Hi peeps,
I’ve been struggling to figure this out for several days now.
I’m trying to get the total number of validation batches such that I may produce an average accuracy measure across all the batches at validation_epoch_end
Upon running self.trainer.num_batch_length
I’m getting a value which seems to be a function of X GPUs and P CPU workers per GPU.
Is there a simple way for me to get the total?
Many thanks