{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"#This Kernel is done after going through all the ML courses on Kaggle Learn\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport os\nfrom sklearn.ensemble import RandomForestRegressor\nfrom sklearn.pipeline import make_pipeline\n\nfrom sklearn.pipeline import Pipeline\nfrom sklearn.model_selection import GridSearchCV\n\nfrom sklearn.preprocessing import Imputer\nfrom sklearn.model_selection import train_test_split\nfrom xgboost import XGBRegressor","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"c20b2f8ef9f98635306a4053e1e7482ac3889804"},"cell_type":"code","source":"# List of all the functions that we are going to use\n\ndef get_rmse(y_predicted,y_real):\n    return np.mean(np.sqrt((np.log(y_predicted)-np.log(y_real))**2))","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"train_data = pd.read_csv('../input/train.csv')\ntest_data = pd.read_csv('../input/test.csv')\n\ny = train_data.SalePrice","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"a3846cf288f83acd2c4c7232c93d30cad4ff87c8"},"cell_type":"code","source":"train_data.head()\ntrain_data.columns","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"7cb7ee5438358ab26e38486b5130af5bd84f2d74"},"cell_type":"code","source":"#One-hot encoding (using categorical data)\n\ncols_with_missing = [col for col in train_data.columns \n                                 if train_data[col].isnull().any()]                                  \ncandidate_train_predictors = train_data.drop(['Id', 'SalePrice'] + cols_with_missing, axis=1)\ncandidate_test_predictors = test_data.drop(['Id'] + cols_with_missing, axis=1)\n\n# \"cardinality\" means the number of unique values in a column.\n# We use it as our only way to select categorical columns here. This is convenient, though\n# a little arbitrary.\nlow_cardinality_cols = [cname for cname in candidate_train_predictors.columns if \n                                candidate_train_predictors[cname].nunique() < 10 and\n                                candidate_train_predictors[cname].dtype == \"object\"]\nnumeric_cols = [cname for cname in candidate_train_predictors.columns if \n                                candidate_train_predictors[cname].dtype in ['int64', 'float64']]\nmy_cols = low_cardinality_cols + numeric_cols\ntrain_predictors = candidate_train_predictors[my_cols]\ntest_predictors = candidate_test_predictors[my_cols]\n\none_hot_encoded_training_predictors = pd.get_dummies(train_predictors)\none_hot_encoded_test_predictors = pd.get_dummies(test_predictors)\nfinal_train, final_test = one_hot_encoded_training_predictors.align(one_hot_encoded_test_predictors,\n                                                                    join='left', \n                                                                    axis=1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"f15433616feda4bf8415a87958d1e35b76aa778d"},"cell_type":"code","source":"X = np.array(final_train)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"a6847c3e0848a3020cbeb574890cc25fb8a4f104"},"cell_type":"code","source":"train_X, val_X, train_y, val_y = train_test_split(X, y)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"aba9511c2d1d08cf984fe041063b58b7c5656a29"},"cell_type":"code","source":"my_pipeline = Pipeline([('imputer', Imputer()), ('xgbrg', XGBRegressor())])\n\nparam_grid = {\n    \"xgbrg__n_estimators\": [10, 20, 30, 40, 50, 60, 70, 80, 80],\n    \"xgbrg__learning_rate\": [0.1, 0.125, 0.15, 0.175, 0.2, 0.225, 0.25, 0.275, 0.3],\n}\n\nfit_params = {\"xgbrg__eval_set\": [(val_X, val_y)], \n              \"xgbrg__early_stopping_rounds\": 10, \n              \"xgbrg__verbose\": False} ;\n\n#5-fold cross validation by passing the argument cv=5\nsearchCV = GridSearchCV(my_pipeline, cv=5, param_grid=param_grid, fit_params=fit_params);\nsearchCV.fit(train_X, train_y)  ;","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"1afe1bc9a86f73ab359e5e1f8e86ccb472f8b0da"},"cell_type":"code","source":"searchCV.best_params_ ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"c777fb714c7cb5efc06f52ae7432b1f5c903a7e8"},"cell_type":"code","source":"best_learn_rate = searchCV.best_params_['xgbrg__learning_rate']\nbest_nb_est = searchCV.best_params_ ['xgbrg__n_estimators']","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"8bc32670a91ca3b4cda936231b9d105d1517d6bf","scrolled":false},"cell_type":"code","source":"my_pipeline = make_pipeline(Imputer(), XGBRegressor(n_estimators=best_nb_est, learning_rate = best_learn_rate))\n\nmy_pipeline.fit(X,y)\ntrain_y_predicted = my_pipeline.predict(train_X)\nval_y_predicted = my_pipeline.predict(val_X)\nprint('Score on training set:',get_rmse(train_y_predicted,train_y))\nprint('Score on validation set:',get_rmse(val_y_predicted,val_y))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"1567da08e533965fa854c031246b1ded30a4aac0"},"cell_type":"code","source":"predictions = my_pipeline.predict(final_test)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"0786b4da8b28adb93aeb98ce7315758d65564aa5"},"cell_type":"code","source":"output = pd.DataFrame({'Id': test_data.Id,\n                       'SalePrice': predictions})\n\noutput.to_csv('submission.csv', index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"88e91a4273914493d8417a617c1f627407039c8d"},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}