{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":81933,"databundleVersionId":9643020,"sourceType":"competition"}],"dockerImageVersionId":30786,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"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":"2024-12-05T13:28:15.061552Z","iopub.execute_input":"2024-12-05T13:28:15.061905Z","iopub.status.idle":"2024-12-05T13:28:16.320284Z","shell.execute_reply.started":"2024-12-05T13:28:15.061867Z","shell.execute_reply":"2024-12-05T13:28:16.318687Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"#Siapkan semua library yang dibutuhkan\nimport lightgbm as lgb\nfrom sklearn.metrics import cohen_kappa_score\nfrom sklearn.model_selection import train_test_split, StratifiedKFold\nimport pandas as pd\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nimport warnings\nfrom sklearn.preprocessing import LabelEncoder, StandardScaler\nimport tensorflow as tf\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.layers import Dense, Dropout, BatchNormalization\nfrom tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau\n\n# Mengabaikan peringatan yang tidak diperlukan agar tidak duplikasi\nwarnings.filterwarnings('ignore', category=UserWarning, message=\".*is a future warning\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:16.322386Z","iopub.execute_input":"2024-12-05T13:28:16.323773Z","iopub.status.idle":"2024-12-05T13:28:21.406063Z","shell.execute_reply.started":"2024-12-05T13:28:16.323711Z","shell.execute_reply":"2024-12-05T13:28:21.404495Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Memuat dataset\ntrain_df = pd.read_csv('/kaggle/input/child-mind-institute-problematic-internet-use/train.csv')\n\ntrain_df.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.413500Z","iopub.execute_input":"2024-12-05T13:28:21.414023Z","iopub.status.idle":"2024-12-05T13:28:21.496210Z","shell.execute_reply.started":"2024-12-05T13:28:21.413972Z","shell.execute_reply":"2024-12-05T13:28:21.494997Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"test_df = pd.read_csv('/kaggle/input/child-mind-institute-problematic-internet-use/test.csv')\ntest_df.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.497732Z","iopub.execute_input":"2024-12-05T13:28:21.498100Z","iopub.status.idle":"2024-12-05T13:28:21.528750Z","shell.execute_reply.started":"2024-12-05T13:28:21.498065Z","shell.execute_reply":"2024-12-05T13:28:21.527483Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Menampilkan kolom yang ada pada train dan test untuk membandingkan\ntrain_columns = set(train_df.columns)\ntest_columns = set(test_df.columns)\n\n# Menyaring kolom yang ada di train tapi tidak ada di test dan sebaliknya\nmissing_in_test = train_columns - test_columns\nmissing_in_train = test_columns - train_columns\n\nprint(f\"Kolom yang ditemukan di training tetapi tidak ada di test: {missing_in_test}\")\nprint(f\"Kolom yang ditemukan di test tetapi tidak ada di training: {missing_in_train}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.530443Z","iopub.execute_input":"2024-12-05T13:28:21.530915Z","iopub.status.idle":"2024-12-05T13:28:21.538584Z","shell.execute_reply.started":"2024-12-05T13:28:21.530863Z","shell.execute_reply":"2024-12-05T13:28:21.537467Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Menghapus baris dengan nilai 'sii' yang hilang\ntrain_df_clean = train_df.dropna(subset=['sii'])\n\n# Menangani kolom numerik yang hilang dengan imputasi median\nnumeric_cols = train_df_clean.select_dtypes(include=['float', 'int']).columns\nfor col in numeric_cols:\n    train_df_clean[col].fillna(train_df_clean[col].median(), inplace=True)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.539934Z","iopub.execute_input":"2024-12-05T13:28:21.540312Z","iopub.status.idle":"2024-12-05T13:28:21.624754Z","shell.execute_reply.started":"2024-12-05T13:28:21.540272Z","shell.execute_reply":"2024-12-05T13:28:21.609381Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Menangani kolom kategorikal yang hilang dengan imputasi modus\ncategorical_cols = train_df_clean.select_dtypes(include=['object']).columns\nfor col in categorical_cols:\n    train_df_clean[col].fillna(train_df_clean[col].mode()[0], inplace=True)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.626448Z","iopub.execute_input":"2024-12-05T13:28:21.626844Z","iopub.status.idle":"2024-12-05T13:28:21.649514Z","shell.execute_reply.started":"2024-12-05T13:28:21.626806Z","shell.execute_reply":"2024-12-05T13:28:21.648154Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Menghapus kolom yang tidak ada di test set\ncolumns_to_drop = list(missing_in_test - {'sii'})\ntrain_df_clean = train_df_clean.drop(columns=columns_to_drop)\n\n# Menghapus kolom yang tidak relevan seperti 'sii' (target) dan 'id'\nX_train_cleaned = train_df_clean.drop(columns=['sii', 'id'])\nX_test_cleaned = test_df.drop(columns=['id'])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.650778Z","iopub.execute_input":"2024-12-05T13:28:21.651099Z","iopub.status.idle":"2024-12-05T13:28:21.667749Z","shell.execute_reply.started":"2024-12-05T13:28:21.651069Z","shell.execute_reply":"2024-12-05T13:28:21.666432Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Mengonversi kolom kategorikal menjadi kode kategori untuk konsistensi dengan data pelatihan\nfor col in X_test_cleaned.select_dtypes(include=['object']).columns:\n    X_test_cleaned[col] = X_test_cleaned[col].astype('category').cat.codes\n\n# Mengisi nilai yang hilang di test set menggunakan statistik dari data pelatihan\nfor col in X_test_cleaned.select_dtypes(include=['float', 'int']).columns:\n    X_test_cleaned[col].fillna(X_train_cleaned[col].median(), inplace=True)\n\nfor col in X_test_cleaned.select_dtypes(include=['category']).columns:\n    X_test_cleaned[col].fillna(X_test_cleaned[col].mode()[0], inplace=True)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.669408Z","iopub.execute_input":"2024-12-05T13:28:21.669887Z","iopub.status.idle":"2024-12-05T13:28:21.719865Z","shell.execute_reply.started":"2024-12-05T13:28:21.669837Z","shell.execute_reply":"2024-12-05T13:28:21.718758Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Menerapkan LabelEncoder pada fitur kategorikal di set pelatihan\nlabel_encoder = LabelEncoder()\ncategorical_columns = X_train_cleaned.select_dtypes(include=['object']).columns\nfor col in categorical_columns:\n    X_train_cleaned[col] = label_encoder.fit_transform(X_train_cleaned[col])\n\n# Memisahkan data menjadi fitur (X) dan target (y)\nX = X_train_cleaned\ny = train_df_clean['sii']\n\n# Normalisasi data menggunakan StandardScaler\nscaler = StandardScaler()\nX_scaled = scaler.fit_transform(X)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.721169Z","iopub.execute_input":"2024-12-05T13:28:21.721536Z","iopub.status.idle":"2024-12-05T13:28:21.748550Z","shell.execute_reply.started":"2024-12-05T13:28:21.721504Z","shell.execute_reply":"2024-12-05T13:28:21.747533Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Mendefinisikan model MLP sederhana untuk klasifikasi\ndef create_mlp(input_shape):\n    model = Sequential()\n    model.add(Dense(128, activation='relu', input_shape=(input_shape,)))\n    model.add(BatchNormalization())\n    model.add(Dropout(0.3))\n\n    model.add(Dense(64, activation='relu'))\n    model.add(BatchNormalization())\n    model.add(Dropout(0.3))\n\n    model.add(Dense(32, activation='relu'))\n    model.add(BatchNormalization())\n    model.add(Dropout(0.2))\n\n    model.add(Dense(4, activation='softmax'))  # Layer output dengan aktivasi softmax\n    return model\n\n# Setup cross-validation K-fold\nkf = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\nqwk_scores = []\n\n# Loop cross-validation\nfor train_index, val_index in kf.split(X, y):\n    X_train, X_val = X.iloc[train_index], X.iloc[val_index]\n    y_train, y_val = y.iloc[train_index], y.iloc[val_index]\n\n    # Normalisasi data\n    X_train_scaled = scaler.fit_transform(X_train)\n    X_val_scaled = scaler.transform(X_val)\n\n    # Membuat model\n    mlp_model = create_mlp(X_train_scaled.shape[1])\n    mlp_model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.001),\n                      loss='sparse_categorical_crossentropy',\n                      metrics=['accuracy'])\n\n    # Callback untuk early stopping dan pengurangan learning rate\n    early_stopping = EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)\n    reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.5, patience=5, verbose=1)\n\n    # Melatih model\n    history = mlp_model.fit(X_train_scaled, y_train,\n                            validation_data=(X_val_scaled, y_val),\n                            epochs=100,\n                            batch_size=64,\n                            callbacks=[early_stopping, reduce_lr],\n                            verbose=1)\n\n    # Evaluasi performa\n    y_pred_probs = mlp_model.predict(X_val_scaled)\n    y_pred = np.argmax(y_pred_probs, axis=1)\n    qwk_score = cohen_kappa_score(y_val, y_pred, weights='quadratic')\n    qwk_scores.append(qwk_score)\n\n# Menampilkan rata-rata skor QWK dari semua fold\nprint(f\"Rata-rata skor Quadratic Weighted Kappa: {np.mean(qwk_scores):.4f}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:28:21.749737Z","iopub.execute_input":"2024-12-05T13:28:21.750058Z","iopub.status.idle":"2024-12-05T13:29:10.037572Z","shell.execute_reply.started":"2024-12-05T13:28:21.750026Z","shell.execute_reply":"2024-12-05T13:29:10.035627Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Prediksi model pada test set\nX_test_scaled = scaler.transform(X_test_cleaned)\ntest_pred_probs = mlp_model.predict(X_test_scaled)\ntest_pred = np.argmax(test_pred_probs, axis=1)\n\n# Menyiapkan dataframe untuk pengiriman\nsubmission = pd.DataFrame({\n    'id': test_df['id'],\n    'sii': test_pred\n})\n\n# Menyimpan file submission\nsubmission.to_csv('submission.csv', index=False)\nprint(submission)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:29:10.042790Z","iopub.execute_input":"2024-12-05T13:29:10.043239Z","iopub.status.idle":"2024-12-05T13:29:10.142396Z","shell.execute_reply.started":"2024-12-05T13:29:10.043184Z","shell.execute_reply":"2024-12-05T13:29:10.141196Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Visualisasi hasil prediksi (Grafik Batang)\nplt.figure(figsize=(10, 6))\nsns.countplot(x=test_pred)\nplt.title(\"Distribusi Prediksi SII pada Test Set\")\nplt.xlabel(\"Kelas Prediksi\")\nplt.ylabel(\"Frekuensi\")\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:29:10.143757Z","iopub.execute_input":"2024-12-05T13:29:10.144223Z","iopub.status.idle":"2024-12-05T13:29:10.352704Z","shell.execute_reply.started":"2024-12-05T13:29:10.144185Z","shell.execute_reply":"2024-12-05T13:29:10.351567Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Visualisasi akurasi pelatihan\nplt.figure(figsize=(12, 6))\nplt.plot(history.history['accuracy'], label='Akurasi Training')\nplt.plot(history.history['val_accuracy'], label='Akurasi Validasi')\nplt.title(\"Akurasi Model selama Pelatihan\")\nplt.xlabel(\"Epochs\")\nplt.ylabel(\"Akurasi\")\nplt.legend()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:29:10.354161Z","iopub.execute_input":"2024-12-05T13:29:10.354531Z","iopub.status.idle":"2024-12-05T13:29:10.688761Z","shell.execute_reply.started":"2024-12-05T13:29:10.354499Z","shell.execute_reply":"2024-12-05T13:29:10.686812Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Visualisasi kehilangan (loss) pelatihan\nplt.figure(figsize=(12, 6))\nplt.plot(history.history['loss'], label='Loss Training')\nplt.plot(history.history['val_loss'], label='Loss Validasi')\nplt.title(\"Kehilangan Model selama Pelatihan\")\nplt.xlabel(\"Epochs\")\nplt.ylabel(\"Loss\")\nplt.legend()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-05T13:29:10.691285Z","iopub.execute_input":"2024-12-05T13:29:10.691916Z","iopub.status.idle":"2024-12-05T13:29:11.061390Z","shell.execute_reply.started":"2024-12-05T13:29:10.691853Z","shell.execute_reply":"2024-12-05T13:29:11.057843Z"}},"outputs":[],"execution_count":null}]}