AdaBound-Tensorflow

AdaBound-Tensorflow copied to clipboard

I have seen a performance boost switching from Adam to AdaBound. I have tuned my model and found that a range of 2e-4 to 2e-2 works best. I am interested...

JohnMBrandt

In line 85, beta power1 is None type, causing errors.

1

beta1_power = math_ops.cast(self._get_non_slot_variable("beta1_power", graph=graph), var.dtype.base_dtype) No configuration changes on my end, throwing an error that beta power1 is None.

ben-arnao

Hi, your works help me a lot and I have a question on your code! I would like to know how to get the Learning rate in your code. I'm...

HugoDel

thanks for your great job. I found it didn't work for large scale model with huge sparse embedding parameters. for the apply_sparse function may be not very efficient as AdamOptimizer...

muye5

x and y must have the same dtype, got tf.float32 != tf.resource

7

Version: Tensorflow 1.12 Stable. When I run the code, it appears the error: > x and y must have the same dtype, got tf.float32 != tf.resource in "AdaBound.py", line 132,...

Chuong98CC