.3602 - val_accuracy: 0.5305 - val_loss: 1.9082 - learning_rate: 0.0010\n",
      "Epoch 7/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m292s\u001b[0m 5s/step - accuracy: 0.9095 - loss: 0.2674 - val_accuracy: 0.5579 - val_loss: 1.7558 - learning_rate: 0.0010\n",
      "Epoch 8/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m292s\u001b[0m 5s/step - accuracy: 0.9241 - loss: 0.2314 - val_accuracy: 0.6284 - val_loss: 1.3278 - learning_rate: 0.0010\n",
      "Epoch 9/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m295s\u001b[0m 5s/step - accuracy: 0.9516 - loss: 0.1474 - val_accuracy: 0.7768 - val_loss: 0.8220 - learning_rate: 0.0010\n",
      "Epoch 10/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m302s\u001b[0m 5s/step - accuracy: 0.9685 - loss: 0.1059 - val_accuracy: 0.8632 - val_loss: 0.4253 - learning_rate: 0.0010\n",
      "Epoch 11/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m293s\u001b[0m 5s/step - accuracy: 0.9827 - loss: 0.0690 - val_accuracy: 0.8547 - val_loss: 0.4567 - learning_rate: 0.0010\n",
      "Epoch 12/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m307s\u001b[0m 5s/step - accuracy: 0.9941 - loss: 0.0315 - val_accuracy: 0.8979 - val_loss: 0.3752 - learning_rate: 0.0010\n",
      "Epoch 13/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m294s\u001b[0m 5s/step - accuracy: 0.9935 - loss: 0.0392 - val_accuracy: 0.8989 - val_loss: 0.3571 - learning_rate: 0.0010\n",
      "Epoch 14/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m291s\u001b[0m 5s/step - accuracy: 0.9896 - loss: 0.0379 - val_accuracy: 0.9053 - val_loss: 0.3434 - learning_rate: 0.0010\n",
      "Epoch 15/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m290s\u001b[0m 5s/step - accuracy: 0.9923 - loss: 0.0300 - val_accuracy: 0.9137 - val_loss: 0.2866 - learning_rate: 0.0010\n",
      "Epoch 16/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m290s\u001b[0m 5s/step - accuracy: 0.9945 - loss: 0.0240 - val_accuracy: 0.9295 - val_loss: 0.2533 - learning_rate: 0.0010\n",
      "Epoch 17/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m290s\u001b[0m 5s/step - accuracy: 0.9959 - loss: 0.0241 - val_accuracy: 0.9337 - val_loss: 0.2318 - learning_rate: 0.0010\n",
      "Epoch 18/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m291s\u001b[0m 5s/step - accuracy: 0.9952 - loss: 0.0243 - val_accuracy: 0.9295 - val_loss: 0.2280 - learning_rate: 0.0010\n",
      "Epoch 19/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m290s\u001b[0m 5s/step - accuracy: 0.9805 - loss: 0.0613 - val_accuracy: 0.9253 - val_loss: 0.3005 - learning_rate: 0.0010\n",
      "Epoch 20/20\n",
      "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m294s\u001b[0m 5s/step - accuracy: 0.9894 - loss: 0.0475 - val_accuracy: 0.9221 - val_loss: 0.3113 - learning_rate: 0.0010\n"
     ]
    }
   ],
   "source": [
    "history=model.fit(x_traincnn, y_train, epochs=20, validation_data=(x_testcnn, y_test), batch_size=64,callbacks=[lr_reduction,model_checkpoint])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m30/30\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 477ms/step - accuracy: 0.9160 - loss: 0.3274\n",
      "Accuracy of our model on test data :  92.21052527427673 %\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Accuracy of our model on test data : \" , model.evaluate(x_testcnn,y_test)[1]*100 , \"%\")\n",
    "\n",
    "epochs = [i for i in range(20)]\n",
    "fig , ax = plt.subplots(1,2)\n",
    "train_acc = history.history['accuracy']\n",
    "train_loss = history.history['loss']\n",
    "test_acc = history.history['val_accuracy']\n",
    "test_loss = history.history['val_loss']\n",
    "\n",
    "fig.set_size_inches(20,6)\n",
    "ax[0].plot(epochs , train_loss , label = 'Training Loss')\n",
    "ax[0].plot(epochs , test_loss , label = 'Testing Loss')\n",
    "ax[0].set_title('Training & Testing Loss')\n",
    "ax[0].legend()\n",
    "ax[0].set_xlabel(\"Epochs\")\n",
    "\n",
    "ax[1].plot(epochs , train_acc , label = 'Training Accuracy')\n",
    "ax[1].plot(epochs , test_acc , label = 'Testing Accuracy')\n",
    "ax[1].set_title('Training & Testing Accuracy')\n",
    "ax[1].legend()\n",
    "ax[1].set_xlabel(\"Epochs\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m30/30\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 484ms/step\n"
     ]
    }
   ],
   "source": [
    "# predicting on test data.\n",
    "pred_test0 = model.predict(x_testcnn)\n",
    "y_pred0 = encoder.inverse_transform(pred_test0)\n",
    "y_test0 = encoder.inverse_transform(y_test)\n",
    "\n",
    "# Check for random predictions\n",
    "df0 = pd.DataFrame(columns=['Predicted Labels', 'Actual Labels'])\n",
    "df0['Predicted Labels'] = y_pred0.flatten()\n",
    "df0['Actual Labels'] = y_test0.flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Predicted Labels</th>\n",
       "      <th>Actual Labels</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>happy</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>angry</td>\n",
       "      <td>angry</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>disgust</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>fear</td>\n",
       "      <td>angry</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>sad</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>945</th>\n",
       "      <td>happy</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>946</th>\n",
       "      <td>disgust</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>947</th>\n",
       "      <td>happy</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>948</th>\n",
       "      <td>disgust</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>949</th>\n",
       "      <td>happy</td>\n",
       "      <td>sad</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>950 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Predicted Labels Actual Labels\n",
       "0              happy         happy\n",
       "1              angry         angry\n",
       "2            disgust       disgust\n",
       "3               fear         angry\n",
       "4                sad         happy\n",
       "..               ...           ...\n",
       "945            happy         happy\n",
       "946          disgust       disgust\n",
       "947            happy         happy\n",
       "948          disgust       disgust\n",
       "949            happy           sad\n",
       "\n",
       "[950 rows x 2 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       angry       0.95      0.89      0.92       117\n",
      "     disgust       0.96      0.89      0.93       123\n",
      "        fear       0.81      0.94      0.87        87\n",
      "       happy       0.93      0.95      0.94       403\n",
      "     neutral       0.90      0.93      0.92       132\n",
      "         sad       0.93      0.86      0.89        88\n",
      "\n",
      "    accuracy                           0.92       950\n",
      "   macro avg       0.92      0.91      0.91       950\n",
      "weighted avg       0.92      0.92      0.92       950\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix,classification_report\n",
    "cm = confusion_matrix(y_test0, y_pred0)\n",
    "plt.figure(figsize = (12, 10))\n",
    "cm = pd.DataFrame(cm , index = [i for i in encoder.categories_] , columns = [i for i in encoder.categories_])\n",
    "#cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "sns.heatmap(cm, linecolor='white', cmap='Blues', linewidth=1, annot=True, fmt='.2f')\n",
    "plt.title('Confusion Matrix', size=20)\n",
    "plt.xlabel('Predicted Labels', size=14)\n",
    "plt.ylabel('Actual Labels', size=14)\n",
    "plt.show()\n",
    "print(classification_report(y_test0, y_pred0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved model to disk\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.models import Sequential, model_from_json\n",
    "model_json = model.to_json()\n",
    "with open(\"CNN_model.json\", \"w\") as json_file:\n",
    "    json_file.write(model_json)\n",
    "model.save_weights(\"CNN_model.weights.h5\")\n",
    "print(\"Saved model to disk\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded model from disk\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.models import Sequential, model_from_json\n",
    "json_file = open('./CNN_model.json', 'r')\n",
    "loaded_model_json = json_file.read()\n",
    "json_file.close()\n",
    "loaded_model = model_from_json(loaded_model_json)\n",
    "# load weights into new model\n",
    "loaded_model.load_weights(\"./best_model1_weights.keras\")\n",
    "print(\"Loaded model from disk\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m30/30\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 478ms/step - accuracy: 0.9438 - loss: 0.1997\n",
      "compile_metrics: 93.37%\n"
     ]
    }
   ],
   "source": [
    "loaded_model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "score = loaded_model.evaluate(x_testcnn,y_test)\n",
    "print(\"%s: %.2f%%\" % (loaded_model.metrics_names[1], score[1]*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}