Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Performance issues in utils/data_loader.py (by P3) #67

Open
DLPerf opened this issue Aug 22, 2021 · 1 comment
Open

Performance issues in utils/data_loader.py (by P3) #67

DLPerf opened this issue Aug 22, 2021 · 1 comment

Comments

@DLPerf
Copy link

DLPerf commented Aug 22, 2021

Hello! I've found a performance issue in utils/data_loader.py: batch() should be called before map(), which could make your program more efficient. Here is the tensorflow document to support it.

Detailed description is listed below:

  • data_loader.py: train_dataset.batch(batchsize)(line 103) should be called before train_dataset.map(parse_data, num_parallel_calls=2)(line 101).
  • data_loader.py: val_dataset.batch(batchsize)(line 140) should be called before val_dataset.map(parse_data_without_augmentation)(line 138).
  • data_loader.py: train_dataset.batch(batchsize)(line 194) should be called before train_dataset.map(parse_single_record, num_parallel_calls=4)(line 192).

Besides, you need to check the function called in map()(e.g., parse_single_record called in train_dataset.map(parse_single_record, num_parallel_calls=4)) whether to be affected or not to make the changed code work properly. For example, if parse_single_record 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
Copy link
Author

DLPerf commented Nov 4, 2021

Hello, I'm looking forward to your reply~

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant