search

tensorlayer / SRGAN

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
https://github.com/tensorlayer/tensorlayerx
3.29k stars 810 forks source link

Performance issue in train.py #238

Open DLPerf opened 3 years ago

DLPerf commented 3 years ago

Hello! I've found a performance issue in train.py: train_ds = train_ds.batch(batch_size)(here) should be called before train_ds = train_ds.map(_map_fn_train, num_parallel_calls=multiprocessing.cpu_count())(here), which would make your program more efficient.

Here is the tensorflow document to support it.

To reproduce the behavior, you need to swap the order of train_ds = train_ds.batch(batch_size) and train_ds = train_ds.map(_map_fn_train, num_parallel_calls=multiprocessing.cpu_count()) in train.py. Besides, you need to check the function _map_fn_train(here) called in train_ds.map() whether to be affected or not to make the changed code work properly. For example, if _map_fn_train needs data with shape (x, y, z) as its input before fix, it would require data with shape (batch_size, x, y, z).

Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.

DLPerf commented 2 years ago

Hello, I'm looking forward to your reply~