A MLP must be trained with a supervised learning algorithm in order
to work. The gradient backpropagation is by far the most used algorithm
used to train MLPs.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
defaultLearningParameters(self,
param_dict) |
source code
|
|
|
finetune(self,
shared_x_train,
shared_y_train,
batch_size,
learning_rate,
epochs,
growth_factor,
growth_threshold,
badmove_threshold,
verbose)
Performs the supervised learning step of the MultiLayerPerceptron,
using a batch-gradient backpropagation algorithm. |
source code
|
|
|
|
Inherited from module.Sequential :
prepareGeometry ,
prepareOutput ,
prepareParams
Inherited from module.Container :
add
Inherited from module.Module :
criterionFunction ,
forward ,
forwardFunction ,
holdFunction ,
linkInputs ,
linkModule ,
prepare ,
prepareBackup ,
restoreFunction ,
save ,
trainFunction
|