GPU out of memory error

[humayunr7]

I am getting out of memory error. How much GPU mem is required to train with batch size 32?

[kentaroy47]

depends on your data, but bs=32 may take up about 16GB of memory.
try halving the batch size?

[alezanga]

I have a similar problem, but how can I change the batch size?
Thanks.

[kentaroy47]

If you run out of memory, try reducing the number of ROIs that are processed simultaneously. Try passing a lower -n to train_frcnn.py. Alternatively, try reducing the image size from the default value of 600 (this setting is found in config.py.