codecov-commenter
commented
1 year ago Codecov Report
Patch coverage: 100.00% and project coverage change: +0.02 :tada:
Comparison is base (
474dcf6) 82.89% compared to head (9da80d1) 82.92%.
Additional details and impacted files
```diff @@ Coverage Diff @@ ## master #158 +/- ## ========================================== + Coverage 82.89% 82.92% +0.02% ========================================== Files 18 18 Lines 3104 3109 +5 Branches 360 361 +1 ========================================== + Hits 2573 2578 +5 Misses 478 478 Partials 53 53 ``` | [Impacted Files](https://codecov.io/gh/biolab/orange3-educational/pull/158?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=biolab) | Coverage Δ | | |---|---|---| | [...trib/educational/widgets/owpolynomialregression.py](https://codecov.io/gh/biolab/orange3-educational/pull/158?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=biolab#diff-b3JhbmdlY29udHJpYi9lZHVjYXRpb25hbC93aWRnZXRzL293cG9seW5vbWlhbHJlZ3Jlc3Npb24ucHk=) | `93.92% <100.00%> (+0.09%)` | :arrow_up: | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=biolab). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=biolab):umbrella: View full report in Codecov by Sentry.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.
Issue
Polynomial features can become huge due to the exponentiation of numbers. Huge numbers (and number differences) can cause model algorithms to be numerically unstable. For example, a higher degree of polynomial features should result in a better fit to the curve, but when numbers are high and linear regression weight becomes really small, it is not the case anymore (because of numeric instability):
Description of changes
It can be fixed with normalization, but as discussed with Tomaz, it is better just to warn since it is an educational add-on, and we do not want to do any values changing in the background (which would also affect the model weights).
Includes