Skip to content

qwen3-emb 微调的时候,传到info_nce损失的labels是在哪里构建的啊? #8039

@dangweili

Description

@dangweili

Checklist / 检查清单

  • I have searched existing issues, and this is a new question or discussion topic. / 我已经搜索过现有的 issues,确认这是一个新的问题与讨论。

Question Description / 问题描述

有两个问题,
1)qwen3-emb训练的时候,传到info_nce损失的labels是怎么构建的,代码在哪个位置?
2)info_nce损失里面,如果label格式是1001000这种样式的,看起来做样本张量切分的时候,_parse_multi_negative_sentences 的实现有bug,
split_indices = np.array(split_indices) + np.array(list(range(len(split_indices)))) -> split_indices = np.array(split_indices)

@Jintao-Huang

Metadata

Metadata

Assignees

No one assigned

    Labels

    questionFurther information is requested

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions