GitLab

* Add fused_lars optimizer

* Update primitive fused_lars optimizer, working for resnet50 with NHWC/NCHW

* Add flow of using nesterov in FusedLARS

* Add fused mixed precision lamb optimizer.

* Fix device usage in constructor.

* Fix sending param_group tensor state to device.

* Remove unneeded device set.

* Add fused mixed precision lamb optimizer.

* Fix device usage in constructor.

* Fix sending param_group tensor state to device.

* Remove unneeded device set.

* fuse norm into scale

* add fused norm into dlamb
Co-authored-by: Sukru Eryilmaz <seryilmaz@computelab-dgx1v-32.nvidia.com>

correctly not apply bias correction to epsilon(same as recent upstream change)
correctly not apply bias correction to weight decay(consistent with upstream AdamW)
Make adam_w_mode for FusedAdam/LAMB, to do L2 or Weight Decay (Adam vs AdamW)
Correct document reg_inside_moment differently from adam_w_mode in FusedNovoGrad
Removed legacy eps_mode from FusedAdam
Make internal math type float across fused optimizers

* First draft, for discussion

* Fix mistakes in LAMB equations

* Add loop over chunk

* Bug fix

* Bug fix

* Bug fix

* Undo bug fix

* Bug fix

* Add multi tensor LAMB optimizer to setup.py

* Rename step_size to learning_rate

* Fix compilation errors

* Existing tests passing, still need to add per-tensor tests

* Test is passing, still need to measure performance

* ILP for l2norm functor

* Refactor to allow more flexible treatment of multiple optimizers/models/losses

* Adding _process_optimizers.py

* Created L0 tests (now passing).

* fix: minor print typo (#234)

* make L1 results easier to read

* L0 multiple model/optimizer/loss test fleshed out

* Adding test that master params remain synced across distributed processes

* Docstring updates

* Docstring updates

Fix dispatch where we have a parameter group with multiple
combinations of types
Optionally apply weight decay after momentum

Initial implementation, all fp32
Tested against torch.optim.sgd