├── 1. Day 1 - Intro to DeepLearning
    ├── Deep Learning.pdf
    └── deep-learning.jpg
├── 10. Day 10 - Implementation of Loss Function
    ├── loss-functions-implementation.ipynb
    └── loss-functions-implementation.pdf
├── 11. Day 11- Intro to Transfer Learning
    └── Intro to Transfer Learning.pdf
├── 12. Day 12 - Convolutional Neural Network
    └── Convolutional Neural Network.pdf
├── 13. Day 13 - Digit Recognition CNN
    ├── digit-recognizer-cnn.ipynb
    └── digit-recognizer-cnn.pdf
├── 14. Day 14 - Brain Tumor MRI Classification
    ├── brain-tumor-mri-classification.ipynb
    └── brain-tumor-mri-classification.pdf
├── 15. Day 15 - Image Classification Resnet
    ├── image-classification-resnet.ipynb
    └── image-classification-resnet.pdf
├── 16. Day 16 - AlexNet Implementation
    ├── alexnet-implementation.ipynb
    └── alexnet-implementation.pdf
├── 17. Day 17 - LeNet Digit Recognizer
    ├── digit-recognizer-lenet.ipynb
    └── digit-recognizer-lenet.pdf
├── 18. Day 18 - X-Ray Images (Pneumonia) - VGG
    ├── x-ray-images-pneumonia (1).ipynb
    └── x-ray-images-pneumonia-1.pdf
├── 19. Day 19 - Brain Tumor Detection DenseNet
    ├── brain-tumor-detection-densenet.ipynb
    └── brain-tumor-detection-densenet.pdf
├── 2. Day 2 - Intro to Neural Network
    └── Neural Network.pdf
├── 20. Day 20 - Intro to RNN LSTM GRU
    └── RNN LSTM GRU.pdf
├── 21. Day 21 - Lstm For Text Classification
    ├── lstm-for-text-classification.ipynb
    └── lstm-for-text-classification.pdf
├── 3. Day 3 - Perceptron
    └── Perceptron.pdf
├── 4. Day 4 - Activation Functions
    └── Activation Functions in Neural Networks.pdf
├── 5. Day 5 - Customer Churn Prediction
    ├── churn-prediction-using-ann.ipynb
    └── churn-prediction-using-ann.pdf
├── 6. Day 6 - Malware Detection Using MLP
    ├── malware-detection-using-mlp.ipynb
    └── malware-detection-using-mlp.pdf
├── 7. Day 7 - Optimizers in Deep Learning
    └── Optimizers in Deep Learning.pdf
├── 8. Day 8 - Optimizers Comparision
    ├── optimizers-comparison-in-dl.ipynb
    └── optimizers-comparison-in-dl.pdf
└── 9. Day 9 - Loss Function
    └── Loss Function.pdf


/1. Day 1 - Intro to DeepLearning/Deep Learning.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/1. Day 1 - Intro to DeepLearning/Deep Learning.pdf


--------------------------------------------------------------------------------
/1. Day 1 - Intro to DeepLearning/deep-learning.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/1. Day 1 - Intro to DeepLearning/deep-learning.jpg


--------------------------------------------------------------------------------
/10. Day 10 - Implementation of Loss Function/loss-functions-implementation.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/10. Day 10 - Implementation of Loss Function/loss-functions-implementation.pdf


--------------------------------------------------------------------------------
/11. Day 11- Intro to Transfer Learning/Intro to Transfer Learning.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/11. Day 11- Intro to Transfer Learning/Intro to Transfer Learning.pdf


--------------------------------------------------------------------------------
/12. Day 12 - Convolutional Neural Network/Convolutional Neural Network.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/12. Day 12 - Convolutional Neural Network/Convolutional Neural Network.pdf


--------------------------------------------------------------------------------
/13. Day 13 - Digit Recognition CNN/digit-recognizer-cnn.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/13. Day 13 - Digit Recognition CNN/digit-recognizer-cnn.pdf


--------------------------------------------------------------------------------
/14. Day 14 - Brain Tumor MRI Classification/brain-tumor-mri-classification.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/14. Day 14 - Brain Tumor MRI Classification/brain-tumor-mri-classification.pdf


--------------------------------------------------------------------------------
/15. Day 15 - Image Classification Resnet/image-classification-resnet.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/15. Day 15 - Image Classification Resnet/image-classification-resnet.pdf


--------------------------------------------------------------------------------
/16. Day 16 - AlexNet Implementation/alexnet-implementation.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/16. Day 16 - AlexNet Implementation/alexnet-implementation.pdf


--------------------------------------------------------------------------------
/17. Day 17 - LeNet Digit Recognizer/digit-recognizer-lenet.ipynb:
--------------------------------------------------------------------------------
1 | {"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":3004,"databundleVersionId":861823,"sourceType":"competition"}],"dockerImageVersionId":30664,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np \nimport pandas as pd \nimport keras \nfrom keras.models import Sequential\nfrom keras.layers import Conv2D, Dense, MaxPool2D, Dropout, Flatten\nfrom keras.optimizers import Adam\nfrom keras.callbacks import ReduceLROnPlateau\nfrom sklearn.model_selection import train_test_split\nimport matplotlib.pyplot as plt\nimport seaborn as sns","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-03-13T08:10:05.574737Z","iopub.execute_input":"2024-03-13T08:10:05.575756Z","iopub.status.idle":"2024-03-13T08:10:05.584120Z","shell.execute_reply.started":"2024-03-13T08:10:05.575708Z","shell.execute_reply":"2024-03-13T08:10:05.582860Z"},"trusted":true},"execution_count":22,"outputs":[]},{"cell_type":"code","source":"df_train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')\nX_train = df_train.iloc[:, 1:]\nY_train = df_train.iloc[:, 0]","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:10:05.586624Z","iopub.execute_input":"2024-03-13T08:10:05.587114Z","iopub.status.idle":"2024-03-13T08:10:09.153154Z","shell.execute_reply.started":"2024-03-13T08:10:05.587071Z","shell.execute_reply":"2024-03-13T08:10:09.151759Z"},"trusted":true},"execution_count":23,"outputs":[]},{"cell_type":"code","source":"X_train.head()","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:10:09.155408Z","iopub.execute_input":"2024-03-13T08:10:09.155805Z","iopub.status.idle":"2024-03-13T08:10:09.176902Z","shell.execute_reply.started":"2024-03-13T08:10:09.155773Z","shell.execute_reply":"2024-03-13T08:10:09.175897Z"},"trusted":true},"execution_count":24,"outputs":[{"execution_count":24,"output_type":"execute_result","data":{"text/plain":"   pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \\\n0       0       0       0       0       0       0       0       0       0   \n1       0       0       0       0       0       0       0       0       0   \n2       0       0       0       0       0       0       0       0       0   \n3       0       0       0       0       0       0       0       0       0   \n4       0       0       0       0       0       0       0       0       0   \n\n   pixel9  ...  pixel774  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n0       0  ...         0         0         0         0         0         0   \n1       0  ...         0         0         0         0         0         0   \n2       0  ...         0         0         0         0         0         0   \n3       0  ...         0         0         0         0         0         0   \n4       0  ...         0         0         0         0         0         0   \n\n   pixel780  pixel781  pixel782  pixel783  \n0         0         0         0         0  \n1         0         0         0         0  \n2         0         0         0         0  \n3         0         0         0         0  \n4         0         0         0         0  \n\n[5 rows x 784 columns]","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>pixel0</th>\n      <th>pixel1</th>\n      <th>pixel2</th>\n      <th>pixel3</th>\n      <th>pixel4</th>\n      <th>pixel5</th>\n      <th>pixel6</th>\n      <th>pixel7</th>\n      <th>pixel8</th>\n      <th>pixel9</th>\n      <th>...</th>\n      <th>pixel774</th>\n      <th>pixel775</th>\n      <th>pixel776</th>\n      <th>pixel777</th>\n      <th>pixel778</th>\n      <th>pixel779</th>\n      <th>pixel780</th>\n      <th>pixel781</th>\n      <th>pixel782</th>\n      <th>pixel783</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 784 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"Y_train.head()","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:10:09.178592Z","iopub.execute_input":"2024-03-13T08:10:09.179829Z","iopub.status.idle":"2024-03-13T08:10:09.191369Z","shell.execute_reply.started":"2024-03-13T08:10:09.179782Z","shell.execute_reply":"2024-03-13T08:10:09.190176Z"},"trusted":true},"execution_count":25,"outputs":[{"execution_count":25,"output_type":"execute_result","data":{"text/plain":"0    1\n1    0\n2    1\n3    4\n4    0\nName: label, dtype: int64"},"metadata":{}}]},{"cell_type":"code","source":"X_train = np.array(X_train)\nY_train = np.array(Y_train)\n# Normalize inputs\nX_train = X_train / 255.0","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:10:09.194044Z","iopub.execute_input":"2024-03-13T08:10:09.194442Z","iopub.status.idle":"2024-03-13T08:10:09.615168Z","shell.execute_reply.started":"2024-03-13T08:10:09.194409Z","shell.execute_reply":"2024-03-13T08:10:09.614047Z"},"trusted":true},"execution_count":26,"outputs":[]},{"cell_type":"code","source":"def plot_digits(X, Y):\n    for i in range(16):\n        plt.subplot(5, 4, i+1)\n        plt.tight_layout()\n        plt.imshow(X[i].reshape(28, 28), cmap='gray')\n        plt.title('Digit:{}'.format(Y[i]))\n        plt.xticks([])\n        plt.yticks([])\n    plt.show()","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:10:09.620223Z","iopub.execute_input":"2024-03-13T08:10:09.620625Z","iopub.status.idle":"2024-03-13T08:10:09.627905Z","shell.execute_reply.started":"2024-03-13T08:10:09.620595Z","shell.execute_reply":"2024-03-13T08:10:09.626629Z"},"trusted":true},"execution_count":27,"outputs":[]},{"cell_type":"code","source":"plot_digits(X_train, Y_train)","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:10:09.629582Z","iopub.execute_input":"2024-03-13T08:10:09.630542Z","iopub.status.idle":"2024-03-13T08:10:12.251024Z","shell.execute_reply.started":"2024-03-13T08:10:09.630507Z","shell.execute_reply":"2024-03-13T08:10:12.249657Z"},"trusted":true},"execution_count":28,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 16 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"#Train-Test Split\nX_dev, X_val, Y_dev, Y_val = train_test_split(X_train, Y_train, test_size=0.03, shuffle=True, random_state=2019)","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:10:57.933630Z","iopub.execute_input":"2024-03-13T08:10:57.934053Z","iopub.status.idle":"2024-03-13T08:10:58.720570Z","shell.execute_reply.started":"2024-03-13T08:10:57.934023Z","shell.execute_reply":"2024-03-13T08:10:58.719277Z"},"trusted":true},"execution_count":36,"outputs":[]},{"cell_type":"code","source":"T_dev = pd.get_dummies(Y_dev).values\nT_val = pd.get_dummies(Y_val).values","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:11:00.111077Z","iopub.execute_input":"2024-03-13T08:11:00.111759Z","iopub.status.idle":"2024-03-13T08:11:00.120075Z","shell.execute_reply.started":"2024-03-13T08:11:00.111725Z","shell.execute_reply":"2024-03-13T08:11:00.118827Z"},"trusted":true},"execution_count":37,"outputs":[]},{"cell_type":"code","source":"#Reshape the input \nX_dev = X_dev.reshape(X_dev.shape[0], 28, 28, 1)\nX_val = X_val.reshape(X_val.shape[0], 28, 28, 1)","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:11:01.583369Z","iopub.execute_input":"2024-03-13T08:11:01.584081Z","iopub.status.idle":"2024-03-13T08:11:01.589739Z","shell.execute_reply.started":"2024-03-13T08:11:01.584039Z","shell.execute_reply":"2024-03-13T08:11:01.588647Z"},"trusted":true},"execution_count":38,"outputs":[]},{"cell_type":"code","source":"model = Sequential()\nmodel.add(Conv2D(filters=32, kernel_size=(5,5), padding='same', activation='relu', input_shape=(28, 28, 1)))\nmodel.add(MaxPool2D(strides=2))\nmodel.add(Conv2D(filters=48, kernel_size=(5,5), padding='valid', activation='relu'))\nmodel.add(MaxPool2D(strides=2))\nmodel.add(Flatten())\nmodel.add(Dense(256, activation='relu'))\nmodel.add(Dense(84, activation='relu'))\nmodel.add(Dense(10, activation='softmax'))\nmodel.build()\nmodel.summary()","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:11:03.513936Z","iopub.execute_input":"2024-03-13T08:11:03.514708Z","iopub.status.idle":"2024-03-13T08:11:03.645827Z","shell.execute_reply.started":"2024-03-13T08:11:03.514671Z","shell.execute_reply":"2024-03-13T08:11:03.644634Z"},"trusted":true},"execution_count":39,"outputs":[{"output_type":"display_data","data":{"text/plain":"\u001b[1mModel: \"sequential_2\"\u001b[0m\n","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_2\"</span>\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"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│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │           \u001b[38;5;34m832\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_4 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m, \u001b[38;5;34m10\u001b[0m, \u001b[38;5;34m48\u001b[0m)     │        \u001b[38;5;34m38,448\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_5 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m48\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ flatten_2 (\u001b[38;5;33mFlatten\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1200\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;34m256\u001b[0m)            │       \u001b[38;5;34m307,456\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_7 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m84\u001b[0m)             │        \u001b[38;5;34m21,588\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_8 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │           \u001b[38;5;34m850\u001b[0m │\n└─────────────────────────────────┴────────────────────────┴───────────────┘\n","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│ conv2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">832</span> │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ conv2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">48</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">38,448</span> │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">48</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ flatten_2 (<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\">1200</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\">256</span>)            │       <span style=\"color: #00af00; text-decoration-color: #00af00\">307,456</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\">84</span>)             │        <span style=\"color: #00af00; text-decoration-color: #00af00\">21,588</span> │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_8 (<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\">10</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">850</span> │\n└─────────────────────────────────┴────────────────────────┴───────────────┘\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m369,174\u001b[0m (1.41 MB)\n","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\">369,174</span> (1.41 MB)\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m369,174\u001b[0m (1.41 MB)\n","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\">369,174</span> (1.41 MB)\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n","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"},"metadata":{}}]},{"cell_type":"code","source":"adam = Adam(learning_rate=5e-4)  \nmodel.compile(loss='categorical_crossentropy', metrics=['accuracy'], optimizer=adam)\n","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:11:43.426137Z","iopub.execute_input":"2024-03-13T08:11:43.426951Z","iopub.status.idle":"2024-03-13T08:11:43.448342Z","shell.execute_reply.started":"2024-03-13T08:11:43.426896Z","shell.execute_reply":"2024-03-13T08:11:43.447173Z"},"trusted":true},"execution_count":40,"outputs":[]},{"cell_type":"code","source":"# Set a learning rate annealer\nreduce_lr = ReduceLROnPlateau(monitor='val_acc', \n                                patience=3, \n                                verbose=1, \n                                factor=0.2, \n                                min_lr=1e-6)","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:11:45.549676Z","iopub.execute_input":"2024-03-13T08:11:45.550501Z","iopub.status.idle":"2024-03-13T08:11:45.557428Z","shell.execute_reply.started":"2024-03-13T08:11:45.550445Z","shell.execute_reply":"2024-03-13T08:11:45.555985Z"},"trusted":true},"execution_count":41,"outputs":[]},{"cell_type":"code","source":"from tensorflow.keras.preprocessing.image import ImageDataGenerator\n\n# Data Augmentation\ndatagen = ImageDataGenerator(\n            rotation_range=10, \n            width_shift_range=0.1, \n            height_shift_range=0.1, \n            zoom_range=0.1)","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:13:23.135791Z","iopub.execute_input":"2024-03-13T08:13:23.136817Z","iopub.status.idle":"2024-03-13T08:13:23.147569Z","shell.execute_reply.started":"2024-03-13T08:13:23.136775Z","shell.execute_reply":"2024-03-13T08:13:23.146406Z"},"trusted":true},"execution_count":44,"outputs":[]},{"cell_type":"code","source":"datagen.fit(X_dev)\nmodel.fit(datagen.flow(X_dev, T_dev, batch_size=100), steps_per_epoch=int(len(X_dev)/100), \n          epochs=30, validation_data=(X_val, T_val), callbacks=[reduce_lr])","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:14:15.872692Z","iopub.execute_input":"2024-03-13T08:14:15.873197Z","iopub.status.idle":"2024-03-13T08:23:24.861457Z","shell.execute_reply.started":"2024-03-13T08:14:15.873160Z","shell.execute_reply":"2024-03-13T08:23:24.860074Z"},"trusted":true},"execution_count":47,"outputs":[{"name":"stdout","text":"Epoch 1/30\n","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/keras/src/trainers/data_adapters/py_dataset_adapter.py:122: UserWarning: Your `PyDataset` class should call `super().__init__(**kwargs)` in its constructor. `**kwargs` can include `workers`, `use_multiprocessing`, `max_queue_size`. Do not pass these arguments to `fit()`, as they will be ignored.\n  self._warn_if_super_not_called()\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m35s\u001b[0m 80ms/step - accuracy: 0.7171 - loss: 0.8715 - val_accuracy: 0.9667 - val_loss: 0.1378 - learning_rate: 5.0000e-04\nEpoch 2/30\n\u001b[1m  1/407\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m25s\u001b[0m 63ms/step - accuracy: 0.9100 - loss: 0.2468","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/keras/src/callbacks/callback_list.py:97: 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/opt/conda/lib/python3.10/contextlib.py:153: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n  self.gen.throw(typ, value, traceback)\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 945us/step - accuracy: 0.9100 - loss: 0.1237 - val_accuracy: 0.9675 - val_loss: 0.1301 - learning_rate: 5.0000e-04\nEpoch 3/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 79ms/step - accuracy: 0.9478 - loss: 0.1653 - val_accuracy: 0.9738 - val_loss: 0.0898 - learning_rate: 5.0000e-04\nEpoch 4/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 903us/step - accuracy: 0.9600 - loss: 0.0473 - val_accuracy: 0.9746 - val_loss: 0.0852 - learning_rate: 5.0000e-04\nEpoch 5/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 79ms/step - accuracy: 0.9654 - loss: 0.1078 - val_accuracy: 0.9825 - val_loss: 0.0654 - learning_rate: 5.0000e-04\nEpoch 6/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 906us/step - accuracy: 0.9700 - loss: 0.0563 - val_accuracy: 0.9802 - val_loss: 0.0689 - learning_rate: 5.0000e-04\nEpoch 7/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 79ms/step - accuracy: 0.9719 - loss: 0.0875 - val_accuracy: 0.9889 - val_loss: 0.0515 - learning_rate: 5.0000e-04\nEpoch 8/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 884us/step - accuracy: 0.9700 - loss: 0.0259 - val_accuracy: 0.9881 - val_loss: 0.0505 - learning_rate: 5.0000e-04\nEpoch 9/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 78ms/step - accuracy: 0.9790 - loss: 0.0686 - val_accuracy: 0.9802 - val_loss: 0.0681 - learning_rate: 5.0000e-04\nEpoch 10/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 898us/step - accuracy: 0.9800 - loss: 0.0521 - val_accuracy: 0.9817 - val_loss: 0.0652 - learning_rate: 5.0000e-04\nEpoch 11/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 78ms/step - accuracy: 0.9782 - loss: 0.0641 - val_accuracy: 0.9873 - val_loss: 0.0517 - learning_rate: 5.0000e-04\nEpoch 12/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 863us/step - accuracy: 0.9800 - loss: 0.0390 - val_accuracy: 0.9881 - val_loss: 0.0515 - learning_rate: 5.0000e-04\nEpoch 13/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 78ms/step - accuracy: 0.9814 - loss: 0.0574 - val_accuracy: 0.9905 - val_loss: 0.0422 - learning_rate: 5.0000e-04\nEpoch 14/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856us/step - accuracy: 1.0000 - loss: 0.0027 - val_accuracy: 0.9897 - val_loss: 0.0430 - learning_rate: 5.0000e-04\nEpoch 15/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 78ms/step - accuracy: 0.9845 - loss: 0.0508 - val_accuracy: 0.9889 - val_loss: 0.0478 - learning_rate: 5.0000e-04\nEpoch 16/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1ms/step - accuracy: 0.9700 - loss: 0.0351 - val_accuracy: 0.9889 - val_loss: 0.0483 - learning_rate: 5.0000e-04\nEpoch 17/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 77ms/step - accuracy: 0.9849 - loss: 0.0478 - val_accuracy: 0.9937 - val_loss: 0.0259 - learning_rate: 5.0000e-04\nEpoch 18/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 894us/step - accuracy: 0.9900 - loss: 0.0114 - val_accuracy: 0.9937 - val_loss: 0.0267 - learning_rate: 5.0000e-04\nEpoch 19/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 77ms/step - accuracy: 0.9868 - loss: 0.0423 - val_accuracy: 0.9889 - val_loss: 0.0524 - learning_rate: 5.0000e-04\nEpoch 20/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 889us/step - accuracy: 0.9900 - loss: 0.0081 - val_accuracy: 0.9865 - val_loss: 0.0539 - learning_rate: 5.0000e-04\nEpoch 21/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 77ms/step - accuracy: 0.9875 - loss: 0.0394 - val_accuracy: 0.9929 - val_loss: 0.0346 - learning_rate: 5.0000e-04\nEpoch 22/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 898us/step - accuracy: 1.0000 - loss: 0.0088 - val_accuracy: 0.9921 - val_loss: 0.0357 - learning_rate: 5.0000e-04\nEpoch 23/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 78ms/step - accuracy: 0.9893 - loss: 0.0348 - val_accuracy: 0.9937 - val_loss: 0.0287 - learning_rate: 5.0000e-04\nEpoch 24/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 944us/step - accuracy: 1.0000 - loss: 0.0051 - val_accuracy: 0.9944 - val_loss: 0.0287 - learning_rate: 5.0000e-04\nEpoch 25/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 78ms/step - accuracy: 0.9881 - loss: 0.0375 - val_accuracy: 0.9921 - val_loss: 0.0417 - learning_rate: 5.0000e-04\nEpoch 26/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 923us/step - accuracy: 1.0000 - loss: 0.0070 - val_accuracy: 0.9905 - val_loss: 0.0461 - learning_rate: 5.0000e-04\nEpoch 27/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 77ms/step - accuracy: 0.9892 - loss: 0.0352 - val_accuracy: 0.9905 - val_loss: 0.0397 - learning_rate: 5.0000e-04\nEpoch 28/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 894us/step - accuracy: 0.9900 - loss: 0.0198 - val_accuracy: 0.9905 - val_loss: 0.0394 - learning_rate: 5.0000e-04\nEpoch 29/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 77ms/step - accuracy: 0.9906 - loss: 0.0315 - val_accuracy: 0.9897 - val_loss: 0.0475 - learning_rate: 5.0000e-04\nEpoch 30/30\n\u001b[1m407/407\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 907us/step - accuracy: 0.9800 - loss: 0.0811 - val_accuracy: 0.9905 - val_loss: 0.0482 - learning_rate: 5.0000e-04\n","output_type":"stream"},{"execution_count":47,"output_type":"execute_result","data":{"text/plain":"<keras.src.callbacks.history.History at 0x7ca65eaf31c0>"},"metadata":{}}]},{"cell_type":"code","source":"score = model.evaluate(X_val, T_val, batch_size=32)\nscore","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:23:31.505678Z","iopub.execute_input":"2024-03-13T08:23:31.506100Z","iopub.status.idle":"2024-03-13T08:23:32.234145Z","shell.execute_reply.started":"2024-03-13T08:23:31.506069Z","shell.execute_reply":"2024-03-13T08:23:32.233193Z"},"trusted":true},"execution_count":48,"outputs":[{"name":"stdout","text":"\u001b[1m40/40\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 9ms/step - accuracy: 0.9916 - loss: 0.0218\n","output_type":"stream"},{"execution_count":48,"output_type":"execute_result","data":{"text/plain":"[0.048181574791669846, 0.9904761910438538]"},"metadata":{}}]},{"cell_type":"code","source":"df_test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:23:48.826059Z","iopub.execute_input":"2024-03-13T08:23:48.826473Z","iopub.status.idle":"2024-03-13T08:23:51.528091Z","shell.execute_reply.started":"2024-03-13T08:23:48.826444Z","shell.execute_reply":"2024-03-13T08:23:51.526686Z"},"trusted":true},"execution_count":50,"outputs":[]},{"cell_type":"code","source":"X_test = np.array(df_test)\nX_test = X_test/255.0","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:23:58.498602Z","iopub.execute_input":"2024-03-13T08:23:58.499101Z","iopub.status.idle":"2024-03-13T08:23:58.702993Z","shell.execute_reply.started":"2024-03-13T08:23:58.499063Z","shell.execute_reply":"2024-03-13T08:23:58.701129Z"},"trusted":true},"execution_count":51,"outputs":[]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\n# Define a function to display images and their predictions\ndef display_images(images, predictions, num_images=5):\n    plt.figure(figsize=(10, 4))\n    for i in range(num_images):\n        plt.subplot(1, num_images, i+1)\n        plt.imshow(images[i].reshape(28, 28), cmap='gray')\n        plt.title(f'Predicted: {predictions[i]}', fontsize=12)\n        plt.axis('off')\n    plt.show()\n\n# Display the first 5 images along with their predictions\ndisplay_images(X_test[:5], Y_test[:5])\n","metadata":{"execution":{"iopub.status.busy":"2024-03-13T08:26:06.368534Z","iopub.execute_input":"2024-03-13T08:26:06.368994Z","iopub.status.idle":"2024-03-13T08:26:06.793867Z","shell.execute_reply.started":"2024-03-13T08:26:06.368963Z","shell.execute_reply":"2024-03-13T08:26:06.792901Z"},"trusted":true},"execution_count":55,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x400 with 5 Axes>","image/png":"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"},"metadata":{}}]}]}


--------------------------------------------------------------------------------
/17. Day 17 - LeNet Digit Recognizer/digit-recognizer-lenet.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/17. Day 17 - LeNet Digit Recognizer/digit-recognizer-lenet.pdf


--------------------------------------------------------------------------------
/18. Day 18 - X-Ray Images (Pneumonia) - VGG/x-ray-images-pneumonia-1.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/18. Day 18 - X-Ray Images (Pneumonia) - VGG/x-ray-images-pneumonia-1.pdf


--------------------------------------------------------------------------------
/19. Day 19 - Brain Tumor Detection DenseNet/brain-tumor-detection-densenet.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/19. Day 19 - Brain Tumor Detection DenseNet/brain-tumor-detection-densenet.pdf


--------------------------------------------------------------------------------
/2. Day 2 - Intro to Neural Network/Neural Network.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/2. Day 2 - Intro to Neural Network/Neural Network.pdf


--------------------------------------------------------------------------------
/20. Day 20 - Intro to RNN LSTM GRU/RNN LSTM GRU.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/20. Day 20 - Intro to RNN LSTM GRU/RNN LSTM GRU.pdf


--------------------------------------------------------------------------------
/21. Day 21 - Lstm For Text Classification/lstm-for-text-classification.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/21. Day 21 - Lstm For Text Classification/lstm-for-text-classification.pdf


--------------------------------------------------------------------------------
/3. Day 3 - Perceptron/Perceptron.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/3. Day 3 - Perceptron/Perceptron.pdf


--------------------------------------------------------------------------------
/4. Day 4 - Activation Functions/Activation Functions in Neural Networks.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/4. Day 4 - Activation Functions/Activation Functions in Neural Networks.pdf


--------------------------------------------------------------------------------
/5. Day 5 - Customer Churn Prediction/churn-prediction-using-ann.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/5. Day 5 - Customer Churn Prediction/churn-prediction-using-ann.pdf


--------------------------------------------------------------------------------
/6. Day 6 - Malware Detection Using MLP/malware-detection-using-mlp.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/6. Day 6 - Malware Detection Using MLP/malware-detection-using-mlp.pdf


--------------------------------------------------------------------------------
/7. Day 7 - Optimizers in Deep Learning/Optimizers in Deep Learning.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/7. Day 7 - Optimizers in Deep Learning/Optimizers in Deep Learning.pdf


--------------------------------------------------------------------------------
/8. Day 8 - Optimizers Comparision/optimizers-comparison-in-dl.ipynb:
--------------------------------------------------------------------------------
1 | {"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[],"dockerImageVersionId":30664,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"#IMPORTING LIBRARIES\n\nimport tensorflow as tf\nfrom tensorflow.keras.datasets import mnist\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.layers import Conv2D, MaxPooling2D\nfrom tensorflow.keras.layers import Dense, Dropout,Flatten","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-03-02T06:26:05.116139Z","iopub.execute_input":"2024-03-02T06:26:05.116496Z","iopub.status.idle":"2024-03-02T06:26:19.254900Z","shell.execute_reply.started":"2024-03-02T06:26:05.116467Z","shell.execute_reply":"2024-03-02T06:26:19.253678Z"},"trusted":true},"execution_count":1,"outputs":[{"name":"stderr","text":"2024-03-02 06:26:07.111044: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n2024-03-02 06:26:07.111192: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n2024-03-02 06:26:07.257062: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### LOADING DATA","metadata":{}},{"cell_type":"code","source":"(X_train,y_train),(X_test,y_test)=mnist.load_data()","metadata":{"execution":{"iopub.status.busy":"2024-03-02T06:26:23.117097Z","iopub.execute_input":"2024-03-02T06:26:23.118192Z","iopub.status.idle":"2024-03-02T06:26:23.574993Z","shell.execute_reply.started":"2024-03-02T06:26:23.118151Z","shell.execute_reply":"2024-03-02T06:26:23.573763Z"},"trusted":true},"execution_count":2,"outputs":[{"name":"stdout","text":"Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n\u001b[1m11490434/11490434\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 0us/step\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### DATA RESHAPE","metadata":{}},{"cell_type":"code","source":"X_train=X_train.reshape(X_train.shape[0],28,28,1)\nX_test=X_test.reshape(X_test.shape[0],28,28,1)","metadata":{"execution":{"iopub.status.busy":"2024-03-02T06:27:09.586165Z","iopub.execute_input":"2024-03-02T06:27:09.586578Z","iopub.status.idle":"2024-03-02T06:27:09.592214Z","shell.execute_reply.started":"2024-03-02T06:27:09.586546Z","shell.execute_reply":"2024-03-02T06:27:09.591011Z"},"trusted":true},"execution_count":3,"outputs":[]},{"cell_type":"code","source":"X_train=X_train.astype('float32')\nX_test=X_test.astype('float32')\nX_train /=255\nX_test /=255\ny_train=tf.keras.utils.to_categorical(y_train)\ny_test=tf.keras.utils.to_categorical(y_test)","metadata":{"execution":{"iopub.status.busy":"2024-03-02T06:27:19.747880Z","iopub.execute_input":"2024-03-02T06:27:19.748349Z","iopub.status.idle":"2024-03-02T06:27:19.860483Z","shell.execute_reply.started":"2024-03-02T06:27:19.748309Z","shell.execute_reply":"2024-03-02T06:27:19.859325Z"},"trusted":true},"execution_count":4,"outputs":[]},{"cell_type":"markdown","source":"### BUILD OPTIMIZER CALL","metadata":{}},{"cell_type":"code","source":"def build_optimizer(op):\n    model=tf.keras.Sequential()\n    model.add(tf.keras.Input(shape=(28,28,1)))\n    model.add(tf.keras.layers.Conv2D(filters=32, kernel_size=(3,3), strides=1, activation='relu'))\n    model.add(tf.keras.layers.MaxPool2D())\n    model.add(tf.keras.layers.Conv2D(filters=64, kernel_size=(3,3), strides=1, activation='relu'))\n    model.add(tf.keras.layers.Dropout(0.25))\n    model.add(tf.keras.layers.Flatten())\n    model.add(tf.keras.layers.Dense(128, activation='relu'))\n    model.add(tf.keras.layers.Dense(256, activation='relu'))\n    model.add(tf.keras.layers.Dropout(0.5))\n    model.add(tf.keras.layers.Dense(10, activation='softmax'))\n    model.compile(optimizer=op, loss='binary_crossentropy', metrics=['accuracy'])\n    return model","metadata":{"execution":{"iopub.status.busy":"2024-03-02T06:27:40.830113Z","iopub.execute_input":"2024-03-02T06:27:40.830508Z","iopub.status.idle":"2024-03-02T06:27:40.839320Z","shell.execute_reply.started":"2024-03-02T06:27:40.830479Z","shell.execute_reply":"2024-03-02T06:27:40.838012Z"},"trusted":true},"execution_count":5,"outputs":[]},{"cell_type":"markdown","source":"### COMPARING EACH OPTIMIZER ACCURACY","metadata":{}},{"cell_type":"code","source":"import os, gc\n\noptimizers=['Adam', 'RMSprop','Adadelta', 'Adagrad',  'SGD']\nopt_res=[]\nmodel_res=[]\nfor i in optimizers:\n    model=build_optimizer(i)\n    print(\"Accuracy for: \",i)\n    print(\"\\n\")\n    history=model.fit(X_train,y_train, epochs=5, batch_size=64,verbose=1, validation_data=(X_test, y_test))\n    print(\"\\n\")\n    gc.collect()\n    model_res.append(history)\n    opt_res.append(history.history['accuracy'])","metadata":{"execution":{"iopub.status.busy":"2024-03-02T06:32:20.491806Z","iopub.execute_input":"2024-03-02T06:32:20.493213Z","iopub.status.idle":"2024-03-02T06:47:03.236793Z","shell.execute_reply.started":"2024-03-02T06:32:20.493162Z","shell.execute_reply":"2024-03-02T06:47:03.235583Z"},"trusted":true},"execution_count":7,"outputs":[{"name":"stdout","text":"Accuracy for:  Adam\n\n\nEpoch 1/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m34s\u001b[0m 34ms/step - accuracy: 0.8167 - loss: 0.0992 - val_accuracy: 0.9815 - val_loss: 0.0103\nEpoch 2/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 33ms/step - accuracy: 0.9837 - loss: 0.0110 - val_accuracy: 0.9893 - val_loss: 0.0067\nEpoch 3/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 33ms/step - accuracy: 0.9885 - loss: 0.0076 - val_accuracy: 0.9889 - val_loss: 0.0070\nEpoch 4/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m30s\u001b[0m 32ms/step - accuracy: 0.9920 - loss: 0.0058 - val_accuracy: 0.9900 - val_loss: 0.0066\nEpoch 5/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 32ms/step - accuracy: 0.9937 - loss: 0.0042 - val_accuracy: 0.9912 - val_loss: 0.0057\n\n\nAccuracy for:  RMSprop\n\n\nEpoch 1/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 32ms/step - accuracy: 0.7934 - loss: 0.1095 - val_accuracy: 0.9810 - val_loss: 0.0106\nEpoch 2/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 33ms/step - accuracy: 0.9780 - loss: 0.0140 - val_accuracy: 0.9865 - val_loss: 0.0077\nEpoch 3/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 33ms/step - accuracy: 0.9868 - loss: 0.0089 - val_accuracy: 0.9900 - val_loss: 0.0061\nEpoch 4/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 34ms/step - accuracy: 0.9911 - loss: 0.0061 - val_accuracy: 0.9915 - val_loss: 0.0053\nEpoch 5/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 32ms/step - accuracy: 0.9939 - loss: 0.0044 - val_accuracy: 0.9900 - val_loss: 0.0063\n\n\nAccuracy for:  Adadelta\n\n\nEpoch 1/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 34ms/step - accuracy: 0.1001 - loss: 0.6806 - val_accuracy: 0.1069 - val_loss: 0.6311\nEpoch 2/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 33ms/step - accuracy: 0.1005 - loss: 0.5991 - val_accuracy: 0.1278 - val_loss: 0.4308\nEpoch 3/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 33ms/step - accuracy: 0.1077 - loss: 0.4143 - val_accuracy: 0.1536 - val_loss: 0.3345\nEpoch 4/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 34ms/step - accuracy: 0.1176 - loss: 0.3604 - val_accuracy: 0.2520 - val_loss: 0.3248\nEpoch 5/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 35ms/step - accuracy: 0.1237 - loss: 0.3522 - val_accuracy: 0.3556 - val_loss: 0.3196\n\n\nAccuracy for:  Adagrad\n\n\nEpoch 1/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 33ms/step - accuracy: 0.0961 - loss: 0.6229 - val_accuracy: 0.1029 - val_loss: 0.3300\nEpoch 2/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 33ms/step - accuracy: 0.1181 - loss: 0.3461 - val_accuracy: 0.4434 - val_loss: 0.3171\nEpoch 3/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m30s\u001b[0m 31ms/step - accuracy: 0.1546 - loss: 0.3326 - val_accuracy: 0.6627 - val_loss: 0.3044\nEpoch 4/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 31ms/step - accuracy: 0.2225 - loss: 0.3183 - val_accuracy: 0.6934 - val_loss: 0.2851\nEpoch 5/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 30ms/step - accuracy: 0.3294 - loss: 0.2986 - val_accuracy: 0.7023 - val_loss: 0.2536\n\n\nAccuracy for:  SGD\n\n\nEpoch 1/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 29ms/step - accuracy: 0.1162 - loss: 0.5056 - val_accuracy: 0.6243 - val_loss: 0.2982\nEpoch 2/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 30ms/step - accuracy: 0.3258 - loss: 0.2965 - val_accuracy: 0.7799 - val_loss: 0.1752\nEpoch 3/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 30ms/step - accuracy: 0.6496 - loss: 0.1944 - val_accuracy: 0.8570 - val_loss: 0.1079\nEpoch 4/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 30ms/step - accuracy: 0.7635 - loss: 0.1414 - val_accuracy: 0.8869 - val_loss: 0.0815\nEpoch 5/5\n\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m30s\u001b[0m 32ms/step - accuracy: 0.8178 - loss: 0.1143 - val_accuracy: 0.9006 - val_loss: 0.0687\n\n\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### PLOTTING OPTIMIZERS ACCURACY","metadata":{}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\nfully_nested = [list(zip(*[(ix+1,y) for ix,y in enumerate(x)])) for x in opt_res]\nnames = ['sublist%d'%(i+1) for i in range(len(fully_nested))]\n\nfig = plt.figure(figsize=(15,10))\n\nfor l in fully_nested:\n    plt.plot(*l)\n\nplt.xlabel(\"Epochs\")\nplt.ylabel(\"Accuracy\")\nplt.legend(optimizers, fontsize=9, loc = 'upper right', bbox_to_anchor=(1.1, 1.01))\nplt.title(\"Optimizer Performance Comparison\", fontsize=25)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-03-02T06:47:57.385509Z","iopub.execute_input":"2024-03-02T06:47:57.385938Z","iopub.status.idle":"2024-03-02T06:47:57.831971Z","shell.execute_reply.started":"2024-03-02T06:47:57.385904Z","shell.execute_reply":"2024-03-02T06:47:57.830841Z"},"trusted":true},"execution_count":8,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1500x1000 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"Adam and RMSprop performed the best in terms of accuracy and loss reduction, with Adam being slightly better. Adadelta performed poorly, while Adagrad and SGD showed improvement over epochs but didn't reach the same level of performance as Adam and RMSprop within the given epochs.","metadata":{}}]}


--------------------------------------------------------------------------------
/8. Day 8 - Optimizers Comparision/optimizers-comparison-in-dl.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/8. Day 8 - Optimizers Comparision/optimizers-comparison-in-dl.pdf


--------------------------------------------------------------------------------
/9. Day 9 - Loss Function/Loss Function.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/Abdullah-khan0/50-Days-OfCode-DeepLearning/16571829d4595c9dbb4039e27c22e0b952bd4f9a/9. Day 9 - Loss Function/Loss Function.pdf


--------------------------------------------------------------------------------