Hello! I encountered an issue during pretraining where the forward method doesn't recognize the masked_lm_labels parameter.
"/home/mickey/Tools/miniconda3/envs/methylbert/lib/python3.11/site-packages/methylbert/trainer.py", line 287
mask_lm_output = self.model.forward(input_ids = data["input"],
masked_lm_labels = data["label"])
Questions:
Should the parameter name be labels instead of masked_lm_labels?