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
Currently, ggml_arange can only generate f32 tensors. I am wondering if there is a way to modify or extend this function to allow ggml_arange to produce tensors of different data types, specifically i32. Any guidance or pointers on where to make these changes in the codebase would be greatly appreciated.
Or if there is any cast method like pytorch's to that I can directly cast it.
Would also love to contribute to that method if there is no quick way to achieve that in the current implementation.
The text was updated successfully, but these errors were encountered:
Currently,
ggml_arange
can only generate f32 tensors. I am wondering if there is a way to modify or extend this function to allow ggml_arange to produce tensors of different data types, specifically i32. Any guidance or pointers on where to make these changes in the codebase would be greatly appreciated.Or if there is any cast method like pytorch's
to
that I can directly cast it.Would also love to contribute to that method if there is no quick way to achieve that in the current implementation.
The text was updated successfully, but these errors were encountered: