{"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":84493,"databundleVersionId":9871156,"sourceType":"competition"},{"sourceId":203522,"sourceType":"modelInstanceVersion","modelInstanceId":173627,"modelId":195958},{"sourceId":205501,"sourceType":"modelInstanceVersion","isSourceIdPinned":true,"modelInstanceId":175256,"modelId":197610},{"sourceId":205503,"sourceType":"modelInstanceVersion","isSourceIdPinned":true,"modelInstanceId":175258,"modelId":197613},{"sourceId":206699,"sourceType":"modelInstanceVersion","isSourceIdPinned":true,"modelInstanceId":176223,"modelId":198552},{"sourceId":207306,"sourceType":"modelInstanceVersion","isSourceIdPinned":true,"modelInstanceId":176745,"modelId":199054}],"dockerImageVersionId":30804,"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\nimport polars as pl\nimport joblib\nimport lightgbm as lgb\nimport kaggle_evaluation.jane_street_inference_server","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-23T03:29:08.812208Z","iopub.execute_input":"2024-12-23T03:29:08.812597Z","iopub.status.idle":"2024-12-23T03:29:10.544716Z","shell.execute_reply.started":"2024-12-23T03:29:08.812563Z","shell.execute_reply":"2024-12-23T03:29:10.543533Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"multi_index = ['date_id', 'time_id', 'symbol_id']\nfeature_col = [f'feature_{i:02d}' for i in range(79)]\nmost_na_drop = ['feature_00', 'feature_01', 'feature_02', 'feature_03', 'feature_04',\n                'feature_21', 'feature_26', 'feature_27', 'feature_31', ]\ntarget = 'responder_6'","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-23T03:29:10.547052Z","iopub.execute_input":"2024-12-23T03:29:10.547798Z","iopub.status.idle":"2024-12-23T03:29:10.553840Z","shell.execute_reply.started":"2024-12-23T03:29:10.547744Z","shell.execute_reply":"2024-12-23T03:29:10.552730Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"data = joblib.load('/kaggle/input/lgbm_d6_wieght/other/default/1/lgbm_d6_n1000_weight_no_id.pkl')\nmodel = data['model']\nfeature = data['feature']","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-23T03:29:10.555111Z","iopub.execute_input":"2024-12-23T03:29:10.555442Z","iopub.status.idle":"2024-12-23T03:29:10.646558Z","shell.execute_reply.started":"2024-12-23T03:29:10.555411Z","shell.execute_reply":"2024-12-23T03:29:10.645220Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"lags_ : pl.DataFrame | None = None\n\n# Replace this function with your inference code.\n# You can return either a Pandas or Polars dataframe, though Polars is recommended.\n# Each batch of predictions (except the very first) must be returned within 1 minute of the batch features being provided.\ndef predict(test: pl.DataFrame, lags: pl.DataFrame | None) -> pl.DataFrame | pd.DataFrame:\n    \"\"\"Make a prediction.\"\"\"\n    # All the responders from the previous day are passed in at time_id == 0. We save them in a global variable for access at every time_id.\n    # Use them as extra features, if you like.\n    global lags_\n    if lags is not None:\n        lags_ = lags\n\n    X = test.select(feature)\n    y_pred = model.predict(X)\n    \n    # Replace this section with your own predictions\n    predictions = \\\n        test.select('row_id').\\\n        with_columns(\n            pl.Series(\n                name   = 'responder_6', \n                values = np.clip(y_pred, a_min = -5, a_max = 5),\n                dtype  = pl.Float64,\n            )\n        )\n\n    \n    if isinstance(predictions, pl.DataFrame):\n        assert predictions.columns == ['row_id', 'responder_6']\n    elif isinstance(predictions, pd.DataFrame):\n        assert (predictions.columns == ['row_id', 'responder_6']).all()\n    else:\n        raise TypeError('The predict function must return a DataFrame')\n    # Confirm has as many rows as the test data.\n    assert len(predictions) == len(test)\n    \n    return predictions","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-23T03:29:10.648249Z","iopub.execute_input":"2024-12-23T03:29:10.648727Z","iopub.status.idle":"2024-12-23T03:29:10.664436Z","shell.execute_reply.started":"2024-12-23T03:29:10.648676Z","shell.execute_reply":"2024-12-23T03:29:10.662987Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"inference_server = kaggle_evaluation.jane_street_inference_server.JSInferenceServer(predict)\n\nif os.getenv('KAGGLE_IS_COMPETITION_RERUN'):\n    inference_server.serve()\nelse:\n    inference_server.run_local_gateway(\n        (\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/test.parquet',\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/lags.parquet',\n        )\n    )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-23T03:29:10.668706Z","iopub.execute_input":"2024-12-23T03:29:10.669055Z","iopub.status.idle":"2024-12-23T03:29:11.139011Z","shell.execute_reply.started":"2024-12-23T03:29:10.669022Z","shell.execute_reply":"2024-12-23T03:29:11.137850Z"}},"outputs":[],"execution_count":null}]}