paddlets.models.anomaly.dl._mtad_gat.layer

class ConvLayer(feature_dim: int, kernel_size: int = 7)[source]

Bases: Layer

Convolution layer to extract features.

Parameters

feature_dim (int) – The number of features.
kernel_size (int) – Kernel size for Conv1D.

_pad

The pad layer.

Type: paddle.nn.Layer

_conv

The conv layer.

Type: paddle.nn.Layer

_relu

The relu layer.

Type: paddle.nn.Layer

forward(x)[source]

Forward

Parameters: x (paddle.Tensor) – The input data.
Returns: Output of conv layer.
Return type: paddle.Tensor

class GRULayer(input_size: int, hidden_size: int, num_layers: int, dropout: float)[source]

Bases: Layer

GRU layer.

Parameters

input_size (int) – The input size
hidden_size (int) – The hidden size.
num_layers (int) – The number of layer.
dropout (float) – Dropout regularization parameter.

_dropout

Dropout regularization parameter.

Type: float

_gru

The gru layer.

Type: paddle.nn.Layer

forward(x)[source]

Forward

Parameters: x (paddle.Tensor) – The input data.
Returns: Output of grulayer. h(paddle.Tensor): final_states.
Return type: out(paddle.Tensor)