Is there a way to use TTA on new image for prediction?
Currently, I am using : pred_class,pred_idx,outputs = learn.predict(img)
for predicting class of an image. Is there a way to do TTA on the image at run-time.
Appreciate any help.
Is there a way to use TTA on new image for prediction?
Currently, I am using : pred_class,pred_idx,outputs = learn.predict(img)
for predicting class of an image. Is there a way to do TTA on the image at run-time.
Appreciate any help.
I have exactly the same question. Have you ever worked it out?