temporal-shift-module
temporal-shift-module copied to clipboard
optim policy for charades dataset
Did you use the optim policy for traning the model for charades dataset or did you train without optim policy to get the mentioned maP values in the paper?