Pytorch Bert 中文分类 运行代码时候遇到的问题
问题1 bert = AutoModel.from_pretrained('bert-base-chinese')
报错信息如下：
RuntimeError: Error(s) in loading state_dict for BertModel:
size mismatch for bert.embeddings.word_embeddings.weight: copying a param with shape torch.Size([21128, 768]) from checkpoint, the shape in current model is torch.Size([119547, 768]).
后来发现因为以前用的模型版本和token啥的不是一套，所以需要重新下载。
下载地址： bert-base-chinese at main (huggingface.co)
问题2 compute_class_weight('balanced', np.unique(y_train), y_train)
报错 compute_class_weight() takes 1 positional argument but 3 were given
解决方案： compute_class_weight('balanced', classes=np.unique(y_train), y=y_train)
问题3 for batch in tqdm(trainloader):
AttributeError: 'tqdm_notebook' object has no attribute 'disp'
解决方案： pip install ipywidgets