Problems encountered when fine-tuning model validation

#41
by ysy2002 - opened

01/31/2024 22:44:19 - INFO - main - *** Evaluate ***
[INFO|trainer.py:3283] 2024-01-31 22:44:19,667 >> ***** Running Evaluation *****
[INFO|trainer.py:3285] 2024-01-31 22:44:19,667 >> Num examples = 0
[INFO|trainer.py:3288] 2024-01-31 22:44:19,667 >> Batch size = 1
Traceback (most recent call last):
File "run_clm.py", line 670, in
main()
File "run_clm.py", line 636, in main
metrics = trainer.evaluate()
File "/root/autodl-tmp/shiyang/software/protgpt2/lib/python3.8/site-packages/transformers/trainer.py", line 3136, in evaluate
output = eval_loop(
File "/root/autodl-tmp/shiyang/software/protgpt2/lib/python3.8/site-packages/transformers/trainer.py", line 3325, in evaluation_loop
loss, logits, labels = self.prediction_step(model, inputs, prediction_loss_only, ignore_keys=ignore_keys)
File "/root/autodl-tmp/shiyang/software/protgpt2/lib/python3.8/site-packages/transformers/trainer.py", line 3495, in prediction_step
has_labels = False if len(self.label_names) == 0 else all(inputs.get(k) is not None for k in self.label_names)
File "/root/autodl-tmp/shiyang/software/protgpt2/lib/python3.8/site-packages/transformers/trainer.py", line 3495, in
has_labels = False if len(self.label_names) == 0 else all(inputs.get(k) is not None for k in self.label_names)
AttributeError: 'NoneType' object has no attribute 'get'

Dear author, I encountered such an error when fine-tuning the model, and I have been unable to solve it effectively. Sincerely wanting to ask for your help

Owner

It seems you don’t have any sequences in your validation dataset.

Sign up or log in to comment