Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Timestamps in Predictions #16

Open
Himmelxd opened this issue Nov 16, 2024 · 2 comments
Open

Timestamps in Predictions #16

Himmelxd opened this issue Nov 16, 2024 · 2 comments

Comments

@Himmelxd
Copy link

Good day,
I was wondering, when using the .predict function in combination with timestamps (column 'date') there is no 'date' column in the prediction. But what is the resulting frequency of the prediction? I tested some models but never got a definitive answer.
Thanks

@Secilia-Cxy
Copy link
Owner

By default, we use a sliding window of stride 1, and each prediction window moves forward one step at a time. For example, with a seq_len of 96, the first prediction is based on [0, 96), the second on [1, 97), and so on.
Hope this helps!

@Himmelxd
Copy link
Author

Himmelxd commented Jan 7, 2025

I'm afraid not.
My question is.
Inputing a timeseries with a column 'date' - where the timestamps are not equally spaced apart.
I get as predictions only timeseries without a 'date' column (so one column less than the input).
But what exactly are the timestamps/the time intervalls then?
Doing some tests i did not found a definitive answer, probably on my side though.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants