Zhao HG keras-radam Issues

CyberZHG/keras-radam Issues

ModuleNotFoundError: No module named 'keras.legacy'
Closed 3 years ago1
No module name math_ops
Closed 4 years ago2
AttributeError: 'TFOptimizer' object has no attribute 'learning_rate'
Closed 4 years ago7
How can i use radam in tf1.4?
Closed 4 years ago1
ValueError: An operation has `None` for gradient. Please make sure that all of your ops have a gradient defined (i.e. are differentiable). Common ops without gradient: K.argmax, K.round, K.eval.
Closed 4 years ago1
Could not interpret optimizer identifier
Closed 5 years ago3
Warmup causes NAN
Closed 4 years ago1
Any other parameter needed when warmup?
Closed 4 years ago1
TypeError: Unexpected keyword argument passed to optimizer: lr
Closed 5 years ago1
Very slow implementation
Closed 5 years ago1
Unknown optimizer: RAdam
Closed 5 years ago3
TF.keras
Closed 5 years ago
Generate First Release
Closed 5 years ago1
The keras backend support
Closed 5 years ago4
Maintenance
Closed 5 years ago1
amsgrad parameter
Closed 5 years ago3
Issue about the dtype
Closed 5 years ago3
Ranger Optimizer extension
Closed 5 years ago2
epsilon not compatible with Adam
Closed 5 years ago1
ModuleNotFoundError: No module named 'tensorflow.python.keras.optimizer_v2'
Closed 5 years ago1
Numerical instability with higher learning rates & larger networks
Closed 5 years ago3
not compatable with tf 1.8 , AttributeError: 'RAdamOptimizer' object has no attribute '_call_if_callable'
Closed 5 years ago6
What version number is the latest version of keras-rectified-adam?
Closed 5 years ago1
Cannot start training with TensorFlow 2.0 and distribute.MirroredStrategy
Closed 5 years ago6
Usage for Tensorflow 2.0 minimize with var_list
Closed 5 years ago6
AttributeError: 'RAdam' object has no attribute 'apply_gradients'
Closed 5 years ago1
RAdam in Tensorflow
Closed 5 years ago3
ValueError: ('Could not interpret optimizer identifier:', <keras_radam.optimizers.RAdam object at 0x7fd0dab35358>)
Closed 5 years ago4
To use Warmup
Closed 5 years ago1
Typo in README.md
Closed 5 years ago1
Do I need to tune learning rates?
Closed 5 years ago4
Please add weight decay
Closed 5 years ago4
keras ReduceLROnPlateau callback expects lr property
Closed 5 years ago2
__init__() missing 1 required positional argument: 'name'
Closed 5 years ago5
min_lr isn't set properly
Closed 5 years ago1

CyberZHG / keras-radam

CyberZHG/keras-radam Issues

ModuleNotFoundError: No module named 'keras.legacy'

No module name math_ops

AttributeError: 'TFOptimizer' object has no attribute 'learning_rate'

How can i use radam in tf1.4?

ValueError: An operation has `None` for gradient. Please make sure that all of your ops have a gradient defined (i.e. are differentiable). Common ops without gradient: K.argmax, K.round, K.eval.

Could not interpret optimizer identifier

Warmup causes NAN

Any other parameter needed when warmup?

TypeError: Unexpected keyword argument passed to optimizer: lr

Very slow implementation

Unknown optimizer: RAdam

TF.keras

Generate First Release

The keras backend support

Maintenance

amsgrad parameter

Issue about the dtype

Ranger Optimizer extension

epsilon not compatible with Adam

ModuleNotFoundError: No module named 'tensorflow.python.keras.optimizer_v2'

Numerical instability with higher learning rates & larger networks

not compatable with tf 1.8 , AttributeError: 'RAdamOptimizer' object has no attribute '_call_if_callable'

What version number is the latest version of keras-rectified-adam?

Cannot start training with TensorFlow 2.0 and distribute.MirroredStrategy

Usage for Tensorflow 2.0 minimize with var_list

AttributeError: 'RAdam' object has no attribute 'apply_gradients'

RAdam in Tensorflow

ValueError: ('Could not interpret optimizer identifier:', <keras_radam.optimizers.RAdam object at 0x7fd0dab35358>)

To use Warmup

Typo in README.md

Do I need to tune learning rates?

Please add weight decay

keras ReduceLROnPlateau callback expects lr property

init() missing 1 required positional argument: 'name'

min_lr isn't set properly