A question about the hyperparameter Settings of the model Time-Grad on both of Solar and Wikipedia datasets.

[coding-loong]

I am inspired by your outstanding work "Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting". When I tried to reproduce the results in the paper, there was a large discrepancy between the Solar and Wikipedia datasets. My guess is that the hyperparameter Settings are different from those in the Electricity notebook. Therefore, I would like to ask the authors about how to set the right hyperparameters to get consistent results with the paper on the above two datasets.