louchao98 / nner_as_parsing Goto Github PK

您好，目前我正在genia数据集上复现您的NNER模型，但是在复现过程中遇到了一些问题，我想问一下配置文件中涉及到的/data/bio_nlp_vec/PubMed-shuffle-win-30.txt文件该如何获得，目前我只能在网上找到PubMed-shuffle-win-30.bin和PubMed-shuffle-win-2.bin文件，但是好像不能用在该模型中，不知道您是否可以提供一下PubMed-shuffle-win-30.txt词向量文件呢？

RuntimeError: CUDA out of memory. Tried to allocate 14.00 MiB (GPU 0; 44.40 GiB total capacity; 26.45 GiB already allocated; 10.81 MiB free; 27.44 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

I'm using the similar config file as ACE dataset. Please tell me what extra tweaks maybe needed.

File "model4/nner_as_parsing/src/modules/loss/sjl_multilabel.py", line 8, in multilabel_categorical_crossentropy
loss = loss_fn(y_true, y_pred)
File "anaconda3/envs/model4/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "anaconda3/envs/model4/lib/python3.9/site-packages/torch/nn/modules/loss.py", line 720, in forward
return F.binary_cross_entropy_with_logits(input, target,
File "anaconda3/envs/model4/lib/python3.9/site-packages/torch/nn/functional.py", line 3160, in binary_cross_entropy_with_logits
raise ValueError("Target size ({}) must be the same as input size ({})".format(target.size(), input.size()))
ValueError: Target size (torch.Size([13, 10, 51, 6])) must be the same as input size (torch.Size([13, 10, 1, 8]))

My dataset has about 86 types of labels and it is quite small and has only 11k lines and their Span tags. Can you tell how to choose hyperparameters used in config file ?? @LouChao98