{"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":30775,"isInternetEnabled":false,"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 os\nfrom sklearn.impute import SimpleImputer\nfrom sklearn.metrics import cohen_kappa_score\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.preprocessing import OneHotEncoder\nimport xgboost as xgb\n\n# As a simple approach:\n# - Only use labeled tabular data for now.\n# - Ignore parquet data.\n# - Ignore PCIAT columns which don't exist in test set (these are directly used to calculate sii, so we could consider them as intermediate targets to predict)\n# - One-hot encode strings.\n# - Impute missing numbers as mean of that feature. This includes string one-hot encodings for now.\n# - Avoid further prep by using XGBoost as model\n# - Use simple set-aside test set for local evaluation. Do not tune hyperparameters yet.\n\ndf_train = pd.read_csv('/kaggle/input/child-mind-institute-problematic-internet-use/train.csv')\ndf_test = pd.read_csv('/kaggle/input/child-mind-institute-problematic-internet-use/test.csv')\n\ndf_train = df_train.drop(columns=df_train.filter(like='PCIAT').columns)\ndf_train = df_train.drop(columns=['id'], axis=1)\ndf_train_unlabeled = df_train.query('sii!=sii') # TODO - determine what to do with this. Clustering?\ndf_train_labeled = df_train.query('sii==sii')\nX_train_labeled, Y_train_labeled = df_train_labeled.drop('sii', axis=1), df_train_labeled['sii']\n\ndf_test, index_test = df_test.drop(columns=['id'], axis=1), df_test['id']\n\n# Do not use unlabeled data for now.\ntX, vX, tY, vY = train_test_split(X_train_labeled, Y_train_labeled, test_size=0.2, random_state=42)\n\nstring_encoder = OneHotEncoder(drop=None, handle_unknown='ignore')\ntX = string_encoder.fit_transform(tX)\nvX = string_encoder.transform(vX)\ndf_test = string_encoder.transform(df_test)\n\nnumeric_imputer = SimpleImputer(missing_values=np.nan, strategy='mean', copy=True)\ntX = numeric_imputer.fit_transform(tX)\nvX = numeric_imputer.transform(vX)\ndf_test = numeric_imputer.transform(df_test)\n\nmodel = xgb.XGBClassifier(tree_method=\"hist\")\nmodel.fit(tX, tY)\nvY_pred = model.predict(vX)\ntestY_pred = model.predict(df_test)\n\nprint(cohen_kappa_score(vY, vY_pred, weights='quadratic'))\n\npY = pd.DataFrame(testY_pred, index=index_test, columns=['sii'])  # ensure that labels and observations are in corresponding order\npY.astype(int).to_csv('submission.csv', index_label='id')","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-09-26T04:07:01.087787Z","iopub.execute_input":"2024-09-26T04:07:01.088323Z","iopub.status.idle":"2024-09-26T04:07:08.227630Z","shell.execute_reply.started":"2024-09-26T04:07:01.088281Z","shell.execute_reply":"2024-09-26T04:07:08.226355Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}