Skip to content

[BUG] ali-ccp 数据预处理问题 #116

@zhenchuan

Description

@zhenchuan

ali-ccp的数据在处理时
https://github.com/datawhalechina/torch-rechub/blob/main/examples/ranking/data/ali-ccp/preprocess_ali_ccp.py#L45

            for fstr in feat_strs.split('\x01'):
                filed, feat_val = fstr.split('\x02')
                feat, val = feat_val.split('\x03')
                if filed in sparse_columns:
                    feat_dict[filed] = feat
                if filed in dense_columns:
                    feat_dict['D' + filed] = val

这种处理方式在处理数据时对于多值的情况是就是只保留了最后一个值?

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions