|
Learning Rates tab (Training Options dialog box) |
|
Use this tab to modify training rates for existing neural networks.
Select from this list the layer you want to modify.
Learning
Displays the currently set learning rate for the neurons in this layer. To modify this rate, type a learning rate.
Refer to Learning Rate (Layer Definition tab (Network Definition dialog box)) for additional information on learning rates.
Displays the currently set step of the learning rate for the neurons in this layer. To modify this rate step, type a new learning rate step.
Refer to Learning Step (Layer Definition tab (Network Definition dialog box)) for additional information on the learning rate step.
Click this button to copy the current learning rate and step into all previous layers.
 | This command is available when the currently selected layer is other than the first hidden layer. |
Click this button to copy the current learning rate and step into all following layers.
| This command is available when the currently selected layer is other than the output layer. |
Momentum
Displays the currently set momentum rate for the neurons in this layer. To modify this rate, type a new momentum rate.
Refer to Momentum Rate (Layer Definition tab (Network Definition dialog box)) for additional information on momentum rates.
Displays the currently set step of the momentum rate for the neurons in this layer. To modify this rate step, type a new momentum rate step.
Refer to Momentum Step ((Layer Definition tab (Network Definition dialog box)) for additional information on the momentum rate step.
Click this button to copy the current learning rate and step into all previous layers.
| This command is available when the currently selected layer is other than the first hidden layer. |
Click this button to copy the current momentum rate and step into all following layers.
| This command is available when the currently selected layer is other than the output layer. |
Momentum Base
Indicates that the momentum rate is applied on the previous total learning rate change.
Indicates that the momentum rate is applied on the previous partial derivative only.
| These options have no effect when the momentum rate for the neurons in this layer is set to zero. |
Refer to backpropagation for additional information on the momentum base.
Sponsored
Try Predictive Systems Lab
Need Help?
Contact support