“Calling the model on the input returns a 10-dimensional tensor with raw predicted values for each class.” stated at this link. In reality, a 2-dimension tensor is returned with dim=0 corresponding to each tensor and dim=1 corresponding to the 10 values in each tensor. There are 10 values in dim=1, not 10 dimensions. Is there anyone I can contact to help get this fixed?
Thanks for reporting this as the phrasing seems to be indeed confusing.
Yes, you could provide a fix by submitting a PR on GitHub
Thank you, first time trying to correct documentation for something like this
That sounds like a good start! As a general recommendation check the Contribution Guide and let me know if you need more information.