fynance.models.recurrent_neural_network.GatedRecurrentUnit

class fynance.models.recurrent_neural_network.GatedRecurrentUnit(X, y, drop=None, x_type=None, y_type=None, bias=True, forward_activation=<class 'torch.nn.modules.activation.Softmax'>, hidden_activation=<class 'torch.nn.modules.activation.Tanh'>, hidden_state_size=None, reset_activation=<class 'torch.nn.modules.activation.Sigmoid'>, update_activation=<class 'torch.nn.modules.activation.Sigmoid'>)

Gated Recurrent Unit neural network.

Parameters:

X, y : array-like or int

• If it’s an array-like, respectively inputs and outputs data.
• If it’s an integer, respectively dimension of inputs and outputs.

drop : float, optional

Probability of an element to be zeroed.

forward_activation, hidden_activation : torch.nn.Module, optional

Activation functions, default is respectively Softmax and Tanh function.

hidden_state_size : int, optional

Size of hidden states, default is the same size than input.

reset_activation, updated_activation : torch.nn.Module, optional

Activation functions for reset and update gate, default are both Sigmoid function.

See also

RecurrentNeuralNetwork, LongShortTermMemory

Attributes:

criterion : torch.nn.modules.loss

A loss function.

optimizer : torch.optim

An optimizer algorithm.

W_c, W_r, W_u, W_y : torch.nn.Linear

Respectively recurrent (hidden), reset, update and forward wheights.

f_h, f_r, f_u, f_y : torch.nn.Module

Respectively hidden (recurrent), reset, update and forward activation functions.

Methods

__call__(*input, **kwargs)
set_optimizer(criterion, optimizer[, params])	Set the optimizer object.
train_on(X, y, H)	Trains the neural network model.
predict(X, H)	Predicts outputs of neural network model.
set_data(X, y[, x_type, y_type])	Set data inputs and outputs.