-
Notifications
You must be signed in to change notification settings - Fork 904
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
Refactor/hist fc general #1801
Refactor/hist fc general #1801
Conversation
… ForecastingModel
Codecov ReportPatch coverage:
❗ Your organization is not using the GitHub App Integration. As a result you may experience degraded service beginning May 15th. Please install the Github App Integration for your organization. Read more. Additional details and impacted files@@ Coverage Diff @@
## master #1801 +/- ##
==========================================
- Coverage 94.25% 94.13% -0.12%
==========================================
Files 125 125
Lines 11619 11604 -15
==========================================
- Hits 10951 10923 -28
- Misses 668 681 +13
☔ View full report in Codecov by Sentry. |
* simple time reduction from removing redundant slicing * move the min_start_time truncation out of the loop * replace self by model to allow for moving historical forecasts out of ForecastingModel * replace series.drop_*() with series.slice() for performance boost * make same state as test * optimize some time series methods * used improved splitting in historical forecasts * Update CHANGELOG.md
Summary
This is the first part of improving historical forecasting:
This PR reduces the TimeSeries preprocessing time in historical forecasting for all models. Below are the result from old (original) hist fc against new implementation for integer- and datetime indexed series (retrain=False):