HugNLP icon indicating copy to clipboard operation
HugNLP copied to clipboard

CIKM2023 Best Demo Paper Award. HugNLP is a unified and comprehensive NLP library based on HuggingFace Transformer. Please hugging for NLP now!😊

Results 8 HugNLP issues
Sort by recently updated
recently updated
newest added

1、使用run_extractive_unified_ie.sh启动训练,运行正常,结果如下: ***** eval metrics ***** epoch = 2.0 eval_ = 0.0 eval_loss = 8.7072 eval_macro_f1 = 0.0 eval_micro_f1 = 0.0 eval_num = 16 eval_runtime = 0:00:01.11 eval_samples = 16 eval_samples_per_second...

能在代码里面修改,超过512分段计算,给出最后准确的start和end?

作者您好,在`documents/pretraining/Causal LM for Continual Pre-training.md`里面,有这样一句话`输入时只需要直接将input_ids复制一份为label即可`,麻烦问一下因为在计算loss的时候,label需要左移一位,那么这个操作是在哪一部分被完成的呢,是在trainer里面吗,可是trainer如何知道是causal loss呢

api_test.py能运行,但是用了CPU,怎样才能改为用GPU运行呢?

Update: README.md -> cikm23

各位大佬好,想问一下 hugnlp 框架对于 NER 任务的使用,在 readme 中看到了结合了 NER 的关系抽取示例,但没有找到仅执行 NER的示例,请问有同学试过吗,或者作者大大可以给一个相关的案例吗?

你好,我无法找到文件: data_path=/wjn/nlp_task_datasets/kg-pre-trained-corpus/total_pretrain_kgicl_gpt,感觉看的有点模糊,麻烦指个路,谢谢!

Bumps [torch](https://github.com/pytorch/pytorch) from 1.12.1 to 1.13.1. Release notes Sourced from torch's releases. PyTorch 1.13.1 Release, small bug fix release This release is meant to fix the following issues (regressions /...

dependencies