xla_fit_one_cycle fails with an error when using fastai version 2.3.1
Opened this issue · 0 comments
butchland commented
When running an xla fit method, it triggers an error:
Exception in device=TPU:0: Input type (torch.FloatTensor) and weight type (XLAFloatType) should be the same
Note: downgrading to 2.3.0 is a temporary workaround while this is being fixed
Works with fastai==2.3.0
See this gist for a error sample