The correct way of using rand_augment
CoinCheung opened this issue · 2 comments
CoinCheung commented
Describe the question.
I have a code, part of which is like this:
And this:
When I launch training, The program just got killed without any error message:
Do you know what is the cause of this, and how could I solve this problem?
I have tried to shrink the batchsize so that oom is not happening here.
Check for duplicates
- I have searched the open bugs/issues and have found no duplicates for this bug report
JanuszL commented
Hi @CoinCheung,
Thank you for reaching out. Can you provide a minimal and self-contained reproduction example we can run on our end (even using a toy data from https://github.com/NVIDIA/DALI_extra) so we know that we are talking about the same issue?
CoinCheung commented
Sorry for not leave a message for so long. I am closing this since I do not have gpus to use for these days, and I will open a new one when I am ready to solve this problem.