dontcryme / angulargrad-tf

Tensorflow implementation of AngularGrad: A New Optimization Technique for Angular Convergence of Convolutional Neural Networks

AngularGrad-tf

Tensorflow implementation of AngularGrad: A New Optimization Technique for Angular Convergence of Convolutional Neural Networks.

The official implementation of AngularGrad is mhaut/AngularGrad.

How to use

You can import the optimizer as follows:

AngularGrad(cos)

from angular_grad import AngularGrad
...
model = YourModel()
...
model.compile(optimizer=AngularGrad("cos"), ...)
...

Or you can omit a value "cos".

...
model.compile(optimizer=AngularGrad(), ...)
...

AngularGrad(tan)

from angular_grad import AngularGrad
...
model = YourModel()
...
model.compile(optimizer=AngularGrad("tan"), ...)
...

Params

AngularGrad(
    method_angle: str = "cos",
    learning_rate=1e-3,
    beta_1=0.9,
    beta_2=0.999,
    eps=1e-7,
    name: str = "AngularGrad",
    **kwargs
)

Tested version

• Python 3.6.9
• Tensorflow 2.5.0

Developed by Eunchan Lee([email protected]), 2021 Linewalks.

If there is any problem in this repository, please feel free to contact us at the above email address.