def xgboost_make_submission(): train_start_date = '2016-03-10' train_end_date = '2016-04-11' test_start_date = '2016-04-11' test_end_date = '2016-04-16' sub_start_date = '2016-03-15' sub_end_date = '2016-04-16' user_index, training_data, label = make_train_set(train_start_date, train_end_date, test_start_date, test_end_date) X_train, X_test, y_train, y_test = train_test_split(training_data.values, label.values, test_size=0.2, random_state=0) dtrain=xgb.DMatrix(X_train.values, label=y_train) dtest=xgb.DMatrix(X_test.values, label=y_test) param = {'learning_rate' : 0.1, 'n_estimators': 1000, 'max_depth': 3, 'min_child_weight': 5, 'gamma': 0, 'subsample': 1.0, 'colsample_bytree': 0.8, 'scale_pos_weight': 1, 'eta': 0.05, 'silent': 1, 'objective': 'binary:logistic'} num_round = 283 param['nthread'] = 4 #param['eval_metric'] = "auc" plst = param.items() plst += [('eval_metric', 'logloss')] evallist = [(dtest, 'eval'), (dtrain, 'train')] bst=xgb.train(plst, dtrain, num_round, evallist) sub_user_index, sub_trainning_data = make_test_set(sub_start_date, sub_end_date,) sub_trainning_data = xgb.DMatrix(sub_trainning_data.values) y = bst.predict(sub_trainning_data) sub_user_index['label'] = y pred = sub_user_index[sub_user_index['label'] >= 0.03] pred = pred[['user_id', 'sku_id']] pred = pred.groupby('user_id').first().reset_index() pred['user_id'] = pred['user_id'].astype(int) pred.to_csv('./sub/submission.csv', index=False, index_label=False)
def logistic_make_submission(): train_start_date = '2016-03-10' train_end_date = '2016-04-11' test_start_date = '2016-04-11' test_end_date = '2016-04-16' sub_start_date = '2016-03-15' sub_end_date = '2016-04-16' user_index, training_data, label = make_train_set(train_start_date, train_end_date, test_start_date, test_end_date) X_train, X_test, y_train, y_test = train_test_split(training_data.values, label.values, test_size=0.2, random_state=0) y_train = list(map(int, y_train)) # print(np.any(np.isnan(X_train))) # print(np.all(np.isfinite(X_train))) clf = lg() # 使用类,参数全是默认的 clf.fit(X_train, y_train) sub_user_index, sub_trainning_data = make_test_set(sub_start_date, sub_end_date) y_hat = clf.predict(sub_trainning_data.values) sub_user_index['label'] = y_hat pred = sub_user_index[sub_user_index['label'] == 1] pred = pred[['user_id', 'sku_id']] pred = pred.groupby('user_id').first().reset_index() pred['user_id'] = pred['user_id'].astype(int) pred.to_csv('../sub/submissionLOG508.csv', index=False, index_label=False)
def xgboost_make_submission(): train_start_date = '2016-03-10' train_end_date = '2016-04-11' test_start_date = '2016-04-11' test_end_date = '2016-04-16' #测试集构建,根据测试集特征数据集预测后五天的label sub_start_date = '2016-03-15' sub_end_date = '2016-04-16' user_index, training_data, label = make_train_set(train_start_date, train_end_date, test_start_date, test_end_date) X_train, X_test, y_train, y_test = train_test_split(training_data.values, label.values, test_size=0.2, random_state=0) dtrain=xgb.DMatrix(X_train, label=y_train) dtest=xgb.DMatrix(X_test, label=y_test) param = {'learning_rate' : 0.1, 'n_estimators': 1000, 'max_depth': 3, 'min_child_weight': 5, 'gamma': 0, 'subsample': 1.0, 'colsample_bytree': 0.8, 'scale_pos_weight': 1, 'eta': 0.05, 'silent': 1, 'objective': 'binary:logistic'} num_round = 283 param['nthread'] = 4 #param['eval_metric'] = "auc" plst = param.items() plst += [('eval_metric', 'logloss')] evallist = [(dtest, 'eval'), (dtrain, 'train')] bst=xgb.train(plst, dtrain, num_round, evallist) sub_user_index, sub_trainning_data = make_test_set(sub_start_date, sub_end_date,) sub_trainning_data = xgb.DMatrix(sub_trainning_data.values) #预测得到用户-商品对数据标签 y = bst.predict(sub_trainning_data) sub_user_index['label'] = y #将用户-商品对出现概率大于0.03的拿出来 pred = sub_user_index[sub_user_index['label'] >= 0.03] pred = pred[['user_id', 'sku_id']] pred = pred.groupby('user_id').first().reset_index() pred['user_id'] = pred['user_id'].astype(int) pred.to_csv('./sub/submission.csv', index=False, index_label=False)
def xgboost_make_submission(): train_start_date = '2016-03-10' train_end_date = '2016-04-11' test_start_date = '2016-04-11' test_end_date = '2016-04-16' sub_start_date = '2016-03-15' sub_end_date = '2016-04-16' user_index, training_data, label = make_train_set(train_start_date, train_end_date, test_start_date, test_end_date) X_train, X_test, y_train, y_test = train_test_split( training_data.values, label.values, test_size=0.2, random_state=0) # select some features dtrain = xgb.DMatrix(X_train, label=y_train) dtest = xgb.DMatrix(X_test, label=y_test) # don't use these param = { 'learning_rate': 0.05, 'n_estimators': 1000, 'max_depth': 5, 'min_child_weight': 1, 'gamma': 0, 'subsample': 1, 'colsample_bytree': 0.8, 'scale_pos_weight': 1, 'eta': 0.05, 'silent': 1, 'objective': 'binary:logistic' } num_round = 20 param['nthread'] = 5 #param['eval_metric'] = "auc" plst = param.items() plst = list(plst) plst += [('eval_metric', 'auc')] evallist = [(dtest, 'eval'), (dtrain, 'train')] bst = xgb.train(plst, dtrain, num_round, evallist) # make test data sub_user_index, sub_trainning_data = make_test_set(sub_start_date, sub_end_date) sub_trainning_data = xgb.DMatrix( sub_trainning_data.values ) # predict this subdata,the DMatrix Object is array y_hat = bst.predict(sub_trainning_data) sub_user_index['label'] = y_hat pred = sub_user_index[sub_user_index['label'] >= 0.05] pred = pred[['user_id', 'sku_id']] pred = pred.groupby('user_id').first().reset_index() pred['user_id'] = pred['user_id'].astype(int) pred.to_csv('../sub/submission424.csv', index=False, index_label=False)
def gbdt_make_submission(): train_start_date = '2016-03-10' train_end_date = '2016-04-11' test_start_date = '2016-04-11' test_end_date = '2016-04-16' sub_start_date = '2016-03-15' sub_end_date = '2016-04-16' user_index, training_data, label = make_train_set(train_start_date, train_end_date, test_start_date, test_end_date) training_data = training_data.fillna(0) print(training_data.info()) X_train, X_test, y_train, y_test = train_test_split(training_data.values, label.values, test_size=0.2, random_state=0) # X_train = X_train.astype(int) y_train = list(map(int, y_train)) param = { 'n_estimators': 1200, 'max_depth': 3, 'subsample': 1.0, 'learning_rate': 0.01, 'min_samples_leaf': 1, 'random_state': 3, 'max_features': 0.8 } clf = gbdt(param) clf.fit(X_train, y_train) sub_user_index, sub_trainning_data = make_test_set(sub_start_date, sub_end_date) sub_trainning_data = sub_trainning_data.fillna(0) y_hat = clf.predict(sub_trainning_data.values) sub_user_index['label'] = y_hat pred = sub_user_index[sub_user_index['label'] == 1] pred = pred[['user_id', 'sku_id']] pred = pred.groupby('user_id').first().reset_index() pred['user_id'] = pred['user_id'].astype(int) pred.to_csv('../sub/submissionGBDT508.csv', index=False, index_label=False)
def xgboost_make_submission(): sub_start_date = '2016-03-15' sub_end_date = '2016-04-16' if os.path.exists('./cache/bstmodel.bin'): bst = xgb.Booster({'ntheard': 4}) bst.load_model('./cache/bstmodel.bin') else: bst = xgboost_train() sub_user_index, sub_trainning_data = make_test_set( sub_start_date, sub_end_date, ) sub_trainning_data = xgb.DMatrix(sub_trainning_data.values) y = bst.predict(sub_trainning_data) sub_user_index['label'] = y pred = sub_user_index[sub_user_index['label'] >= 0.03] pred = pred[['user_id', 'sku_id']] pred = pred.groupby('user_id').first().reset_index() pred['user_id'] = pred['user_id'].astype(int) pred.to_csv('./sub/submission.csv', index=False, index_label=False)
def xgboost_report_submission(): train_start_date = '2016-03-08' train_end_date = '2016-04-09' result_start_date = '2016-04-09' result_end_date = '2016-04-14' valid_start_date = '2016-03-01' valid_end_date = '2016-04-02' valid_result_start_date = '2016-04-02' valid_result_end_date = '2016-04-07' test_start_date = '2016-03-15' test_end_date = '2016-04-16' user_index, training_data, label = make_train_set(train_start_date, train_end_date, result_start_date, result_end_date) x_train, x_test, y_train, y_test = train_test_split(training_data.values, label.values, test_size=0.2, random_state=0) dtrain = xgb.DMatrix(x_train, label=y_train) dtest = xgb.DMatrix(x_test, label=y_test) param = {'learning_rate': 0.1, 'n_estimators': 1000, 'max_depth': 3, 'min_child_weight': 5, 'gamma': 0, 'subsample': 1.0, 'colsample_bytree': 0.8, 'scale_pos_weight': 1, 'eta': 0.05, 'silent': 1, 'objective': 'binary:logistic'} num_round = 283 param['nthread'] = 4 #param['eval_metric'] = "auc" plst = param.items() plst += [('eval_metric', 'logloss')] evallist = [(dtest, 'eval'), (dtrain, 'train')] bst = xgb.train(plst, dtrain, num_round, evallist) # Report with validation set valid_user_index, valid_trainning_date = make_test_set(valid_start_date, valid_end_date) valid_trainning_date = xgb.DMatrix(valid_trainning_date.values) pred_y = bst.predict(valid_trainning_date) valid_pred = valid_user_index.copy() valid_pred['label'] = pred_y valid_pred = valid_pred[valid_pred['label'] >= 0.014] valid_pred = valid_pred.sort_values('label', ascending=False).groupby('user_id').first().reset_index() valid_true = get_true(valid_result_start_date, valid_result_end_date) report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.016] print 0.016 report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.018] print 0.018 report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.02] print 0.02 report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.022] print 0.022 report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.024] print 0.024 report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.026] print 0.026 report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.028] print 0.028 report(valid_pred, valid_true) valid_pred = valid_pred[valid_pred['label'] >= 0.03] print 0.03 report(valid_pred, valid_true)
def xgboost_make_submission(): train_start_date = '2016-03-31' train_end_date = '2016-04-10' test_start_date = '2016-04-10' test_end_date = '2016-04-16' sub_start_date = '2016-04-06' sub_end_date = '2016-04-16' user_index, training_data, label = make_train_set(train_start_date, train_end_date, test_start_date, test_end_date) list_of_train = list(training_data.columns) print len(list_of_train) X_train, X_test, y_train, y_test = train_test_split(training_data.values, label.values, test_size=0.2, random_state=0) dtrain = xgb.DMatrix(X_train, label=y_train) dtest = xgb.DMatrix(X_test, label=y_test) param = { 'learning_rate': 0.1, 'n_estimators': 1000, 'max_depth': 3, 'min_child_weight': 5, 'gamma': 0, 'subsample': 1.0, 'colsample_bytree': 0.8, 'scale_pos_weight': 1, 'eta': 0.05, 'silent': 1, 'objective': 'binary:logistic' } # num_round = 345 num_round = 511 # param['nthread'] = 8 param['eval_metric'] = "auc" plst = param.items() plst += [('eval_metric', 'logloss')] evallist = [(dtest, 'eval'), (dtrain, 'train')] bst = xgb.train(plst, dtrain, num_round, evallist, early_stopping_rounds=10) importance = bst.get_fscore() print importance feat_importances = [] for ft, score in importance.iteritems(): ft = ft.split('f')[1] feat_importances.append({'Feature': ft, 'Importance': score}) feat_importances = pd.DataFrame(feat_importances) feat_importances = feat_importances.sort_values( by='Importance', ascending=False).reset_index(drop=True) new_columns = [] for index in list(feat_importances['Feature']): index = int(index) # feat_importances[index]['Feature'] = new_columns.append(list_of_train[index]) name_of = pd.DataFrame({'new': new_columns}) feat_importances = pd.concat([feat_importances, name_of], axis=1) feat_importances.to_csv('./sub/fecure.csv') sub_user_index, sub_trainning_data = make_test_set(sub_start_date, sub_end_date) sub_trainning_data = xgb.DMatrix(sub_trainning_data.values) y_label = bst.predict(xgb.DMatrix(X_test)) fpr, tpr, threasholds = roc_curve(y_test, y_label, pos_label=2) print fpr, tpr, threasholds # print auc(fpr,tpr) # plt.plot(threasholds,fpr) # plt.show() y = bst.predict(sub_trainning_data) sub_user_index['label'] = y # print np.median(y) # print sub_user_index # pred = sub_user_index.groupby('user_id').max().reset_index() # print pred pred = sub_user_index[sub_user_index['label'] >= 0.04] pred = pred[['user_id', 'sku_id']] pred = pred.groupby('user_id').max().reset_index() pred['user_id'] = pred['user_id'].astype(int) pred.to_csv('./sub/submission.csv', index=False, index_label=False) buy_cate_8 = np.load('./unique_8/user_id_unique.npy') pred = pred[~pred['user_id'].isin(buy_cate_8)] pred.to_csv('./sub/submission_unique.csv', index=False, index_label=False)