How to enable repeat in data loading?

1

I am studying the data loading tutorial. I am wondering if there is similar utility as repeat() in TensorFlow. It is necessary when the size of the dataset is smaller than my training iterations.

Thank you in advance.

2

AFAIK, there are many ways to achieve your goal. Here are two examples.

dloader = DataLoader(dset, batch_size=batch_size, drop_last=True, shuffle=True)
loader = iter(dloader)
while True:
    try:
        img, label = loader.next()
    except StopIteration:
        loader = iter(dloader)
        img, label = loader.next()

   # ...

and

dloader = DataLoader(dset, batch_size=batch_size, drop_last=True, shuffle=True)
epoch = 3
for e in range(epoch):
   for img, label in dloader:
      # ...