GitLab

* update ngc link and dockerhub container tag

* update

* update

* update

* Update README.md
Co-authored-by: Masaki Kozuki <mkozuki@nvidia.com>

* local_rank and install cuda version fix

- resume() is a nested function and when it loads best_prec1
  it creates a local variable that hides the one from the parent
  function (which refers to the global one). This PR adds `global`
  to modify the global variable as intended
Signed-off-by: Janusz Lisiecki <jlisiecki@nvidia.com>

Actual flag is --opt_level and copy pasting the example results in an unrecognized arguments error.

* initial commit

* add default O1 mode, enable other modes, add README

* add carilli's review suggestions to README

* remove unused tens tensor in example/imagenet/main_amp.py

* remove unused tens tensor in deprecated examples and tests/L1

Updated default sizes to be multiples of 8 to enable Tensor Core use.  Added performance guidelines to README.