Skip to content

Hardcoded frequency in PytorchForecastingTransform #106

Closed
4 tasks done
Mr-Geekman opened this issue Sep 28, 2021 · 1 comment · Fixed by #107
Closed
4 tasks done

Hardcoded frequency in PytorchForecastingTransform #106

Mr-Geekman opened this issue Sep 28, 2021 · 1 comment · Fixed by #107
Assignees
@Mr-Geekman
Labels
bug Something isn't working

Comments

@Mr-Geekman
Copy link
Contributor

🐛 Bug Report

How we have this line in PytorchForecastingTransform.fit:

ts = TSDataset(df, "1d")

The problem is that this df can have any frequency.

Expected behavior

Frequency is calculated instead of hardcoding using pd.infer_freq.

How To Reproduce

Use not daily dataset with neural models.

Environment

No response

Additional context

No response

Checklist

  • Bug appears at the latest library version
  • Bug description added
  • Steps to reproduce added
  • Expected behavior added
@Mr-Geekman Mr-Geekman added the bug Something isn't working label Sep 28, 2021
@Mr-Geekman Mr-Geekman self-assigned this Sep 28, 2021
@martins0n
Copy link
Contributor

Method transform has the same issue аdditionaly.

@Mr-Geekman Mr-Geekman mentioned this issue Sep 28, 2021
Merged
9 tasks
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees

@Mr-Geekman Mr-Geekman

Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Fix bug with hardcoded frequency in PytorchForecastingTransform tinkoff-ai/etna
2 participants
@martins0n @Mr-Geekman