Adam Optimizer

neualpy.optimizer.Adam(learning_rate=0.001, betas=(0.9, 0.999), eps=1e-08, weight_decay=0.0, amsgrad=False)
info

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

Applies Adam algorithm, Adam: A Method for Stochastic Optimization

For more information, check this page.

Supported Arguments

  • learning_rate=0.001: (Float) Learning Rate for the optimizer
  • betas=(0.9,0.999) : (Tuple[Float, Float]) coefficients used for computing running averages of gradient and its square
  • eps=0 : (Float) Term added to the denominator to improve numerical stability
  • weight_decay=0 : (Float) Weight decay for the optimizer
  • amsgrad=False : (Bool) if true, then uses AMSGrad various of the optimizer

Code Example

from neuralpy.models import Sequential
from neuralpy.optimizer import Adam
...
# Rest of the imports
...
model = Sequential()
...
# Rest of the architecture
...
model.compile(optimizer=Adam(), ...))
Edit this page