Skip to content

Train_TransE 训练样本量过大会导致 Segmentation fault,怎么解决? #93

Description

@SHK92

我用自己的数据生成的train.txt, entity2id.txt 和 relation2id.txt。train的样本量在1000w级。会报错。train.txt抽一个和FB15k中train.txt样本量相当的sample,是可以正常跑出结果的。但是数据量上去就不得行了。

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions