urban_sound_classification.ipynb

{
	"cells": [
		{
			"cell_type": "code",
			"execution_count": 3,
			"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>slice_file_name</th>\n",
							"      <th>fsID</th>\n",
							"      <th>start</th>\n",
							"      <th>end</th>\n",
							"      <th>salience</th>\n",
							"      <th>fold</th>\n",
							"      <th>classID</th>\n",
							"      <th>class</th>\n",
							"    </tr>\n",
							"  </thead>\n",
							"  <tbody>\n",
							"    <tr>\n",
							"      <th>0</th>\n",
							"      <td>100032-3-0-0.wav</td>\n",
							"      <td>100032</td>\n",
							"      <td>0.0</td>\n",
							"      <td>0.317551</td>\n",
							"      <td>1</td>\n",
							"      <td>5</td>\n",
							"      <td>3</td>\n",
							"      <td>dog_bark</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>1</th>\n",
							"      <td>100263-2-0-117.wav</td>\n",
							"      <td>100263</td>\n",
							"      <td>58.5</td>\n",
							"      <td>62.500000</td>\n",
							"      <td>1</td>\n",
							"      <td>5</td>\n",
							"      <td>2</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>2</th>\n",
							"      <td>100263-2-0-121.wav</td>\n",
							"      <td>100263</td>\n",
							"      <td>60.5</td>\n",
							"      <td>64.500000</td>\n",
							"      <td>1</td>\n",
							"      <td>5</td>\n",
							"      <td>2</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>3</th>\n",
							"      <td>100263-2-0-126.wav</td>\n",
							"      <td>100263</td>\n",
							"      <td>63.0</td>\n",
							"      <td>67.000000</td>\n",
							"      <td>1</td>\n",
							"      <td>5</td>\n",
							"      <td>2</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>4</th>\n",
							"      <td>100263-2-0-137.wav</td>\n",
							"      <td>100263</td>\n",
							"      <td>68.5</td>\n",
							"      <td>72.500000</td>\n",
							"      <td>1</td>\n",
							"      <td>5</td>\n",
							"      <td>2</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"  </tbody>\n",
							"</table>\n",
							"</div>"
						],
						"text/plain": [
							"      slice_file_name    fsID  start        end  salience  fold  classID  \\\n",
							"0    100032-3-0-0.wav  100032    0.0   0.317551         1     5        3   \n",
							"1  100263-2-0-117.wav  100263   58.5  62.500000         1     5        2   \n",
							"2  100263-2-0-121.wav  100263   60.5  64.500000         1     5        2   \n",
							"3  100263-2-0-126.wav  100263   63.0  67.000000         1     5        2   \n",
							"4  100263-2-0-137.wav  100263   68.5  72.500000         1     5        2   \n",
							"\n",
							"              class  \n",
							"0          dog_bark  \n",
							"1  children_playing  \n",
							"2  children_playing  \n",
							"3  children_playing  \n",
							"4  children_playing  "
						]
					},
					"execution_count": 3,
					"metadata": {},
					"output_type": "execute_result"
				}
			],
			"source": [
				"import pandas as pd\n",
				"\n",
				"metadata=pd.read_csv(\"UrbanSound8K/metadata/UrbanSound8K.csv\")\n",
				"metadata.head()"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 4,
			"metadata": {},
			"outputs": [],
			"source": [
				"import pandas as pd\n",
				"import os\n",
				"import librosa\n",
				"import numpy as np\n",
				"import librosa.display\n",
				"\n",
				"def extract_features(file_name):\n",
				"    data,sample_rate=librosa.load(file_name)\n",
				"    mfccs = librosa.feature.mfcc(y=data,sr=sample_rate,n_mfcc=40)\n",
				"    pad_width = 174 - mfccs.shape[1]\n",
				"    mfccs = np.pad(mfccs, pad_width=((0, 0), (0, pad_width)), mode='constant')\n",
				"    return mfccs\n"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 5,
			"metadata": {},
			"outputs": [
				{
					"name": "stderr",
					"output_type": "stream",
					"text": [
						"3555it [07:59,  7.18it/s]C:\\Users\\NITHISH\\AppData\\Local\\Programs\\Python\\Python39\\lib\\site-packages\\librosa\\util\\decorators.py:88: UserWarning: n_fft=2048 is too small for input signal of length=1323\n",
						"  return f(*args, **kwargs)\n",
						"8326it [18:51, 10.21it/s]C:\\Users\\NITHISH\\AppData\\Local\\Programs\\Python\\Python39\\lib\\site-packages\\librosa\\util\\decorators.py:88: UserWarning: n_fft=2048 is too small for input signal of length=1103\n",
						"  return f(*args, **kwargs)\n",
						"8328it [18:52, 11.88it/s]C:\\Users\\NITHISH\\AppData\\Local\\Programs\\Python\\Python39\\lib\\site-packages\\librosa\\util\\decorators.py:88: UserWarning: n_fft=2048 is too small for input signal of length=1523\n",
						"  return f(*args, **kwargs)\n",
						"8732it [19:44,  7.37it/s]\n"
					]
				},
				{
					"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>feature</th>\n",
							"      <th>class_label</th>\n",
							"    </tr>\n",
							"  </thead>\n",
							"  <tbody>\n",
							"    <tr>\n",
							"      <th>0</th>\n",
							"      <td>[[-333.38168, -172.86313, -94.09414, -60.86848...</td>\n",
							"      <td>dog_bark</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>1</th>\n",
							"      <td>[[-489.11716, -451.05603, -445.70062, -440.020...</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>2</th>\n",
							"      <td>[[-494.69894, -468.22916, -476.6907, -482.8309...</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>3</th>\n",
							"      <td>[[-444.25037, -408.51923, -404.58514, -404.187...</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"    <tr>\n",
							"      <th>4</th>\n",
							"      <td>[[-470.15686, -434.04797, -438.31635, -437.627...</td>\n",
							"      <td>children_playing</td>\n",
							"    </tr>\n",
							"  </tbody>\n",
							"</table>\n",
							"</div>"
						],
						"text/plain": [
							"                                             feature       class_label\n",
							"0  [[-333.38168, -172.86313, -94.09414, -60.86848...          dog_bark\n",
							"1  [[-489.11716, -451.05603, -445.70062, -440.020...  children_playing\n",
							"2  [[-494.69894, -468.22916, -476.6907, -482.8309...  children_playing\n",
							"3  [[-444.25037, -408.51923, -404.58514, -404.187...  children_playing\n",
							"4  [[-470.15686, -434.04797, -438.31635, -437.627...  children_playing"
						]
					},
					"execution_count": 5,
					"metadata": {},
					"output_type": "execute_result"
				}
			],
			"source": [
				"from tqdm import tqdm\n",
				"audio_data=[]\n",
				"for i,row in tqdm(metadata.iterrows()):\n",
				"    file_name = f'UrbanSound8K/audio/fold{row[\"fold\"]}/{row[\"slice_file_name\"]}'\n",
				"    features=extract_features(file_name)\n",
				"    class_name=row[\"class\"]\n",
				"    audio_data.append([features,class_name])\n",
				"\n",
				"featuresdf = pd.DataFrame(audio_data, columns=['feature','class_label'])\n",
				"featuresdf.head()"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 6,
			"metadata": {},
			"outputs": [],
			"source": [
				"from sklearn.preprocessing import LabelEncoder\n",
				"from tensorflow.keras.utils import to_categorical\n",
				"\n",
				"X = np.array(featuresdf.feature.tolist())\n",
				"y = np.array(featuresdf.class_label.tolist())\n",
				"le = LabelEncoder()\n",
				"yy = to_categorical(le.fit_transform(y))"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 7,
			"metadata": {},
			"outputs": [
				{
					"name": "stdout",
					"output_type": "stream",
					"text": [
						"Stored 'x_train' (ndarray)\n",
						"Stored 'x_test' (ndarray)\n",
						"Stored 'y_train' (ndarray)\n",
						"Stored 'y_test' (ndarray)\n",
						"Stored 'yy' (ndarray)\n",
						"Stored 'le' (LabelEncoder)\n"
					]
				}
			],
			"source": [
				"from sklearn.model_selection import train_test_split \n",
				"\n",
				"x_train, x_test, y_train, y_test = train_test_split(X, yy, test_size=0.2, random_state = 42)\n",
				"%store x_train \n",
				"%store x_test \n",
				"%store y_train \n",
				"%store y_test \n",
				"%store yy \n",
				"%store le"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 8,
			"metadata": {},
			"outputs": [],
			"source": [
				"from keras.models import Sequential\n",
				"from keras.layers import Dense, Dropout, Activation, Flatten\n",
				"from keras.layers import Convolution2D, Conv2D, MaxPooling2D, GlobalAveragePooling2D\n",
				"from tensorflow.keras.optimizers import Adam\n",
				"from keras.utils import np_utils\n",
				"from sklearn import metrics \n",
				"\n",
				"num_rows = 40\n",
				"num_columns = 174\n",
				"num_channels = 1\n",
				"\n",
				"x_train = x_train.reshape(x_train.shape[0], num_rows, num_columns, num_channels)\n",
				"x_test = x_test.reshape(x_test.shape[0], num_rows, num_columns, num_channels)\n",
				"\n",
				"num_labels = yy.shape[1]\n",
				"filter_size = 2\n",
				"\n",
				"model = Sequential()\n",
				"model.add(Conv2D(filters=16, kernel_size=2, input_shape=(num_rows, num_columns, num_channels), activation='relu'))\n",
				"model.add(MaxPooling2D(pool_size=2))\n",
				"model.add(Dropout(0.2))\n",
				"\n",
				"model.add(Conv2D(filters=32, kernel_size=2, activation='relu'))\n",
				"model.add(MaxPooling2D(pool_size=2))\n",
				"model.add(Dropout(0.2))\n",
				"\n",
				"model.add(Conv2D(filters=64, kernel_size=2, activation='relu'))\n",
				"model.add(MaxPooling2D(pool_size=2))\n",
				"model.add(Dropout(0.2))\n",
				"\n",
				"model.add(Conv2D(filters=128, kernel_size=2, activation='relu'))\n",
				"model.add(MaxPooling2D(pool_size=2))\n",
				"model.add(Dropout(0.2))\n",
				"model.add(GlobalAveragePooling2D())\n",
				"\n",
				"model.add(Dense(num_labels, activation='softmax')) "
			]
		},
		{
			"cell_type": "code",
			"execution_count": 9,
			"metadata": {},
			"outputs": [
				{
					"name": "stdout",
					"output_type": "stream",
					"text": [
						"Model: \"sequential\"\n",
						"_________________________________________________________________\n",
						"Layer (type)                 Output Shape              Param #   \n",
						"=================================================================\n",
						"conv2d (Conv2D)              (None, 39, 173, 16)       80        \n",
						"_________________________________________________________________\n",
						"max_pooling2d (MaxPooling2D) (None, 19, 86, 16)        0         \n",
						"_________________________________________________________________\n",
						"dropout (Dropout)            (None, 19, 86, 16)        0         \n",
						"_________________________________________________________________\n",
						"conv2d_1 (Conv2D)            (None, 18, 85, 32)        2080      \n",
						"_________________________________________________________________\n",
						"max_pooling2d_1 (MaxPooling2 (None, 9, 42, 32)         0         \n",
						"_________________________________________________________________\n",
						"dropout_1 (Dropout)          (None, 9, 42, 32)         0         \n",
						"_________________________________________________________________\n",
						"conv2d_2 (Conv2D)            (None, 8, 41, 64)         8256      \n",
						"_________________________________________________________________\n",
						"max_pooling2d_2 (MaxPooling2 (None, 4, 20, 64)         0         \n",
						"_________________________________________________________________\n",
						"dropout_2 (Dropout)          (None, 4, 20, 64)         0         \n",
						"_________________________________________________________________\n",
						"conv2d_3 (Conv2D)            (None, 3, 19, 128)        32896     \n",
						"_________________________________________________________________\n",
						"max_pooling2d_3 (MaxPooling2 (None, 1, 9, 128)         0         \n",
						"_________________________________________________________________\n",
						"dropout_3 (Dropout)          (None, 1, 9, 128)         0         \n",
						"_________________________________________________________________\n",
						"global_average_pooling2d (Gl (None, 128)               0         \n",
						"_________________________________________________________________\n",
						"dense (Dense)                (None, 10)                1290      \n",
						"=================================================================\n",
						"Total params: 44,602\n",
						"Trainable params: 44,602\n",
						"Non-trainable params: 0\n",
						"_________________________________________________________________\n"
					]
				}
			],
			"source": [
				"model.compile(loss='categorical_crossentropy', metrics=['accuracy'], optimizer='adam') \n",
				"model.summary()"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 10,
			"metadata": {},
			"outputs": [
				{
					"name": "stdout",
					"output_type": "stream",
					"text": [
						"55/55 [==============================] - 2s 21ms/step - loss: 15.8002 - accuracy: 0.0927: 0s - loss: 15.8993 - ac\n",
						"Pre-training accuracy: 9.2730%\n"
					]
				}
			],
			"source": [
				"score = model.evaluate(x_test, y_test, verbose=1)\n",
				"accuracy = 100*score[1]\n",
				"\n",
				"print(\"Pre-training accuracy: %.4f%%\" % accuracy)"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 11,
			"metadata": {},
			"outputs": [
				{
					"name": "stdout",
					"output_type": "stream",
					"text": [
						"Epoch 1/100\n",
						"28/28 [==============================] - 25s 861ms/step - loss: 7.8874 - accuracy: 0.1462 - val_loss: 2.5033 - val_accuracy: 0.2290\n",
						"Epoch 2/100\n",
						"28/28 [==============================] - 25s 910ms/step - loss: 2.3503 - accuracy: 0.2712 - val_loss: 1.8719 - val_accuracy: 0.3589\n",
						"Epoch 3/100\n",
						"28/28 [==============================] - 24s 875ms/step - loss: 1.8422 - accuracy: 0.3752 - val_loss: 1.6606 - val_accuracy: 0.4596\n",
						"Epoch 4/100\n",
						"28/28 [==============================] - 24s 852ms/step - loss: 1.5979 - accuracy: 0.4402 - val_loss: 1.5573 - val_accuracy: 0.5009\n",
						"Epoch 5/100\n",
						"28/28 [==============================] - 25s 902ms/step - loss: 1.4534 - accuracy: 0.4862 - val_loss: 1.4635 - val_accuracy: 0.5169\n",
						"Epoch 6/100\n",
						"28/28 [==============================] - 25s 904ms/step - loss: 1.3617 - accuracy: 0.5251 - val_loss: 1.3934 - val_accuracy: 0.5392\n",
						"Epoch 7/100\n",
						"28/28 [==============================] - 25s 880ms/step - loss: 1.2907 - accuracy: 0.5477 - val_loss: 1.3396 - val_accuracy: 0.5587\n",
						"Epoch 8/100\n",
						"28/28 [==============================] - 25s 889ms/step - loss: 1.2242 - accuracy: 0.5656 - val_loss: 1.3084 - val_accuracy: 0.5718\n",
						"Epoch 9/100\n",
						"28/28 [==============================] - 25s 881ms/step - loss: 1.1813 - accuracy: 0.5904 - val_loss: 1.2569 - val_accuracy: 0.5947\n",
						"Epoch 10/100\n",
						"28/28 [==============================] - 25s 905ms/step - loss: 1.1291 - accuracy: 0.6106 - val_loss: 1.1886 - val_accuracy: 0.6119\n",
						"Epoch 11/100\n",
						"28/28 [==============================] - 25s 881ms/step - loss: 1.0902 - accuracy: 0.6296 - val_loss: 1.1712 - val_accuracy: 0.6148\n",
						"Epoch 12/100\n",
						"28/28 [==============================] - 25s 900ms/step - loss: 1.0406 - accuracy: 0.6371 - val_loss: 1.1650 - val_accuracy: 0.6131\n",
						"Epoch 13/100\n",
						"28/28 [==============================] - 25s 896ms/step - loss: 1.0064 - accuracy: 0.6518 - val_loss: 1.1128 - val_accuracy: 0.6342\n",
						"Epoch 14/100\n",
						"28/28 [==============================] - 25s 882ms/step - loss: 0.9776 - accuracy: 0.6628 - val_loss: 1.0770 - val_accuracy: 0.6606\n",
						"Epoch 15/100\n",
						"28/28 [==============================] - 25s 910ms/step - loss: 0.9406 - accuracy: 0.6730 - val_loss: 1.0384 - val_accuracy: 0.6686\n",
						"Epoch 16/100\n",
						"28/28 [==============================] - 25s 896ms/step - loss: 0.9126 - accuracy: 0.6886 - val_loss: 1.0093 - val_accuracy: 0.6840\n",
						"Epoch 17/100\n",
						"28/28 [==============================] - 25s 896ms/step - loss: 0.8870 - accuracy: 0.6963 - val_loss: 0.9914 - val_accuracy: 0.6932\n",
						"Epoch 18/100\n",
						"28/28 [==============================] - 25s 878ms/step - loss: 0.8650 - accuracy: 0.7077 - val_loss: 0.9104 - val_accuracy: 0.7224\n",
						"Epoch 19/100\n",
						"28/28 [==============================] - 25s 881ms/step - loss: 0.8478 - accuracy: 0.7077 - val_loss: 0.9443 - val_accuracy: 0.7035\n",
						"Epoch 20/100\n",
						"28/28 [==============================] - 26s 908ms/step - loss: 0.8238 - accuracy: 0.7144 - val_loss: 0.9111 - val_accuracy: 0.7127\n",
						"Epoch 21/100\n",
						"28/28 [==============================] - 24s 863ms/step - loss: 0.7995 - accuracy: 0.7294 - val_loss: 0.8957 - val_accuracy: 0.7155\n",
						"Epoch 22/100\n",
						"28/28 [==============================] - 25s 892ms/step - loss: 0.7720 - accuracy: 0.7346 - val_loss: 0.8451 - val_accuracy: 0.7459\n",
						"Epoch 23/100\n",
						"28/28 [==============================] - 24s 864ms/step - loss: 0.7640 - accuracy: 0.7387 - val_loss: 0.9043 - val_accuracy: 0.7132\n",
						"Epoch 24/100\n",
						"28/28 [==============================] - 25s 882ms/step - loss: 0.7431 - accuracy: 0.7466 - val_loss: 0.8410 - val_accuracy: 0.7481\n",
						"Epoch 25/100\n",
						"28/28 [==============================] - 24s 863ms/step - loss: 0.7261 - accuracy: 0.7513 - val_loss: 0.8288 - val_accuracy: 0.7436\n",
						"Epoch 26/100\n",
						"28/28 [==============================] - 24s 856ms/step - loss: 0.7241 - accuracy: 0.7505 - val_loss: 0.8000 - val_accuracy: 0.7499\n",
						"Epoch 27/100\n",
						"28/28 [==============================] - 25s 897ms/step - loss: 0.6835 - accuracy: 0.7638 - val_loss: 0.7815 - val_accuracy: 0.7504\n",
						"Epoch 28/100\n",
						"28/28 [==============================] - 24s 863ms/step - loss: 0.6715 - accuracy: 0.7738 - val_loss: 0.7465 - val_accuracy: 0.7699\n",
						"Epoch 29/100\n",
						"28/28 [==============================] - 25s 885ms/step - loss: 0.6542 - accuracy: 0.7775 - val_loss: 0.7656 - val_accuracy: 0.7619\n",
						"Epoch 30/100\n",
						"28/28 [==============================] - 24s 859ms/step - loss: 0.6306 - accuracy: 0.7890 - val_loss: 0.7451 - val_accuracy: 0.7653\n",
						"Epoch 31/100\n",
						"28/28 [==============================] - 25s 882ms/step - loss: 0.6268 - accuracy: 0.7875 - val_loss: 0.7175 - val_accuracy: 0.7739\n",
						"Epoch 32/100\n",
						"28/28 [==============================] - 25s 886ms/step - loss: 0.6197 - accuracy: 0.7868 - val_loss: 0.6761 - val_accuracy: 0.7853\n",
						"Epoch 33/100\n",
						"28/28 [==============================] - 24s 872ms/step - loss: 0.5874 - accuracy: 0.8053 - val_loss: 0.6388 - val_accuracy: 0.8037\n",
						"Epoch 34/100\n",
						"28/28 [==============================] - 25s 891ms/step - loss: 0.5995 - accuracy: 0.7940 - val_loss: 0.6331 - val_accuracy: 0.8094\n",
						"Epoch 35/100\n",
						"28/28 [==============================] - 25s 875ms/step - loss: 0.5723 - accuracy: 0.8033 - val_loss: 0.6284 - val_accuracy: 0.8025\n",
						"Epoch 36/100\n",
						"28/28 [==============================] - 24s 873ms/step - loss: 0.5661 - accuracy: 0.8066 - val_loss: 0.6432 - val_accuracy: 0.8025\n",
						"Epoch 37/100\n",
						"28/28 [==============================] - 25s 897ms/step - loss: 0.5486 - accuracy: 0.8106 - val_loss: 0.6623 - val_accuracy: 0.7922\n",
						"Epoch 38/100\n",
						"28/28 [==============================] - 24s 869ms/step - loss: 0.5267 - accuracy: 0.8238 - val_loss: 0.6143 - val_accuracy: 0.8111\n",
						"Epoch 39/100\n",
						"28/28 [==============================] - 25s 891ms/step - loss: 0.5376 - accuracy: 0.8219 - val_loss: 0.6019 - val_accuracy: 0.8226\n",
						"Epoch 40/100\n",
						"28/28 [==============================] - 25s 878ms/step - loss: 0.5108 - accuracy: 0.8242 - val_loss: 0.5876 - val_accuracy: 0.8208\n",
						"Epoch 41/100\n",
						"28/28 [==============================] - 25s 882ms/step - loss: 0.5188 - accuracy: 0.8222 - val_loss: 0.6141 - val_accuracy: 0.8042\n",
						"Epoch 42/100\n",
						"28/28 [==============================] - 24s 867ms/step - loss: 0.4854 - accuracy: 0.8293 - val_loss: 0.5990 - val_accuracy: 0.8071\n",
						"Epoch 43/100\n",
						"28/28 [==============================] - 24s 863ms/step - loss: 0.4949 - accuracy: 0.8322 - val_loss: 0.5850 - val_accuracy: 0.8168\n",
						"Epoch 44/100\n",
						"28/28 [==============================] - 25s 885ms/step - loss: 0.4878 - accuracy: 0.8314 - val_loss: 0.5770 - val_accuracy: 0.8157\n",
						"Epoch 45/100\n",
						"28/28 [==============================] - 24s 862ms/step - loss: 0.4810 - accuracy: 0.8342 - val_loss: 0.6136 - val_accuracy: 0.8014\n",
						"Epoch 46/100\n",
						"28/28 [==============================] - 25s 893ms/step - loss: 0.4777 - accuracy: 0.8361 - val_loss: 0.5969 - val_accuracy: 0.8111\n",
						"Epoch 47/100\n",
						"28/28 [==============================] - 24s 872ms/step - loss: 0.4644 - accuracy: 0.8405 - val_loss: 0.5243 - val_accuracy: 0.8392\n",
						"Epoch 48/100\n",
						"28/28 [==============================] - 24s 866ms/step - loss: 0.4464 - accuracy: 0.8464 - val_loss: 0.5485 - val_accuracy: 0.8283\n",
						"Epoch 49/100\n",
						"28/28 [==============================] - 25s 884ms/step - loss: 0.4456 - accuracy: 0.8481 - val_loss: 0.5400 - val_accuracy: 0.8283\n",
						"Epoch 50/100\n",
						"28/28 [==============================] - 24s 860ms/step - loss: 0.4376 - accuracy: 0.8478 - val_loss: 0.5180 - val_accuracy: 0.8374\n",
						"Epoch 51/100\n",
						"28/28 [==============================] - 25s 890ms/step - loss: 0.4202 - accuracy: 0.8577 - val_loss: 0.5322 - val_accuracy: 0.8266\n",
						"Epoch 52/100\n",
						"28/28 [==============================] - 24s 874ms/step - loss: 0.4165 - accuracy: 0.8604 - val_loss: 0.5002 - val_accuracy: 0.8432\n",
						"Epoch 53/100\n",
						"28/28 [==============================] - 26s 936ms/step - loss: 0.4191 - accuracy: 0.8586 - val_loss: 0.4904 - val_accuracy: 0.8535\n",
						"Epoch 54/100\n",
						"28/28 [==============================] - 26s 925ms/step - loss: 0.4058 - accuracy: 0.8593 - val_loss: 0.5155 - val_accuracy: 0.8334\n",
						"Epoch 55/100\n",
						"28/28 [==============================] - 24s 865ms/step - loss: 0.4132 - accuracy: 0.8573 - val_loss: 0.5176 - val_accuracy: 0.8380\n",
						"Epoch 56/100\n",
						"28/28 [==============================] - 25s 891ms/step - loss: 0.3930 - accuracy: 0.8684 - val_loss: 0.5085 - val_accuracy: 0.8426\n",
						"Epoch 57/100\n",
						"28/28 [==============================] - 24s 863ms/step - loss: 0.3839 - accuracy: 0.8671 - val_loss: 0.4786 - val_accuracy: 0.8472\n",
						"Epoch 58/100\n",
						"28/28 [==============================] - 29s 1s/step - loss: 0.3818 - accuracy: 0.8694 - val_loss: 0.4538 - val_accuracy: 0.8609\n",
						"Epoch 59/100\n",
						"28/28 [==============================] - 28s 1s/step - loss: 0.3917 - accuracy: 0.8628 - val_loss: 0.5116 - val_accuracy: 0.8369\n",
						"Epoch 60/100\n",
						"28/28 [==============================] - 27s 964ms/step - loss: 0.3744 - accuracy: 0.8727 - val_loss: 0.4733 - val_accuracy: 0.8517\n",
						"Epoch 61/100\n",
						"28/28 [==============================] - 25s 883ms/step - loss: 0.3594 - accuracy: 0.8776 - val_loss: 0.4642 - val_accuracy: 0.8586\n",
						"Epoch 62/100\n",
						"28/28 [==============================] - 24s 869ms/step - loss: 0.3584 - accuracy: 0.8772 - val_loss: 0.4452 - val_accuracy: 0.8552\n",
						"Epoch 63/100\n",
						"28/28 [==============================] - 25s 891ms/step - loss: 0.3367 - accuracy: 0.8822 - val_loss: 0.5626 - val_accuracy: 0.8140\n",
						"Epoch 64/100\n",
						"28/28 [==============================] - 27s 964ms/step - loss: 0.3507 - accuracy: 0.8779 - val_loss: 0.4718 - val_accuracy: 0.8495\n",
						"Epoch 65/100\n",
						"28/28 [==============================] - 26s 931ms/step - loss: 0.3571 - accuracy: 0.8754 - val_loss: 0.4681 - val_accuracy: 0.8414\n",
						"Epoch 66/100\n",
						"28/28 [==============================] - 22s 795ms/step - loss: 0.3342 - accuracy: 0.8860 - val_loss: 0.4717 - val_accuracy: 0.8489\n",
						"Epoch 67/100\n",
						"28/28 [==============================] - 22s 792ms/step - loss: 0.3270 - accuracy: 0.8880 - val_loss: 0.4606 - val_accuracy: 0.8575\n",
						"Epoch 68/100\n",
						"28/28 [==============================] - 22s 801ms/step - loss: 0.3376 - accuracy: 0.8827 - val_loss: 0.4187 - val_accuracy: 0.8620\n",
						"Epoch 69/100\n",
						"28/28 [==============================] - 23s 815ms/step - loss: 0.3295 - accuracy: 0.8870 - val_loss: 0.5074 - val_accuracy: 0.8317\n",
						"Epoch 70/100\n",
						"28/28 [==============================] - 23s 814ms/step - loss: 0.3169 - accuracy: 0.8898 - val_loss: 0.4924 - val_accuracy: 0.8443\n",
						"Epoch 71/100\n",
						"28/28 [==============================] - 23s 815ms/step - loss: 0.3345 - accuracy: 0.8870 - val_loss: 0.4644 - val_accuracy: 0.8540\n",
						"Epoch 72/100\n",
						"28/28 [==============================] - 23s 830ms/step - loss: 0.3169 - accuracy: 0.8915 - val_loss: 0.4473 - val_accuracy: 0.8580\n",
						"Epoch 73/100\n",
						"28/28 [==============================] - 23s 815ms/step - loss: 0.3127 - accuracy: 0.8915 - val_loss: 0.4355 - val_accuracy: 0.8626\n",
						"Epoch 74/100\n",
						"28/28 [==============================] - 25s 880ms/step - loss: 0.3224 - accuracy: 0.8865 - val_loss: 0.4150 - val_accuracy: 0.8706\n",
						"Epoch 75/100\n",
						"28/28 [==============================] - 23s 819ms/step - loss: 0.3000 - accuracy: 0.8995 - val_loss: 0.4092 - val_accuracy: 0.8643\n",
						"Epoch 76/100\n",
						"28/28 [==============================] - 26s 941ms/step - loss: 0.3035 - accuracy: 0.8898 - val_loss: 0.4204 - val_accuracy: 0.8672\n",
						"Epoch 77/100\n",
						"28/28 [==============================] - 26s 936ms/step - loss: 0.3000 - accuracy: 0.8962 - val_loss: 0.3973 - val_accuracy: 0.8729\n",
						"Epoch 78/100\n",
						"28/28 [==============================] - 30s 1s/step - loss: 0.2927 - accuracy: 0.8985 - val_loss: 0.4355 - val_accuracy: 0.8615\n",
						"Epoch 79/100\n",
						"28/28 [==============================] - 31s 1s/step - loss: 0.2934 - accuracy: 0.9021 - val_loss: 0.4045 - val_accuracy: 0.8701\n",
						"Epoch 80/100\n",
						"28/28 [==============================] - 30s 1s/step - loss: 0.2806 - accuracy: 0.9009 - val_loss: 0.4058 - val_accuracy: 0.8672\n",
						"Epoch 81/100\n",
						"28/28 [==============================] - 29s 1s/step - loss: 0.2784 - accuracy: 0.9024 - val_loss: 0.4075 - val_accuracy: 0.8643\n",
						"Epoch 82/100\n",
						"28/28 [==============================] - 29s 1s/step - loss: 0.2729 - accuracy: 0.9022 - val_loss: 0.4388 - val_accuracy: 0.8569\n",
						"Epoch 83/100\n",
						"28/28 [==============================] - 31s 1s/step - loss: 0.2652 - accuracy: 0.9082 - val_loss: 0.4084 - val_accuracy: 0.8683\n",
						"Epoch 84/100\n",
						"28/28 [==============================] - 29s 1s/step - loss: 0.2658 - accuracy: 0.9072 - val_loss: 0.3962 - val_accuracy: 0.8809\n",
						"Epoch 85/100\n",
						"28/28 [==============================] - 27s 957ms/step - loss: 0.2597 - accuracy: 0.9097 - val_loss: 0.4561 - val_accuracy: 0.8643\n",
						"Epoch 86/100\n",
						"28/28 [==============================] - 26s 946ms/step - loss: 0.2731 - accuracy: 0.9051 - val_loss: 0.3931 - val_accuracy: 0.8792\n",
						"Epoch 87/100\n",
						"28/28 [==============================] - 26s 912ms/step - loss: 0.2438 - accuracy: 0.9140 - val_loss: 0.3647 - val_accuracy: 0.8798\n",
						"Epoch 88/100\n",
						"28/28 [==============================] - 25s 881ms/step - loss: 0.2417 - accuracy: 0.9145 - val_loss: 0.4130 - val_accuracy: 0.8678\n",
						"Epoch 89/100\n",
						"28/28 [==============================] - 28s 989ms/step - loss: 0.2630 - accuracy: 0.9084 - val_loss: 0.4099 - val_accuracy: 0.8712\n",
						"Epoch 90/100\n",
						"28/28 [==============================] - 26s 917ms/step - loss: 0.2444 - accuracy: 0.9171 - val_loss: 0.4064 - val_accuracy: 0.8718\n",
						"Epoch 91/100\n",
						"28/28 [==============================] - 25s 879ms/step - loss: 0.2481 - accuracy: 0.9131 - val_loss: 0.4081 - val_accuracy: 0.8729\n",
						"Epoch 92/100\n",
						"28/28 [==============================] - 27s 952ms/step - loss: 0.2356 - accuracy: 0.9181 - val_loss: 0.3977 - val_accuracy: 0.8775\n",
						"Epoch 93/100\n",
						"28/28 [==============================] - 31s 1s/step - loss: 0.2301 - accuracy: 0.9207 - val_loss: 0.3556 - val_accuracy: 0.8878\n",
						"Epoch 94/100\n",
						"28/28 [==============================] - 23s 835ms/step - loss: 0.2238 - accuracy: 0.9211 - val_loss: 0.3671 - val_accuracy: 0.8832\n",
						"Epoch 95/100\n",
						"28/28 [==============================] - 24s 878ms/step - loss: 0.2396 - accuracy: 0.9137 - val_loss: 0.4064 - val_accuracy: 0.8666\n",
						"Epoch 96/100\n",
						"28/28 [==============================] - 27s 979ms/step - loss: 0.2388 - accuracy: 0.9165 - val_loss: 0.3640 - val_accuracy: 0.8884\n",
						"Epoch 97/100\n",
						"28/28 [==============================] - 27s 960ms/step - loss: 0.2333 - accuracy: 0.9175 - val_loss: 0.3831 - val_accuracy: 0.8804\n",
						"Epoch 98/100\n",
						"28/28 [==============================] - 29s 1s/step - loss: 0.2166 - accuracy: 0.9250 - val_loss: 0.3888 - val_accuracy: 0.8781\n",
						"Epoch 99/100\n",
						"28/28 [==============================] - 28s 1s/step - loss: 0.2178 - accuracy: 0.9244 - val_loss: 0.3554 - val_accuracy: 0.8855\n",
						"Epoch 100/100\n",
						"28/28 [==============================] - 27s 973ms/step - loss: 0.2265 - accuracy: 0.9194 - val_loss: 0.3992 - val_accuracy: 0.8746\n"
					]
				},
				{
					"data": {
						"text/plain": [
							"<keras.callbacks.History at 0x2dc22f564f0>"
						]
					},
					"execution_count": 11,
					"metadata": {},
					"output_type": "execute_result"
				}
			],
			"source": [
				"from datetime import datetime \n",
				"\n",
				"model.fit(x_train, y_train, batch_size=256, epochs=70, validation_data=(x_test, y_test), verbose=1)\n"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 12,
			"metadata": {},
			"outputs": [
				{
					"name": "stdout",
					"output_type": "stream",
					"text": [
						"Training Accuracy:  0.9322834610939026\n",
						"Testing Accuracy:  0.8746422529220581\n"
					]
				}
			],
			"source": [
				"score = model.evaluate(x_train, y_train, verbose=0)\n",
				"print(\"Training Accuracy: \", score[1])\n",
				"\n",
				"score = model.evaluate(x_test, y_test, verbose=0)\n",
				"print(\"Testing Accuracy: \", score[1])"
			]
		},
		{
			"cell_type": "code",
			"execution_count": 93,
			"metadata": {},
			"outputs": [],
			"source": [
				"def print_prediction(file_name):\n",
				"    prediction_feature = extract_features(file_name) \n",
				"    prediction_feature = prediction_feature.reshape(1, num_rows, num_columns, num_channels)\n",
				"\n",
				"    predicted_vector = np.argmax(model.predict(prediction_feature))\n",
				"    predicted_class = metadata[metadata[\"classID\"]==predicted_vector][\"class\"]\n",
				"    print(\"The predicted class is:\", predicted_class.iloc[0], '\\n') "
			]
		},
		{
			"cell_type": "code",
			"execution_count": 102,
			"metadata": {},
			"outputs": [
				{
					"name": "stdout",
					"output_type": "stream",
					"text": [
						"The predicted class is: siren \n",
						"\n"
					]
				}
			],
			"source": [
				"filename = 'UrbanSound8K/audio/fold1/157867-8-0-17.wav' \n",
				"print_prediction(filename)"
			]
		}
	],
	"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": "3.9.6"
		}
	},
	"nbformat": 4,
	"nbformat_minor": 2
}