Info | 说明
Download the zip files and build models.
zip压缩文件可用于模型开发。
Extra notes | 备注:
- The first column in each dataset is the target variable, prefixed with "target_", e.g., "target_survived" or "target_Class".
每个数据集中第1列为目标变量,以 "target_" 开头,例如,"target_survived" 或者 "target_Class"。 - The original class variable in some datasets may have more than 2 values, we have binarized them for binary classification modeling. Read source code for the treatments.
某些数据集的原始分类变量可能含有2个以上的取值,我们进行了二值化处理使数据集能够进行二分类建模。源代码中有详细处理逻辑。 - Due to the lack of domain knowledge, data treatments may not be best-practices. Please create a new zipped data file by modifying the source code.
由于缺少相关领域知识,某些数据处理步骤可能并非最佳实践。请修改源代码后重新生成压缩的数据文件。
Datasets | 数据集