Implement BasePredictionIntervals
#86
Conversation
|
🚀 Deployed on https://deploy-preview-86--etna-docs.netlify.app |
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## master #86 +/- ##
==========================================
+ Coverage 89.27% 89.28% +0.01%
==========================================
Files 206 208 +2
Lines 12869 12913 +44
==========================================
+ Hits 11489 11530 +41
- Misses 1380 1383 +3
☔ View full report in Codecov by Sentry. |
| def test_forecast_intervals_exists(example_tsds, pipeline_name, expected_columns, request): | ||
| pipeline = request.getfixturevalue(pipeline_name) | ||
|
|
||
| intervals_pipeline = DummyPredictionIntervals(pipeline=pipeline) |
There was a problem hiding this comment.
How is this code different from run_base_pipeline_compat_check?
| ) | ||
| def test_default_params_to_tune(pipeline): | ||
| intervals_pipeline = DummyPredictionIntervals(pipeline=pipeline) | ||
| assert intervals_pipeline.params_to_tune() == pipeline.params_to_tune() |
There was a problem hiding this comment.
It shouldn't work like this. Because the pipeline. prefix should be added to each param in case of intervals_pipeline.
There was a problem hiding this comment.
What are we going to do with this check? Are we going to fix it in some other task related to params_to_tune?
| "expected_columns", | ||
| ({"target_lower", "target_upper"},), | ||
| ) | ||
| @pytest.mark.parametrize("pipeline_name", ("naive_pipeline", "naive_pipeline_with_transforms")) |
There was a problem hiding this comment.
Do we even need this test if we have test_pipelines_forecast_intervals.
Before submitting (must do checklist)
Proposed Changes
Closing issues
closes #70