blog
blog copied to clipboard
PatchTSMixer tutorial missing inference example
The tutorial shows how to run evaluation but is missing a bare bone example of how to use the model to run inference .
for inference, you can directly use model.forward() to get the predictions dict, if your input is tensor
otherwise - if input is test dataset, then you can do model.predict(). All our APIs are huggingface based and used the standard predict and forward calls..