Ross Wightman 20d66beead Move RMSpropTF another step closer to Tensorflow impl ...

* init square_avg with one instead of zero as per TF
* match TF order of ops for square_avg accumulation
* move LR scaling to momentum buffer accumulator as per TF
* add decoupled weight decay flag (not in TF)

2019-05-14 18:35:43 -07:00

..

__init__.py

Remove adabound optimizer, never got it working well on larger datasets

2019-05-14 18:35:43 -07:00

nadam.py

Update a few comment, add some references

2019-04-12 23:16:49 -07:00

optim_factory.py

Remove adabound optimizer, never got it working well on larger datasets

2019-05-14 18:35:43 -07:00

rmsprop_tf.py

Move RMSpropTF another step closer to Tensorflow impl

2019-05-14 18:35:43 -07:00