multi-step prediction with iteration

Question

multi-step prediction with iteration

Closed this issue 7 months ago · 3 comments

FortunateStarstar86 commented 7 months ago

I'm sorry to bother you with a problem I can't solve.
when I manipulate the input iteratively by appending the predicted value at T time step, and use the new input to predict the values at T+1,the Losses have increased exponentially.

I used the trained model test data (test.csv) to output "predicted", which resulted in exponentially larger predictions from the fourth round. Can you give some analysis or advice?

Looking forward to your reply!

Answer 1 · 2024-03-14T08:30:25.000Z

the second pic:

Answer 2 · 2024-03-15T09:05:46.000Z

Hi, i see. I guess it might be due to the error propagation in the sequential modeling, but it seems a bit severe from your cases. In this case, you might want to use more time steps as input, or try the way of training with an output of 10 dimensions. Sorry for the inconvenience.

Answer 3 · 2024-03-18T02:42:15.000Z

Hi, i see. I guess it might be due to the error propagation in the sequential modeling, but it seems a bit severe from your cases. In this case, you might want to use more time steps as input, or try the way of training with an output of 10 dimensions. Sorry for the inconvenience.

OK，Thanks for your reply！