brainpy.nn.runners.BPFF

class brainpy.nn.runners.BPFF(target, **kwargs)

The trainer implementing back propagation algorithm for feedforward neural networks.

__init__(target, **kwargs)

Methods

__init__(target, **kwargs)
f_grad([shared_kwargs])	Get gradient function.
f_loss([shared_kwargs])	Get loss function.
f_train([shared_kwargs])	Get training function.
fit(train_data[, test_data, num_batch, ...])	Fit the target model according to the given training and testing data.
predict(xs[, initial_states, ...])	Predict a series of input data with the given target model.

Attributes

mapping_type	Mapping type for the output and the target.
train_losses	Training loss.