-
Notifications
You must be signed in to change notification settings - Fork 0
Features
Garrett Long edited this page May 24, 2020
·
2 revisions
- Fully connected layers, obviously.
- Convolution, both standard and partially strided/transposed/deconvolution. The latter is used to calculate the output value errors.
- A gui: the neural network is fully customizable, as are most (if not all) hyperparameters. Deep networks are supported here, as well.
- Pooling in various sizes.
- Momentum
- Nesterov Momentum
- LeCun initialization
- RMSProp
- Batch normalization