Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

loss 下降到2.7就不下降了 #744

Open
Alexender-Ye opened this issue Apr 30, 2024 · 1 comment
Open

loss 下降到2.7就不下降了 #744

Alexender-Ye opened this issue Apr 30, 2024 · 1 comment

Comments

@Alexender-Ye
Copy link

Alexender-Ye commented Apr 30, 2024

你好,我在做一个QA的数据集的任务,其中query是question,pos是answer,neg是其他的选项,但是在finetune
的时候,loss下降到2.7就不再下降了,并且最后accuracy也大概只有20-30%,请问大佬知道这种事什么问题么?

脚本:torchrun --nproc_per_node 1 -m FlagEmbedding.baai_general_embedding.finetune.run --output_dir finetune_model --model_name_or_path BAAI/bge-small-en-v1.5 --train_data fine_tune_data_10.jsonl --learning_rate 1e-5 --bf16 --num_train_epochs 5 --per_device_train_batch_size 16 --dataloader_drop_last True --normlized True --temperature 0.02 --query_max_len 82 --passage_max_len 56 --train_group_size 10 --negatives_cross_device --logging_steps 10 --save_steps 10000 --query_instruction_for_retrieval ""

微信图片_20240501012041
@staoxiao
Copy link
Collaborator

staoxiao commented May 2, 2024

The loss seems large. I guess there are pseudo-negative examples in the training data (some samples in the negative sample list neg: List[str] are actually positive samples).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants