You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While reviewing the code for the layers used in the CVNets library, my colleague Saanidhya Vats and I observed that certain layers, such as Adaptive Pooling, are inherited from the torch nn module. The maintainer of MLX has suggested that we raise an issue to enable support for CVNets on the MLX backend, which would allow us to take advantage of performance benefits when running on Apple Silicon.
The text was updated successfully, but these errors were encountered:
cvnad1
changed the title
[MLX Backend] CV-Nets
[MLX Backend] CVNets
Oct 25, 2024
@sacmehta We opened an issue to propose some new features in MLX. The discussion with the MLX maintainer can be found in this issue: ml-explore/mlx#1512. If CVNets can integrate with MLX-based features, we'd be glad to contribute any missing functionalities to MLX.
While reviewing the code for the layers used in the CVNets library, my colleague Saanidhya Vats and I observed that certain layers, such as Adaptive Pooling, are inherited from the torch nn module. The maintainer of MLX has suggested that we raise an issue to enable support for CVNets on the MLX backend, which would allow us to take advantage of performance benefits when running on Apple Silicon.
The text was updated successfully, but these errors were encountered: