Hello, thank you very much for sharing TXMixer's multivariate prediction implementation. How is its performance?

[potoyeee]

Hello, thank you very much for sharing TXMixer's multivariate prediction implementation. How is its performance?

[ditschuk]

Hi,

I have not properly benchmarked it and sadly can't provide you any numbers on the performance. I plan to benchmark the performance on the different ETT datasets, but I can't guarantee when I'll find the necessary time.

Best,
Konstantin

[deepsleeplearn]

hello,
self.projection = (
nn.Linear(input_channels, output_channels)
if input_channels != output_channels
else nn.Identity()
)

def forward(self, x: torch.Tensor) -> torch.Tensor:
   """
        x: A 3D tensor with shape (N, C, L) where C is the channel dimension.
    """
    x_proj = self.projection(x)

are you sure the middle channel is C ？

[rebpan]

hello, self.projection = ( nn.Linear(input_channels, output_channels) if input_channels != output_channels else nn.Identity() )

def forward(self, x: torch.Tensor) -> torch.Tensor:
   """
        x: A 3D tensor with shape (N, C, L) where C is the channel dimension.
    """
    x_proj = self.projection(x)

are you sure the middle channel is C ？

actually, shape is (N, L, C)