Skip to content

Commit

Permalink
add llama and nv-embed training and refine evaluation
Browse files Browse the repository at this point in the history
  • Loading branch information
Li-Z-Q committed Nov 18, 2024
1 parent 1fa993f commit 8b14719
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion slm/pipelines/examples/contrastive_training/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -140,7 +140,7 @@ python evaluation/benchmarks.py --model_type bert \
- `query_max_length`: query 的最大长度
- `passage_max_length`: passage 的最大长度
- `evaluate_all`: 是否评估所有的 checkpoint,默认为 False,即只评估指定的 checkpoint
- `checkpoint_dir`: 与`evaluate_all`一起使用
- `checkpoint_dir`: 与`evaluate_all`一起使用

在 MTEB 的 DuRetrieval 上评估,对 RocketQA 的测试代码示例如下:
```
Expand Down

0 comments on commit 8b14719

Please sign in to comment.