How to interpret the trained model?

[landkwon94]

Description

Hello Sir!

First of all, thank you so much for sharing your amazing codes!

I am using TFT model for train/validate/test for time-series forecasting.

After training the model, I want to visualize 'feature importance' and 'attention weights' such as figures below (from the original paper)

Is there any module we can use to perform this visualization in neuralforecast?

I will wait for your reply!

Many thanks for your contributions :)

Link

No response

[marcopeix]

#1104 solves this issue.