Chunngai Ho

Question about the loss calculation of pet

3

Hello. I am not clear about the loss calculation of pet during training (as described in Section 3.1 in the paper: Exploiting Cloze Questions for Few-Shot Text Classification and Natural...

When testing the feature of *background color with rounded corners* (as introduced [here](https://levelup.gitconnected.com/background-with-rounded-corners-in-uitextview-1c095c708d14)), I found two issues with a non-zero line spacing, as shown in the attached screenshot. ![Screenshot 2024-02-21...

Hello. The `max_len` is set to 50 in training (as mentioned [here](https://github.com/grammarly/gector/blob/master/docs/training_parameters.md)). Is it the same for prediction?

自定义 metric 报错：UnpicklingError

1

**Describe the bug** SFT 训练评分模型（分数范围是 1, 1.5, 2, 2.5, 3），需要自定义 metric 计算 ACC 和 QWK。 数据格式（train valid test 格式相同）： ```python { # LLM request messages "messages": [ {"role": "system", "content":...