Training step issue

[wagu0809]

Hi there, here I found that the step += 1 is under the for loop of train_iter.
And in this case, only the first 100 samples of the first batch are used to train, is that correct?

[nlpyang]

I am not sure why do you think like that.

[wagu0809]

I am not sure why do you think like that.

Because I printed the training process and it only gave me 100(default training step is 100) prints and the whole process stopped very soon.
Or do I misunderstand the function of training_step? Does it indicate training epochs or some other use?
Thanks in advance for your time to explain!

[nlpyang]

training_step is the total number of steps for training.

[wagu0809]

Clear now, thanks for your explanation!