6 Pull requests opened by 2 people

• Bump serde_json from 1.0.133 to 1.0.134

  #2635 opened Dec 23, 2024

• Bump thiserror from 2.0.8 to 2.0.9

  #2636 opened Dec 23, 2024

• Bump openblas-src from 0.10.10 to 0.10.11

  #2637 opened Dec 23, 2024

• Bump bytemuck from 1.20.0 to 1.21.0

  #2638 opened Dec 23, 2024

• Bump tokio-tungstenite from 0.24.0 to 0.26.1

  #2639 opened Dec 23, 2024

• Combined PRs

  #2640 opened Dec 23, 2024

5 Issues opened by 3 people

• Invalid gradients on Cuda backend using f32 above certain layer size

  #2646 opened Dec 28, 2024

• Early stopping for customized training loop

  #2645 opened Dec 28, 2024

• convert elements to primitive datatypes

  #2644 opened Dec 28, 2024

• BatchNorm don't work as expected

  #2642 opened Dec 26, 2024

• Boolean algebra and bit counting on tensor elements

  #2641 opened Dec 23, 2024

6 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.

• feat: bitwise-ops-for-tensors

  #2498 commented on Dec 28, 2024 • 6 new comments

• Update two models

  #2630 commented on Dec 25, 2024 • 0 new comments

• Implement ctc loss function

  #1536 commented on Dec 27, 2024 • 0 new comments

• Error when deserializing Enum with mismatched weight tensor shapes across variants

  #2332 commented on Dec 28, 2024 • 0 new comments

• Refactor for wgpu v23 compatibility

  #2435 commented on Dec 25, 2024 • 0 new comments

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

  #2613 commented on Dec 26, 2024 • 0 new comments