LLaMA-Factory icon indicating copy to clipboard operation
LLaMA-Factory copied to clipboard

最新代码评估结果generated_predictions.jsonl的顺序乱了,前天还是正常的

Open johnmai-dev opened this issue 1 year ago • 1 comments

Reminder

  • [X] I have read the README and searched the existing issues.

Reproduction

generated_predictions.jsonl的顺序乱了,之前我是按找generated_predictions.jsonl顺序去eval_data数据集中找对应的数据。

Expected behavior

No response

System Info

No response

Others

No response

johnmai-dev avatar May 30 '24 10:05 johnmai-dev

这个commit导致的:https://github.com/hiyouga/LLaMA-Factory/pull/3829/commits/b55fb611c57be03fb38218c7da1d96f6848496ba

https://github.com/hiyouga/LLaMA-Factory/blob/c4f50865ad798e1e99044480e1ab05abefc30224/src/llamafactory/data/loader.py#L122-L124

目前我先自己改回了

    if data_args.max_samples is not None:  # truncate dataset
        num_samples = min(data_args.max_samples, len(dataset))
        dataset = dataset.select(range(num_samples))

@hiyouga

johnmai-dev avatar May 30 '24 12:05 johnmai-dev