i understood that eval() changes the bn and dropout layer’s behaviour torch.no_grad() deals with the autograd engine and stops it from calculating the gradients, which is not compulsory but the recommended way of doing validation BUT, I didn’t understand the use of with torch.set_grad_enabled() …

[image] 'model.eval()' vs 'with torch.no_grad()' are you saying that torch.no_grad and torch.set_grad_enabled(False) are the same ?

Difference between eval(), torch.no_grad() and with torch.set_grad_enabled()

n0obcoder (n0obcoder) June 4, 2019, 11:21am 2