day5.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"day5.ipynb","provenance":[{"file_id":"1_w4hLkmQ3sa9z7BhDcmaqjWIM8kBHMBY","timestamp":1583518382546}],"collapsed_sections":["wlH__kh_oI1F"],"mount_file_id":"1ROJat7PVvQ7MDxAw2KPB9RdvB50cnDXY","authorship_tag":"ABX9TyOsa4JNjdLo3UaSoCUaIAFQ"},"kernelspec":{"name":"python3","display_name":"Python 3"},"accelerator":"GPU"},"cells":[{"cell_type":"code","metadata":{"id":"rYHp6LVHXZtn","colab_type":"code","colab":{}},"source":["# upgrade tables bo read_hdf wymaga zaraz\n","!pip install --upgrade tables\n","!pip install eli5\n","!pip install xgboost  #model\n","!pip install hyperopt  #optymalizator"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"2GcQV-BLYCz_","colab_type":"code","colab":{}},"source":["import pandas as pd\n","import numpy as np\n","\n","import matplotlib.pyplot as plt\n","\n","import xgboost as xgb\n","\n","from sklearn.metrics import mean_absolute_error as mae\n","from sklearn.model_selection import cross_val_score\n","\n","from hyperopt import hp, fmin, tpe, STATUS_OK\n","\n","import eli5\n","from eli5.sklearn import PermutationImportance\n"],"execution_count":0,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"wlH__kh_oI1F","colab_type":"text"},"source":["## wczytywanie danych"]},{"cell_type":"code","metadata":{"id":"8Qzgu24VWlX4","colab_type":"code","outputId":"38824d68-42f0-4db0-f708-4ca0531c1c71","executionInfo":{"status":"ok","timestamp":1583570283651,"user_tz":-60,"elapsed":22620,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}},"colab":{"base_uri":"https://localhost:8080/","height":0}},"source":["ls \"/content/drive/My Drive/Colab Notebooks/dw_matrix/matrix_two/dw_matrix_car\""],"execution_count":6,"outputs":[{"output_type":"stream","text":["\u001b[0m\u001b[01;34mdata\u001b[0m/       day2_visualisation.ipynb  day4.ipynb  README.md\n","day1.ipynb  day3_first_model.ipynb    LICENSE\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"OlhjUY7ZXONY","colab_type":"code","outputId":"2acf0c26-3386-4c54-ba43-4d6ab1fc530c","executionInfo":{"status":"ok","timestamp":1583570283658,"user_tz":-60,"elapsed":22622,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}},"colab":{"base_uri":"https://localhost:8080/","height":0}},"source":["cd \"/content/drive/My Drive/Colab Notebooks/dw_matrix/matrix_two/dw_matrix_car/data\""],"execution_count":7,"outputs":[{"output_type":"stream","text":["/content/drive/My Drive/Colab Notebooks/dw_matrix/matrix_two/dw_matrix_car/data\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"HDFRQH0hXQf4","colab_type":"code","outputId":"5517e85d-c05c-4813-b65b-d0d55941aec1","executionInfo":{"status":"ok","timestamp":1583570285947,"user_tz":-60,"elapsed":24904,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}},"colab":{"base_uri":"https://localhost:8080/","height":0}},"source":["ls"],"execution_count":8,"outputs":[{"output_type":"stream","text":["car.h5\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"OaD2wXd7YI8u","colab_type":"code","outputId":"aa012492-2b9e-4bef-b21f-22ee668d9505","executionInfo":{"status":"ok","timestamp":1583570291260,"user_tz":-60,"elapsed":30212,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}},"colab":{"base_uri":"https://localhost:8080/","height":0}},"source":["df = pd.read_hdf('car.h5')\n","df.shape"],"execution_count":9,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(106494, 155)"]},"metadata":{"tags":[]},"execution_count":9}]},{"cell_type":"code","metadata":{"id":"HcmHyBySoRiK","colab_type":"code","outputId":"76b81375-1273-4ce3-cd86-1ba11bede7d6","executionInfo":{"status":"ok","timestamp":1583570291262,"user_tz":-60,"elapsed":30209,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}},"colab":{"base_uri":"https://localhost:8080/","height":0}},"source":["df.columns"],"execution_count":10,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Index(['breadcrumb', 'created_at', 'price_currency', 'price_details',\n","       'price_value', 'seller_address', 'seller_name', 'seller_type',\n","       'feature_czujniki-parkowania-przednie',\n","       'feature_poduszka-powietrzna-chroniąca-kolana',\n","       ...\n","       'param_pearl', 'param_stan', 'param_wersja', 'param_emisja-co2',\n","       'param_body-type', 'param_matowy', 'param_bezwypadkowy',\n","       'param_akryl-(niemetalizowany)', 'param_monthly-payment-value',\n","       'car_id'],\n","      dtype='object', length=155)"]},"metadata":{"tags":[]},"execution_count":10}]},{"cell_type":"markdown","metadata":{"id":"QxAmkDAnp23I","colab_type":"text"},"source":["## features factorization"]},{"cell_type":"code","metadata":{"id":"ifCFr4mCp07n","colab_type":"code","colab":{}},"source":["SUFFIX_CAT = '__cat'\n","for feat in df.columns:\n","  if isinstance(df[feat][0], list): continue\n","  \n","  factorized_values = df[feat].factorize()[0]\n","  if SUFFIX_CAT in feat:\n","    df[feat] = factorized_values\n","  else:\n","    df[feat+SUFFIX_CAT] = factorized_values"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"u285LaOWp8K2","colab_type":"code","outputId":"e8018468-4c59-490b-f3bb-c7166cf55f28","executionInfo":{"status":"ok","timestamp":1583570292087,"user_tz":-60,"elapsed":31024,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}},"colab":{"base_uri":"https://localhost:8080/","height":35}},"source":["cat_feats = [x for x in df.columns if SUFFIX_CAT in x]\n","# bez cen\n","cat_feats = [x for x in cat_feats if 'price' not in x]\n","len(cat_feats)"],"execution_count":12,"outputs":[{"output_type":"execute_result","data":{"text/plain":["151"]},"metadata":{"tags":[]},"execution_count":12}]},{"cell_type":"code","metadata":{"id":"TOGI-O7xEI6v","colab_type":"code","colab":{}},"source":["# zmienne numeryczne lepiej nie faktoryzować, bo tracimy informacje o wzajemnych odległościach\n","df['param_rok-produkcji'] = df['param_rok-produkcji'].map(lambda x: -1 if str(x) == 'None' else int(x))\n","df['param_moc'] = df['param_moc'].map(lambda x: -1 if str(x) == 'None' else int(str(x).split(' ')[0]))\n","df['param_pojemność-skokowa'] = df['param_pojemność-skokowa'].map(lambda x: -1 if str(x) == 'None' else int(str(x).split('cm')[0].replace(' ','')))"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"bnPb48D4ro6_","colab_type":"code","colab":{}},"source":["def run_model(model, feats=cat_feats):\n","  X = df[feats].values\n","  y = df['price_value'].values\n","  scores = cross_val_score(model, X, y, cv=3, scoring='neg_mean_absolute_error')\n","  return np.mean(scores), np.std(scores)"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"D-YdHTUiF0-f","colab_type":"code","colab":{}},"source":["feats = ['param_napęd__cat','param_rok-produkcji','param_stan__cat','param_skrzynia-biegów__cat','param_faktura-vat__cat','param_moc','param_marka-pojazdu__cat','feature_kamera-cofania__cat','param_typ__cat','param_pojemność-skokowa','seller_name__cat','feature_wspomaganie-kierownicy__cat','param_model-pojazdu__cat','param_wersja__cat','param_kod-silnika__cat','feature_system-start-stop__cat','feature_asystent-pasa-ruchu__cat','feature_czujniki-parkowania-przednie__cat','feature_łopatki-zmiany-biegów__cat','feature_regulowane-zawieszenie__cat']"],"execution_count":0,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"9XPongjmyWaH","colab_type":"text"},"source":["##XGBoost"]},{"cell_type":"code","metadata":{"id":"reP_aO5-yYj7","colab_type":"code","outputId":"0e39af46-1dd4-479f-9496-ef43c6b1c41a","executionInfo":{"status":"ok","timestamp":1583570345598,"user_tz":-60,"elapsed":11622,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}},"colab":{"base_uri":"https://localhost:8080/","height":90}},"source":["xgb_params = {\n","    'max_depth': 5,\n","    'n_estimators': 50,\n","    'learning_rate': 0.1,\n","    'random_state': 0,\n","    'seed': 0\n","}\n","run_model(xgb.XGBRegressor(**xgb_params), feats)"],"execution_count":16,"outputs":[{"output_type":"stream","text":["[08:38:55] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n","[08:38:59] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n","[08:39:02] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n"],"name":"stdout"},{"output_type":"execute_result","data":{"text/plain":["(-9569.227198767323, 72.83561801421891)"]},"metadata":{"tags":[]},"execution_count":16}]},{"cell_type":"markdown","metadata":{"id":"D7JiqXG-GLhK","colab_type":"text"},"source":["## Hyperopt"]},{"cell_type":"code","metadata":{"id":"_aRmmxSLF-vc","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":1000},"outputId":"1ae9ef9c-b7d2-4208-9b79-fb0446c39ad7","executionInfo":{"status":"ok","timestamp":1583541882096,"user_tz":-60,"elapsed":687340,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}}},"source":["def obj_func(params):\n","  print(\"Training with params: \")\n","  print(params)\n","\n","  mean_mae, score_std = run_model(xgb.XGBRegressor(**params), feats)\n","\n","  return {'loss': np.abs(mean_mae), 'status': STATUS_OK}\n","\n","#space\n","xgb_reg_params = {\n","    'learning_rate':    hp.choice('learning_rate',      np.arange(0.05, 0.31, 0.05)),\n","    'max_depth':        hp.choice('max_depth',          np.arange(5,16,1, dtype=int)),\n","    'subsample':        hp.quniform('subsample',        0.5, 1, 0.05),\n","    'colsaple_bytree':  hp.quniform('colsample_bytree', 0.5, 1, 0.05),\n","    'objective':        'reg:squarederror',\n","    'n_estimators':     100,\n","    'seed':             0,\n","    'tree_method':      'gpu_hist'\n","}\n","# 'tree_method': 'gpu_hist' użyje GPU o ile zmieniony w runtime\n","\n","## run\n","#best = fmin(obj_func, xgb_reg_params, algo=tpe.suggest, max_evals=25)\n","# wg: https://github.com/hyperopt/hyperopt/issues/530\n","best = fmin(obj_func, xgb_reg_params, algo=tpe.suggest, max_evals=25, return_argmin=False)\n","best"],"execution_count":15,"outputs":[{"output_type":"stream","text":["Training with params: \n","{'colsaple_bytree': 0.8500000000000001, 'learning_rate': 0.3, 'max_depth': 9, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8, 'learning_rate': 0.15000000000000002, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.55, 'learning_rate': 0.15000000000000002, 'max_depth': 7, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 1.0, 'learning_rate': 0.3, 'max_depth': 9, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8, 'learning_rate': 0.15000000000000002, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.55, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9500000000000001, 'learning_rate': 0.25, 'max_depth': 14, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.5, 'learning_rate': 0.2, 'max_depth': 7, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.7000000000000001, 'learning_rate': 0.2, 'max_depth': 6, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.55, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 1.0, 'learning_rate': 0.05, 'max_depth': 15, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.65, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 1.0, 'learning_rate': 0.2, 'max_depth': 14, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.6000000000000001, 'learning_rate': 0.1, 'max_depth': 5, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.6000000000000001, 'learning_rate': 0.25, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.6000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.5, 'learning_rate': 0.2, 'max_depth': 8, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.55, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.75, 'learning_rate': 0.05, 'max_depth': 8, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.6000000000000001, 'learning_rate': 0.05, 'max_depth': 13, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9500000000000001, 'learning_rate': 0.2, 'max_depth': 12, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.7000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.55, 'learning_rate': 0.05, 'max_depth': 12, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.65, 'learning_rate': 0.05, 'max_depth': 11, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.6000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.65, 'learning_rate': 0.25, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9, 'learning_rate': 0.15000000000000002, 'max_depth': 7, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8500000000000001, 'learning_rate': 0.1, 'max_depth': 15, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.7000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9, 'learning_rate': 0.1, 'max_depth': 15, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.7000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9, 'learning_rate': 0.1, 'max_depth': 15, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.75, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8500000000000001, 'learning_rate': 0.1, 'max_depth': 15, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.7000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9, 'learning_rate': 0.1, 'max_depth': 15, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.75, 'tree_method': 'gpu_hist'}\n","100%|██████████| 25/25 [10:11<00:00, 47.31s/it, best loss: 7586.455579423336]\n"],"name":"stdout"},{"output_type":"execute_result","data":{"text/plain":["{'colsaple_bytree': 0.8500000000000001,\n"," 'learning_rate': 0.1,\n"," 'max_depth': 15,\n"," 'n_estimators': 100,\n"," 'objective': 'reg:squarederror',\n"," 'seed': 0,\n"," 'subsample': 0.7000000000000001,\n"," 'tree_method': 'gpu_hist'}"]},"metadata":{"tags":[]},"execution_count":15}]},{"cell_type":"code","metadata":{"id":"D3Z333LhPK9o","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":35},"outputId":"5fdc85d2-11c3-4bc0-d334-610ae63e674e","executionInfo":{"status":"ok","timestamp":1583541934002,"user_tz":-60,"elapsed":739239,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}}},"source":["run_model(xgb.XGBRegressor(**best), feats)"],"execution_count":16,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(-7586.455579423336, 53.57928381493533)"]},"metadata":{"tags":[]},"execution_count":16}]},{"cell_type":"markdown","metadata":{"id":"-yR8T70bH-41","colab_type":"text"},"source":["## dodaję moje modyfikacje do features"]},{"cell_type":"code","metadata":{"id":"w2Q0df6aIiZU","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":35},"outputId":"e33aa954-6344-4e21-dbc2-49c30aa2fab4","executionInfo":{"status":"ok","timestamp":1583570356897,"user_tz":-60,"elapsed":979,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}}},"source":["# EUR\n","df = df[ df['price_currency'] != 'EUR' ]\n","df.shape"],"execution_count":17,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(106290, 309)"]},"metadata":{"tags":[]},"execution_count":17}]},{"cell_type":"code","metadata":{"id":"JiifcleJIlo9","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":320},"outputId":"4c6de488-6446-42e8-8866-c3d2baed2075","executionInfo":{"status":"ok","timestamp":1583570357837,"user_tz":-60,"elapsed":1021,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}}},"source":["#param_prebieg_int\n","df['param_przebieg_int'] = df['param_przebieg'].map(lambda x: int(str(x).replace(\"km\",\"\").replace(\" \", \"\")) if x else 0).astype(int)\n","# odcinam odstające wartości powyżej 1mln\n","df['param_przebieg_int'] = df['param_przebieg_int'].map(lambda x: x if x < 1000000 else -1)  #do modelu nie może być None, więc daję -1 znaczące brak\n","df.param_przebieg_int.hist(bins=100, figsize=(20,5));"],"execution_count":18,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABJAAAAEvCAYAAAAadDsuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n","text/plain":["<Figure size 1440x360 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"id":"IvdKsltvJhws","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":90},"outputId":"3d9e3c6a-0676-46a5-f640-0f60b40adfb3","executionInfo":{"status":"ok","timestamp":1583570411584,"user_tz":-60,"elapsed":11179,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}}},"source":["# biorę tylko te ~20 o największym wpływie po dodani moich kolumn\n","feats = ['param_rok-produkcji','param_moc','param_przebieg_int','param_pojemność-skokowa','param_marka-pojazdu__cat','param_napęd__cat','param_rodzaj-paliwa__cat','param_kod-silnika__cat','param_wersja__cat','param_typ__cat','param_uszkodzony__cat','param_skrzynia-biegów__cat','seller_name__cat','param_bezwypadkowy__cat','param_model-pojazdu__cat','feature_czujniki-parkowania-przednie__cat','param_faktura-vat__cat','param_pojemność-skokowa__cat','feature_asystent-pasa-ruchu__cat','feature_kamera-cofania__cat']\n","run_model(xgb.XGBRegressor(**xgb_params), feats)"],"execution_count":19,"outputs":[{"output_type":"stream","text":["[08:40:01] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n","[08:40:04] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n","[08:40:08] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n"],"name":"stdout"},{"output_type":"execute_result","data":{"text/plain":["(-8888.965998494477, 61.35790253138393)"]},"metadata":{"tags":[]},"execution_count":19}]},{"cell_type":"markdown","metadata":{"id":"vXC5ru6WJ01f","colab_type":"text"},"source":["tearaz hyperoptem optymalizacja parametrów modelu, \n","przed optymalizacją było:\n","(-8888.965998494477, 61.35790253138393)"]},{"cell_type":"code","metadata":{"id":"IoSEiTq5Jx7t","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":1000},"outputId":"bd20e2c5-a97c-4e71-963a-8dbffc3d2723","executionInfo":{"status":"ok","timestamp":1583570913580,"user_tz":-60,"elapsed":315946,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}}},"source":["def obj_func(params):\n","  print(\"Training with params: \")\n","  print(params)\n","\n","  mean_mae, score_std = run_model(xgb.XGBRegressor(**params), feats)\n","\n","  return {'loss': np.abs(mean_mae), 'status': STATUS_OK}\n","\n","#space\n","xgb_reg_params = {\n","    'learning_rate':    hp.choice('learning_rate',      np.arange(0.05, 0.31, 0.05)),\n","    'max_depth':        hp.choice('max_depth',          np.arange(5,16,1, dtype=int)),\n","    'subsample':        hp.quniform('subsample',        0.5, 1, 0.05),\n","    'colsaple_bytree':  hp.quniform('colsample_bytree', 0.5, 1, 0.05),\n","    'objective':        'reg:squarederror',\n","    'n_estimators':     100,\n","    'seed':             0,\n","    'tree_method':      'gpu_hist'\n","}\n","# 'tree_method': 'gpu_hist' użyje GPU o ile zmieniony w runtime\n","\n","## run\n","#best = fmin(obj_func, xgb_reg_params, algo=tpe.suggest, max_evals=25)\n","# wg: https://github.com/hyperopt/hyperopt/issues/530\n","best = fmin(obj_func, xgb_reg_params, algo=tpe.suggest, max_evals=25, return_argmin=False)\n","best"],"execution_count":21,"outputs":[{"output_type":"stream","text":["Training with params: \n","{'colsaple_bytree': 0.75, 'learning_rate': 0.3, 'max_depth': 15, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.5, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.75, 'learning_rate': 0.2, 'max_depth': 14, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.7000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.5, 'learning_rate': 0.25, 'max_depth': 8, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 1.0, 'learning_rate': 0.2, 'max_depth': 9, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.65, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9, 'learning_rate': 0.2, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 1.0, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.55, 'learning_rate': 0.3, 'max_depth': 7, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 1.0, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8, 'learning_rate': 0.3, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.75, 'learning_rate': 0.3, 'max_depth': 9, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.6000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.7000000000000001, 'learning_rate': 0.25, 'max_depth': 7, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.5, 'learning_rate': 0.2, 'max_depth': 6, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8500000000000001, 'learning_rate': 0.25, 'max_depth': 8, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.65, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8500000000000001, 'learning_rate': 0.3, 'max_depth': 11, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.6000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.5, 'learning_rate': 0.2, 'max_depth': 8, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.55, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.65, 'learning_rate': 0.15000000000000002, 'max_depth': 11, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.7000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.7000000000000001, 'learning_rate': 0.2, 'max_depth': 11, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.6000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9500000000000001, 'learning_rate': 0.1, 'max_depth': 9, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9500000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.7000000000000001, 'learning_rate': 0.05, 'max_depth': 9, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.8500000000000001, 'learning_rate': 0.25, 'max_depth': 13, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.6000000000000001, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 1.0, 'learning_rate': 0.1, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.55, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.55, 'learning_rate': 0.05, 'max_depth': 10, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.9, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.6000000000000001, 'learning_rate': 0.15000000000000002, 'max_depth': 12, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.75, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.6000000000000001, 'learning_rate': 0.15000000000000002, 'max_depth': 12, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9500000000000001, 'learning_rate': 0.15000000000000002, 'max_depth': 5, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 0.8, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9500000000000001, 'learning_rate': 0.15000000000000002, 'max_depth': 12, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 1.0, 'tree_method': 'gpu_hist'}\n","Training with params: \n","{'colsaple_bytree': 0.9, 'learning_rate': 0.15000000000000002, 'max_depth': 12, 'n_estimators': 100, 'objective': 'reg:squarederror', 'seed': 0, 'subsample': 1.0, 'tree_method': 'gpu_hist'}\n","100%|██████████| 25/25 [05:15<00:00, 19.38s/it, best loss: 6705.428548550925]\n"],"name":"stdout"},{"output_type":"execute_result","data":{"text/plain":["{'colsaple_bytree': 0.9500000000000001,\n"," 'learning_rate': 0.15000000000000002,\n"," 'max_depth': 12,\n"," 'n_estimators': 100,\n"," 'objective': 'reg:squarederror',\n"," 'seed': 0,\n"," 'subsample': 1.0,\n"," 'tree_method': 'gpu_hist'}"]},"metadata":{"tags":[]},"execution_count":21}]},{"cell_type":"code","metadata":{"id":"4uq5j8psXWxQ","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":35},"outputId":"da631f28-16a3-4bed-a20f-76aecccaedc4","executionInfo":{"status":"ok","timestamp":1583570971396,"user_tz":-60,"elapsed":27897,"user":{"displayName":"Pawel B","photoUrl":"","userId":"07297855696876178085"}}},"source":["run_model(xgb.XGBRegressor(**best), feats)"],"execution_count":22,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(-6705.428548550925, 68.4404322368238)"]},"metadata":{"tags":[]},"execution_count":22}]},{"cell_type":"code","metadata":{"id":"bm7-Sz7-Lv1t","colab_type":"code","colab":{}},"source":[""],"execution_count":0,"outputs":[]}]}