1 Release published by 1 person

• v0.16.0

  published Jan 14, 2025

11 Pull requests merged by 6 people

• Code generation bug fix for ONNX import

  #2708 merged Jan 16, 2025

• Clean up train system metrics

  #2707 merged Jan 16, 2025

• Fix GRU to match pytorch (#2701).

  #2704 merged Jan 16, 2025

• Add hardsigmoid formula and fix WGAN doc + default lr

  #2706 merged Jan 16, 2025

• Remove deprecated Data and DataSerialize

  #2703 merged Jan 16, 2025

• Wasserstein Generative Adversarial Network

  #2660 merged Jan 15, 2025

• Feature add new one hot function meeting multi-dimensions (ranks)

  #2613 merged Jan 15, 2025

• Bump next version of Burn to 0.17.0

  #2698 merged Jan 14, 2025

• Set cubecl version

  #2697 merged Jan 14, 2025

• Fix reduce autotune key no anchor

  #2696 merged Jan 14, 2025

• Add dropout prob check

  #2695 merged Jan 14, 2025

1 Pull request opened by 1 person

• Clean up -jit suffix in feature flags and modules

  #2705 opened Jan 15, 2025

3 Issues closed by 2 people

• ONNX import generates integer for Interpolate2DConfig when it should be f32

  #2702 closed Jan 16, 2025

• GRU implementation does not use hidden state in iteration, differs from PyTorch

  #2701 closed Jan 16, 2025

• zero dropout rate

  #2694 closed Jan 15, 2025

1 Issue opened by 1 person

• Scalars have caused some issues with onnx import

  #2699 opened Jan 15, 2025

3 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Add "load pytorch tensor" section into the burn book

  #2316 commented on Jan 14, 2025 • 0 new comments

• Implement `grid_sample` for sampling from a spatial tensor at arbitrary locations, similar to PyTorch

  #2674 commented on Jan 16, 2025 • 0 new comments

• expected `*mut u8`, found `*mut i8`

  #2596 commented on Jan 16, 2025 • 0 new comments