BUG

[StarxSky]

Hello, I was running the code and found a place where the parameters were incorrect causing the stack to report an error.

Stacks

File "C:\Users\xbj0916\Desktop\CREAD\modeling\main.py", line 136, in train
_ = run_one_epoch('dev', dev_dataloader, trainer, -1, 'teacher_force')
File "C:\Users\xbj0916\Desktop\CREAD\modeling\main.py", line 59, in run_one_epoch
loss, _, _, _, _, _, _ = model(input_ids=batch['input_ids'], attention_mask=batch['attention_mask'],
File "C:\Program Files\Anaconda\envs\ML\lib\site-packages\torch\nn\modules\module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\xbj0916\Desktop\CREAD\modeling\model.py", line 326, in forward
transformer_outputs = self.transformer(input_ids, past=past, attention_mask=attention_mask,
File "C:\Program Files\Anaconda\envs\ML\lib\site-packages\torch\nn\modules\module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
TypeError: forward() got an unexpected keyword argument 'past'
Epoch -1 teacher_force: 0%| | 0/1901 [00:00<?, ?it/s]

Error Points: https://github.com/apple/ml-cread/blob/main/modeling/model.py#L325

Fix

• Change

transformer_outputs = self.transformer(input_ids, attention_mask=attention_mask,
	token_type_ids=token_type_ids, position_ids=position_ids,
	head_mask=head_mask, inputs_embeds=inputs_embeds, use_cache=use_cache,
	output_attentions=True, output_hidden_states=True)

• Removepastargument

def forward(self, input_ids=None, past=None, attention_mask=None, token_type_ids=None, position_ids=None,
				head_mask=None, inputs_embeds=None, labels=None, use_cache=None, output_attentions=None,
				output_hidden_states=None, step=None, mention_labels=None, predict_mention=True, predict_lm=True,
				coref_attn=None, batch=None, coref_links=None）。

I hope this is useful to you, and finally, I wish Apple better and better!