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
After huggingface/candle#1620 is resolve, update the burn-candle backend. Currently default unoptimized implementation is used for float tensor. To use with int tensor, one should convert to float (tensor.float()) and use prod or prod_dim methods.
The text was updated successfully, but these errors were encountered:
After huggingface/candle#1620 is resolve, update the burn-candle backend. Currently default unoptimized implementation is used for float tensor. To use with int tensor, one should convert to float (
tensor.float()
) and useprod
orprod_dim
methods.The text was updated successfully, but these errors were encountered: