RileyShe

"crf.py", line 177, in _compute_score score = self.start_transitions[tags[0]] IndexError: index -100 is out of bounds for dimension 0 with size 5 上面是报错信息，但是-100 是label_id的padding值哦，理论上没啥问题吧。 ``` score = self.start_transitions[tags[0]] score += emissions[0,...

[BUG/Help] ds_train_finetune.sh 多卡训练 需要多少资源才行？

34

### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior ds_train_finetune.sh 多卡训练时，我用到24G 3090 四张显卡还是`torch.cuda.OutOfMemoryError: CUDA out of memory.` 1. 感觉不应该啊，96G的显存，用模型并行的方式还是不够用。 所以...

[bug or question]数据的输入和输出处理部分的疑问，麻烦作者帮忙解答一下

4

“”“ def preprocess(tokenizer, example, max_seq_length): prompt = example["context"] target = example["target"] prompt_ids = tokenizer.encode(prompt, max_length=max_seq_length, truncation=True) target_ids = tokenizer.encode( target, max_length=max_seq_length, truncation=True, add_special_tokens=False ) input_ids = prompt_ids + target_ids +...

bug?

1

https://github.com/PaddlePaddle/PaddleNLP/blob/6f1c0bcc69fbc522716225cf9eac325de0c16433/model_zoo/bert/create_pretraining_data.py#L80 这行代码逻辑是不是写错了，应该是 next_sentence_label = 1 if not instance.is_random_next else 0