Adagrad Optimizer

neualpy.optimizer.Adagrad(learning_rate=0.001, learning_rate_decay=0.0, eps=1e-08, weight_decay=0.0)

info

Adagrad Optimizer is mostly stable and can be used for any project. In the future, any chance of breaking changes is very low.

Applies Adagrad algorithm, Adagrad:Adaptive Subgradient Methods for Online Learning and Stochastic Optimization

For more information, check this page.

Supported Arguments

• learning_rate=0.001: (Float) Learning Rate for the optimizer
• learning_rate_decay=(0.9,0.999) : (Float) Learning Rate decay
• eps=0 : (Float) Term added to the denominator to improve numerical stability
• weight_decay=0 : (Float) Weight decay for the optimizer

Code Example

from neuralpy.models import Sequential

from neuralpy.optimizer import Adagrad

...

# Rest of the imports

...

model = Sequential()

...

# Rest of the architecture

...

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