{"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":38128,"databundleVersionId":4230952,"sourceType":"competition"}],"dockerImageVersionId":30732,"isInternetEnabled":true,"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        \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","execution":{"iopub.status.busy":"2024-06-14T02:25:14.589277Z","iopub.execute_input":"2024-06-14T02:25:14.589965Z","iopub.status.idle":"2024-06-14T02:25:14.967844Z","shell.execute_reply.started":"2024-06-14T02:25:14.589933Z","shell.execute_reply":"2024-06-14T02:25:14.966950Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os, gc, pickle\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nimport numpy as np\nfrom colorama import Fore, Back, Style\nfrom matplotlib.ticker import MaxNLocator\nimport warnings\nwarnings.simplefilter('ignore')\nfrom sklearn.base import BaseEstimator, TransformerMixin\nfrom sklearn.model_selection import KFold\nfrom sklearn.preprocessing import StandardScaler, scale\nfrom sklearn.decomposition import PCA\nfrom sklearn.dummy import DummyRegressor\nfrom sklearn.pipeline import make_pipeline, Pipeline\nfrom sklearn.linear_model import Ridge\nfrom sklearn.linear_model import Lasso\nfrom sklearn.multioutput import MultiOutputRegressor\nimport lightgbm as lgb\nfrom sklearn.compose import TransformedTargetRegressor\nfrom sklearn.metrics import mean_squared_error\n\nDATA_DIR = \"/kaggle/input/open-problems-multimodal/\"\nFP_CELL_METADATA = os.path.join(DATA_DIR,\"metadata.csv\")\n\nFP_CITE_TRAIN_INPUTS = os.path.join(DATA_DIR,\"train_cite_inputs.h5\")\nFP_CITE_TRAIN_TARGETS = os.path.join(DATA_DIR,\"train_cite_targets.h5\")\nFP_CITE_TEST_INPUTS = os.path.join(DATA_DIR,\"test_cite_inputs.h5\")\n\nFP_MULTIOME_TRAIN_INPUTS = os.path.join(DATA_DIR,\"train_multi_inputs.h5\")\nFP_MULTIOME_TRAIN_TARGETS = os.path.join(DATA_DIR,\"train_multi_targets.h5\")\nFP_MULTIOME_TEST_INPUTS = os.path.join(DATA_DIR,\"test_multi_inputs.h5\")\n\nFP_SUBMISSION = os.path.join(DATA_DIR,\"sample_submission.csv\")\nFP_EVALUATION_IDS = os.path.join(DATA_DIR,\"evaluation_ids.csv\")","metadata":{"execution":{"iopub.status.busy":"2024-06-14T02:25:15.684555Z","iopub.execute_input":"2024-06-14T02:25:15.685025Z","iopub.status.idle":"2024-06-14T02:25:19.608499Z","shell.execute_reply.started":"2024-06-14T02:25:15.684996Z","shell.execute_reply":"2024-06-14T02:25:19.607532Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"!pip install --quiet tables","metadata":{"execution":{"iopub.status.busy":"2024-06-14T02:25:21.801424Z","iopub.execute_input":"2024-06-14T02:25:21.802543Z","iopub.status.idle":"2024-06-14T02:25:34.314924Z","shell.execute_reply.started":"2024-06-14T02:25:21.802501Z","shell.execute_reply":"2024-06-14T02:25:34.313694Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_cell = pd.read_csv(FP_CELL_METADATA)\ndf_cell_cite = df_cell[df_cell.technology==\"citeseq\"]\ndf_cell_multi = df_cell[df_cell.technology==\"multiome\"]\ndf_cell_cite.shape, df_cell_multi.shape","metadata":{"execution":{"iopub.status.busy":"2024-06-14T02:25:34.317031Z","iopub.execute_input":"2024-06-14T02:25:34.317339Z","iopub.status.idle":"2024-06-14T02:25:34.818533Z","shell.execute_reply.started":"2024-06-14T02:25:34.317311Z","shell.execute_reply":"2024-06-14T02:25:34.817592Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n# Preprocessing\ncol_start = 10000\n\nclass PreprocessCiteseq(BaseEstimator, TransformerMixin):\n    columns_to_use = 13000\n    \n    @staticmethod\n    def take_column_subset(X):\n        return X[:,-(PreprocessCiteseq.columns_to_use+col_start):-col_start]\n    \n    def transform(self, X):\n        print(X.shape)\n        X = X[:,~self.all_zero_columns]\n        print(X.shape)\n        X = PreprocessCiteseq.take_column_subset(X) # use only a part of the columns\n        print(X.shape)\n        gc.collect()\n\n        X = self.pca.transform(X)\n        print(X.shape)\n        return X\n\n    def fit_transform(self, X):\n        gc.collect()\n        print(X.shape)\n        self.all_zero_columns = (X == 0).all(axis=0)\n        X = X[:,~self.all_zero_columns]\n        print(X.shape)\n        X = PreprocessCiteseq.take_column_subset(X) # use only a part of the columns\n        print(X.shape)\n        gc.collect()\n\n        self.pca = PCA(n_components=100, copy=False, random_state=1)\n        X = self.pca.fit_transform(X)\n#         plt.plot(self.pca.explained_variance_ratio_.cumsum())\n#         plt.title(\"Cumulative explained variance ratio\")\n#         plt.gca().xaxis.set_major_locator(MaxNLocator(integer=True))\n#         plt.xlabel('PCA component')\n#         plt.ylabel('Cumulative explained variance ratio')\n#         plt.show()\n        print(X.shape)\n        return X\n\npreprocessor = PreprocessCiteseq()\n\ncite_train_x = None\ncite_train_x = preprocessor.fit_transform(pd.read_hdf(FP_CITE_TRAIN_INPUTS).values)\n\ncite_train_y = pd.read_hdf(FP_CITE_TRAIN_TARGETS).values\nprint(cite_train_y.shape)","metadata":{"execution":{"iopub.status.busy":"2024-06-14T02:25:34.819630Z","iopub.execute_input":"2024-06-14T02:25:34.819915Z","iopub.status.idle":"2024-06-14T02:26:52.649189Z","shell.execute_reply.started":"2024-06-14T02:25:34.819891Z","shell.execute_reply":"2024-06-14T02:26:52.648272Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pip install lightgbm xgboost catboost scikit-learn \n","metadata":{"execution":{"iopub.status.busy":"2024-06-13T22:48:55.841584Z","iopub.execute_input":"2024-06-13T22:48:55.842037Z","iopub.status.idle":"2024-06-13T22:49:07.935314Z","shell.execute_reply.started":"2024-06-13T22:48:55.842003Z","shell.execute_reply":"2024-06-13T22:49:07.934166Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"","metadata":{}},{"cell_type":"code","source":"from sklearn.neighbors import KNeighborsRegressor\nfrom sklearn.metrics import mean_squared_error\nimport gc\n\n# Function to train and evaluate a model\ndef train_and_evaluate_model(model, train_x, train_y):\n    model.fit(train_x, train_y)\n    y_pred = model.predict(train_x)\n    mse = mean_squared_error(train_y, y_pred)\n    return mse, model\n\n# Placeholder for training data\n# cite_train_x and cite_train_y should be your actual training data\n# For demonstration purposes, assume they are already loaded\n\nmodel = KNeighborsRegressor(n_neighbors=5)\n\nprint('Training K-Nearest Neighbors model...')\nmse, trained_model = train_and_evaluate_model(model, cite_train_x, cite_train_y)\nprint(f'K-Nearest Neighbors MSE: {mse}')\n\n# Clean up to free memory\ngc.collect()\n","metadata":{"execution":{"iopub.status.busy":"2024-06-14T02:26:52.651569Z","iopub.execute_input":"2024-06-14T02:26:52.652228Z","iopub.status.idle":"2024-06-14T02:27:10.273895Z","shell.execute_reply.started":"2024-06-14T02:26:52.652190Z","shell.execute_reply":"2024-06-14T02:27:10.273072Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import r2_score","metadata":{"execution":{"iopub.status.busy":"2024-06-14T02:27:10.274969Z","iopub.execute_input":"2024-06-14T02:27:10.275243Z","iopub.status.idle":"2024-06-14T02:27:10.280399Z","shell.execute_reply.started":"2024-06-14T02:27:10.275219Z","shell.execute_reply":"2024-06-14T02:27:10.279557Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.neighbors import KNeighborsRegressor\nfrom sklearn.metrics import r2_score\nimport gc\n\n# Function to train and evaluate a model\ndef train_and_evaluate_model(model, train_x, train_y):\n    model.fit(train_x, train_y)\n    y_pred = model.predict(train_x)\n    r2_scores = r2_score(train_y, y_pred, multioutput='raw_values')\n    r2_total = r2_score(train_y, y_pred)\n    mse =mean_squared_error(train_y, y_pred, multioutput='raw_values')\n    mse_total =mean_squared_error(train_y, y_pred)\n    return r2_scores, r2_total, mse, mse_total, model\n\n# Placeholder for training data\n# cite_train_x and cite_train_y should be your actual training data\n# For demonstration purposes, assume they are already loaded\n\nmodel = KNeighborsRegressor(n_neighbors=5, metric='manhattan')\n\nprint('Training K-Nearest Neighbors model...')\nr2_scores, r2_total, mse, mse_total, trained_model = train_and_evaluate_model(model, cite_train_x, cite_train_y)\n\n# Print R2 scores for each target\nfor i, r2 in enumerate(r2_scores):\n    print(f'R^2 Score for Target {i}: {r2}')\nfor i, mse in enumerate(mse):\n    print(f'R^2 Score for Target {i}: {mse}')\nprint('R^2 score total:', r2_total)\nprint('MSE total :', mse_total)\n# Clean up to free memory\ngc.collect()\n","metadata":{"execution":{"iopub.status.busy":"2024-06-14T02:34:22.867420Z","iopub.execute_input":"2024-06-14T02:34:22.868047Z","iopub.status.idle":"2024-06-14T02:37:57.208455Z","shell.execute_reply.started":"2024-06-14T02:34:22.868016Z","shell.execute_reply":"2024-06-14T02:37:57.207598Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import lightgbm as lgb\nfrom sklearn.multioutput import MultiOutputRegressor\nfrom sklearn.metrics import r2_score, mean_squared_error\nimport gc\n\n# Function to train and evaluate a model\ndef train_and_evaluate_model(model, train_x, train_y):\n    model.fit(train_x, train_y)\n    y_pred = model.predict(train_x)\n    r2_scores = r2_score(train_y, y_pred, multioutput='raw_values')\n    mse_scores = mean_squared_error(train_y, y_pred, multioutput='raw_values')\n    r2_total = r2_score(train_y, y_pred)\n    mse_total = mean_squared_error(train_y, y_pred)\n    return r2_scores, r2_total, mse_scores, mse_total, model\n\n# Placeholder for training data\n# cite_train_x and cite_train_y should be your actual training data\n# For demonstration purposes, assume they are already loaded\n\n# Define LightGBM parameters\nparams = {\n    'n_estimators': 1000,\n    'learning_rate': 0.1,\n    # Add more parameters as needed\n}\n\n# Initialize LightGBM regressor\nlgb_model = lgb.LGBMRegressor(**params)\n\n# Wrap LightGBM model in MultiOutputRegressor for multi-output support\nmulti_output_model = MultiOutputRegressor(lgb_model)\n\nprint('Training LightGBM model for multi-output...')\nr2_scores, r2_total, mse_scores, mse_total, trained_model = train_and_evaluate_model(multi_output_model, cite_train_x, cite_train_y)\n\n# Print R2 scores and MSE for each target\nfor i, r2 in enumerate(r2_scores):\n    print(f'R^2 Score for Target {i}: {r2}')\n    print(f'MSE for Target {i}: {mse_scores[i]}')\n\nprint('R^2 score total:', r2_total)\nprint('MSE total :', mse_total)\n\n# Clean up to free memory\ngc.collect()\n","metadata":{"execution":{"iopub.status.busy":"2024-06-13T22:50:06.909926Z","iopub.execute_input":"2024-06-13T22:50:06.910331Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import xgboost as xgb\nfrom sklearn.multioutput import MultiOutputRegressor\nfrom sklearn.metrics import r2_score, mean_squared_error\nimport gc\n\n# Function to train and evaluate a model\ndef train_and_evaluate_model(model, train_x, train_y):\n    model.fit(train_x, train_y)\n    y_pred = model.predict(train_x)\n    r2_scores = r2_score(train_y, y_pred, multioutput='raw_values')\n    mse_scores = mean_squared_error(train_y, y_pred, multioutput='raw_values')\n    r2_total = r2_score(train_y, y_pred)\n    mse_total = mean_squared_error(train_y, y_pred)\n    return r2_scores, r2_total, mse_scores, mse_total, model\n\n# Placeholder for training data\n# cite_train_x and cite_train_y should be your actual training data\n# For demonstration purposes, assume they are already loaded\n\n# Define XGBoost parameters\nparams = {\n    'objective': 'reg:squarederror',  # for regression task\n    'n_estimators': 1000,\n    'learning_rate': 0.1,\n    # Add more parameters as needed\n}\n\n# Initialize XGBoost regressor\nxgb_model = xgb.XGBRegressor(**params)\n\n# Wrap XGBoost model in MultiOutputRegressor for multi-output support\nmulti_output_model = MultiOutputRegressor(xgb_model)\n\nprint('Training XGBoost model for multi-output...')\nr2_scores, r2_total, mse_scores, mse_total, trained_model = train_and_evaluate_model(multi_output_model, cite_train_x, cite_train_y)\n\n# Print R2 scores and MSE for each target\nfor i, r2 in enumerate(r2_scores):\n    print(f'R^2 Score for Target {i}: {r2}')\n    print(f'MSE for Target {i}: {mse_scores[i]}')\n\nprint('R^2 score total:', r2_total)\nprint('MSE total :', mse_total)\n\n# Clean up to free memory\ngc.collect()\n","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from catboost import CatBoostRegressor\nfrom sklearn.multioutput import MultiOutputRegressor\nfrom sklearn.metrics import r2_score, mean_squared_error\nimport gc\n\n# Function to train and evaluate a model\ndef train_and_evaluate_model(model, train_x, train_y):\n    model.fit(train_x, train_y)\n    y_pred = model.predict(train_x)\n    r2_scores = r2_score(train_y, y_pred, multioutput='raw_values')\n    mse_scores = mean_squared_error(train_y, y_pred, multioutput='raw_values')\n    r2_total = r2_score(train_y, y_pred)\n    mse_total = mean_squared_error(train_y, y_pred)\n    return r2_scores, r2_total, mse_scores, mse_total, model\n\n# Placeholder for training data\n# cite_train_x and cite_train_y should be your actual training data\n# For demonstration purposes, assume they are already loaded\n\n# Define CatBoost parameters\nparams = {\n    'iterations': 1000,\n    'learning_rate': 0.1,\n    'loss_function': 'RMSE',  # for regression task\n    # Add more parameters as needed\n}\n\n# Initialize CatBoost regressor\ncatboost_model = CatBoostRegressor(**params)\n\n# Wrap CatBoost model in MultiOutputRegressor for multi-output support\nmulti_output_catboost = MultiOutputRegressor(catboost_model)\n\nprint('Training CatBoost model with MultiOutputRegressor...')\nr2_scores, r2_total, mse_scores, mse_total, trained_model = train_and_evaluate_model(multi_output_catboost, cite_train_x, cite_train_y)\n\n# Print R2 scores and MSE for each target\nfor i, r2 in enumerate(r2_scores):\n    print(f'R^2 Score for Target {i}: {r2}')\n    print(f'MSE for Target {i}: {mse_scores[i]}')\n\nprint('R^2 score total:', r2_total)\nprint('MSE total :', mse_total)\n\n# Clean up to free memory\ngc.collect()\n","metadata":{},"execution_count":null,"outputs":[]}]}