python - teradataml：如何使用 teradataml 生成 xgboost 模型？

Question

我正在尝试使用 teradataml 生成 XGBoost 模型。此示例将展示如何在 teradataml 中使用 xgboost。假设训练和测试数据存在于 teradata 数据库中。

标签： pythonteradata

Answer 1

1. 导入库

   import teradataml
   from teradataml.context.context import *
   from teradataml.dataframe.dataframe import DataFrame
   from teradataml.analytics.mle.XGBoost import XGBoost
   from teradataml.analytics.mle.XGBoostPredict import XGBoostPredict
   

2. 创建上下文

   housing_train_binary = DataFrame.from_table("housing_train_binary")
   housing_test_binary = DataFrame.from_table("housing_test_binary")
   

3. 生成模型

   xgboostmodel = XGBoost(data=housing_train_binary,
                        id_column='sn',
                        formula=" homestyle ~ driveway +  recroom +  fullbase +  gashw +  airco +  prefarea ",
                        num_boosted_trees=2,
                        loss_function='SOFTMAX',
                        prediction_type='CLASSIFICATION',
                        reg_lambda=1.0,
                        shrinkage_factor=0.1,
                        column_subsampling=1.0,
                        iter_num=10,
                        min_node_size=1,
                        max_depth=12,
                        variance=0.0,
                        seed=1,
                        data_sequence_column=['sn', 'homestyle']
                        );
   

4. 预测

   xgpredict = XGBoostPredict(newdata=housing_test_binary, object=result, object_order_column=['tree_id', 'iter','class_num']);