'nthread':8,'reg_lambda':3,'reg_alpha':0.01, 'objective':'binary:logistic', 'silent':1, 'subsample':0.60, } class ModelV1_stage2(BaseModel): def build_model(self): return XGBClassifier(params=self.params, num_round=5) # ----- END first stage stacking model ----- if __name__ == "__main__": # Create cv-fold index train = pd.read_csv(INPUT_PATH + 'train.csv') create_cv_id(train, n_folds_ = 5, cv_id_name='cv_id', seed=407) ######## stage1 Models ######### print 'Start stage 1 training' m = ModelV1(name="v1_stage1", flist=FEATURE_LIST_stage1, params = PARAMS_V1, kind = 'st' ) m.run() m = ModelV2(name="v2_stage1", flist=FEATURE_LIST_stage1, params = PARAMS_V2,
'objective':'reg:linear', 'seed':407, 'silent':1, 'subsample':0.8 } class ModelV1_stage2(BaseModel): def build_model(self): return XGBRegressor(params=self.params, num_round=50) # ----- END first stage stacking model ----- if __name__ == "__main__": # Create cv-fold index target = pd.read_csv(INPUT_PATH + 'target.csv') create_cv_id(target, n_folds_ = 5, cv_id_name='cv_id', seed=407) ######## stage1 Models ######### print 'Start stage 1 training' m = ModelV1(name="v1_stage1", flist=FEATURE_LIST_stage1, params = PARAMS_V1, kind = 'st' ) m.run() m = ModelV2(name="v2_stage1", flist=FEATURE_LIST_stage1, params = PARAMS_V2,
'subsample': 0.8 } class ModelV1_stage2(BaseModel): def build_model(self): return XGBRegressor(params=self.params, num_round=50) # ----- END first stage stacking model ----- if __name__ == "__main__": # Create cv-fold index train = pd.read_csv(INPUT_PATH + 'train.csv') create_cv_id(train, n_folds_=5, cv_id_name='cv_id', seed=407) ######## stage1 Models ######### print 'Start stage 1 training' m = ModelV1(name="v1_stage1", flist=FEATURE_LIST_stage1, params=PARAMS_V1, kind='st') m.run() m = ModelV2(name="v2_stage1", flist=FEATURE_LIST_stage1, params=PARAMS_V2, kind='st') m.run()