{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.7.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceType":"competition","sourceId":3004,"databundleVersionId":861823}],"dockerImageVersionId":30260,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:19.569179Z","iopub.execute_input":"2026-03-17T08:48:19.569476Z","iopub.status.idle":"2026-03-17T08:48:19.578311Z","shell.execute_reply.started":"2026-03-17T08:48:19.569454Z","shell.execute_reply":"2026-03-17T08:48:19.577310Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"mnist_train = pd.read_csv(\"../input/digit-recognizer/train.csv\")\nmnist_test = pd.read_csv(\"../input/digit-recognizer/test.csv\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:19.579792Z","iopub.execute_input":"2026-03-17T08:48:19.580131Z","iopub.status.idle":"2026-03-17T08:48:22.556666Z","shell.execute_reply.started":"2026-03-17T08:48:19.580108Z","shell.execute_reply":"2026-03-17T08:48:22.555937Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"y_train1 = pd.get_dummies(mnist_train.label)\nprint(y_train1.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:22.557840Z","iopub.execute_input":"2026-03-17T08:48:22.558241Z","iopub.status.idle":"2026-03-17T08:48:22.568495Z","shell.execute_reply.started":"2026-03-17T08:48:22.558205Z","shell.execute_reply":"2026-03-17T08:48:22.567623Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"npx = mnist_train.drop(['label'], axis=1)\nx_train1 = np.array(npx).reshape(42000,28,28,1)/255\nprint(x_train1.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:22.570348Z","iopub.execute_input":"2026-03-17T08:48:22.570589Z","iopub.status.idle":"2026-03-17T08:48:22.850473Z","shell.execute_reply.started":"2026-03-17T08:48:22.570567Z","shell.execute_reply":"2026-03-17T08:48:22.849418Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split #to split our dataset\n# x_train, x_val, y_train, y_val = train_test_split(x_train1, y_train1, test_size=0.1, random_state=101)\nx_train = x_train1\ny_train = y_train1","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:22.851633Z","iopub.execute_input":"2026-03-17T08:48:22.851957Z","iopub.status.idle":"2026-03-17T08:48:22.857406Z","shell.execute_reply.started":"2026-03-17T08:48:22.851923Z","shell.execute_reply":"2026-03-17T08:48:22.856440Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"npx = mnist_test\nx_test = np.array(npx).reshape(28000,28,28,1)/255\nprint(x_test.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:22.858559Z","iopub.execute_input":"2026-03-17T08:48:22.858829Z","iopub.status.idle":"2026-03-17T08:48:22.999955Z","shell.execute_reply.started":"2026-03-17T08:48:22.858792Z","shell.execute_reply":"2026-03-17T08:48:22.999087Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import keras\nfrom keras.models import Sequential\nfrom keras.optimizers import adam_v2\nfrom keras.utils.np_utils import to_categorical\nfrom keras.layers import Dense, Dropout, Flatten, BatchNormalization\nfrom keras.layers.convolutional import Conv2D, MaxPooling2D\nfrom keras.preprocessing import image\nfrom keras.callbacks import ReduceLROnPlateau, EarlyStopping\n\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport cv2 \nfrom PIL import Image \n\nnp.random.seed(99)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:23.001298Z","iopub.execute_input":"2026-03-17T08:48:23.001705Z","iopub.status.idle":"2026-03-17T08:48:23.007788Z","shell.execute_reply.started":"2026-03-17T08:48:23.001670Z","shell.execute_reply":"2026-03-17T08:48:23.006957Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def display(img):\n  plt.figure(figsize=(5,5))\n  plt.axis(\"off\")\n  plt.imshow(img,cmap=\"gray\")\n\ndisplay(x_train[1191])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:23.009513Z","iopub.execute_input":"2026-03-17T08:48:23.010025Z","iopub.status.idle":"2026-03-17T08:48:23.067858Z","shell.execute_reply.started":"2026-03-17T08:48:23.009999Z","shell.execute_reply":"2026-03-17T08:48:23.066915Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from keras.preprocessing.image import ImageDataGenerator\n\ndatagen = ImageDataGenerator(rotation_range = 10, width_shift_range = 0.1, shear_range = 0.1,\n                             height_shift_range = 0.1, zoom_range = [0.9,1.1], fill_mode = 'nearest', \n                             horizontal_flip = False, vertical_flip = False)\ndatagen.fit(x_train)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:23.070412Z","iopub.execute_input":"2026-03-17T08:48:23.070656Z","iopub.status.idle":"2026-03-17T08:48:23.174809Z","shell.execute_reply.started":"2026-03-17T08:48:23.070634Z","shell.execute_reply":"2026-03-17T08:48:23.173779Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def Conv_Net_model():\n    model = Sequential()\n    # Blok Konvolusi 1 (Stacked)\n    model.add(Conv2D(32, kernel_size=(3,3), activation=\"relu\", padding=\"same\", input_shape=(28,28,1)))\n    model.add(BatchNormalization())\n    model.add(Conv2D(32, kernel_size=(3,3), activation=\"relu\", padding=\"same\"))\n    model.add(BatchNormalization())\n    model.add(MaxPooling2D(pool_size=(2,2)))\n    model.add(Dropout(0.25)) # Menggunakan dropout kecil pada fitur spasial\n    \n    # Blok Konvolusi 2 (Stacked)\n    model.add(Conv2D(64, kernel_size=(3,3), activation=\"relu\", padding=\"same\"))\n    model.add(BatchNormalization())\n    model.add(Conv2D(64, kernel_size=(3,3), activation=\"relu\", padding=\"same\"))\n    model.add(BatchNormalization())\n    model.add(MaxPooling2D(pool_size=(2,2)))\n    model.add(Dropout(0.25))\n    \n    # Fully Connected Layer\n    model.add(Flatten())\n    model.add(Dense(256, activation=\"relu\")) # Diperbesar dari 200 ke 256 (power of 2)\n    model.add(BatchNormalization())\n    model.add(Dropout(0.5)) # Dropout tinggi untuk mencegah overfitting di layer dense\n    model.add(Dense(10, activation=\"softmax\"))\n    return model\n\nmodel = Conv_Net_model()\nprint(model.summary())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:23.177209Z","iopub.execute_input":"2026-03-17T08:48:23.177488Z","iopub.status.idle":"2026-03-17T08:48:23.285877Z","shell.execute_reply.started":"2026-03-17T08:48:23.177464Z","shell.execute_reply":"2026-03-17T08:48:23.285081Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"reduce_lr = ReduceLROnPlateau(monitor = 'accuracy', \n                              min_lr = 0.00001, verbose = 1,  patience = 3, factor = 0.5, mode = 'auto', cooldown = 0)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:23.287013Z","iopub.execute_input":"2026-03-17T08:48:23.287351Z","iopub.status.idle":"2026-03-17T08:48:23.292028Z","shell.execute_reply.started":"2026-03-17T08:48:23.287317Z","shell.execute_reply":"2026-03-17T08:48:23.291059Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"opt = adam_v2.Adam(0.0003, beta_1=0.9, beta_2=0.999, amsgrad=True)\nmodel.compile(optimizer=opt, loss = \"categorical_crossentropy\", metrics = [\"accuracy\"])\nh = model.fit(datagen.flow(x_train,y_train, batch_size=64), epochs=40, \n              shuffle=1, verbose=1, callbacks=[reduce_lr])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:48:23.293559Z","iopub.execute_input":"2026-03-17T08:48:23.293834Z","iopub.status.idle":"2026-03-17T08:55:38.329688Z","shell.execute_reply.started":"2026-03-17T08:48:23.293811Z","shell.execute_reply":"2026-03-17T08:55:38.328914Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"y_pred=model.predict(x_test)\ny_test = np.argmax(y_pred,axis=1)\nprint(y_test.shape)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:55:38.330915Z","iopub.execute_input":"2026-03-17T08:55:38.331254Z","iopub.status.idle":"2026-03-17T08:55:39.853156Z","shell.execute_reply.started":"2026-03-17T08:55:38.331225Z","shell.execute_reply":"2026-03-17T08:55:39.852359Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"sample = pd.read_csv(\"../input/digit-recognizer/sample_submission.csv\")\ni = sample.ImageId\n\noutput = pd.DataFrame({'ImageId' : i,'Label': y_test})\noutput.head(20)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:55:39.854296Z","iopub.execute_input":"2026-03-17T08:55:39.854662Z","iopub.status.idle":"2026-03-17T08:55:39.875048Z","shell.execute_reply.started":"2026-03-17T08:55:39.854623Z","shell.execute_reply":"2026-03-17T08:55:39.874181Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"display(x_test[18])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:55:39.876080Z","iopub.execute_input":"2026-03-17T08:55:39.876392Z","iopub.status.idle":"2026-03-17T08:55:39.982209Z","shell.execute_reply.started":"2026-03-17T08:55:39.876369Z","shell.execute_reply":"2026-03-17T08:55:39.980907Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"output.to_csv('submission.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-03-17T08:55:39.983763Z","iopub.execute_input":"2026-03-17T08:55:39.984189Z","iopub.status.idle":"2026-03-17T08:55:40.049256Z","shell.execute_reply.started":"2026-03-17T08:55:39.984145Z","shell.execute_reply":"2026-03-17T08:55:40.048585Z"}},"outputs":[],"execution_count":null}]}