Series decomposition block

Special designed layernorm for the seasonal part

AutoCorrelation Mechanism with the following two phases: (1) period-based dependencies discovery (2) time delay aggregation This bloc

Autoformer encoder

Autoformer decoder layer with the progressive decomposition architecture

Autoformer encoder

Autoformer encoder layer with the progressive decomposition architecture

Paper: Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting Link: https://arxiv.org/abs/1707.01926 Codes are

Paper: Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting Link: https://arxiv.org/abs/1707.01926 Codes are

r"""An implementation of the Diffusion Convolution Layer. For details see: `"Diffusion Convolutional Recurrent Neural Network: Data-Driven Tra

Layer parameters.

Layer parameters.

1D multiwavelet block.

Moving average block to highlight the trend of time series

FEDformer performs the attention mechanism on frequency domain and achieved O(N) complexity

Multi-Layer Perceptron with residual links.

1D Multiwavelet Cross Attention layer.

Standard the input

r"""Scaled Dot-Product Attention module (Attention is all you need by Vaswani et al., 2017) with optional residual attention from previous layer (

Graph convolution network.

Linear layer.

Moving average block to highlight the trend of time series

Moving average block to highlight the trend of time series

Graph conv operation.

Series decomposition block

Series decomposition block

Series decomposition block

Series decomposition block

Extremely slow, please sample < 0.1

Paper: Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting Link: https://arxiv.org/abs/1707.01926 Codes are