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
Thanks for your attention to our work! You can replace FP8Linear with torch.nn.Linear, where the weight and bias with ScalingTensor can be converted to torch.float32 by 'weight = weight.float(), bias = bias.float()'. Then the FP32 model can be converted to onnx.
I have try your example, however when I try to export the model to onnx, it's error。
The text was updated successfully, but these errors were encountered: