Skip to content

2070 user specified date feature #2217

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

Merged
merged 27 commits into from
May 4, 2021
Merged

Conversation

ParthivNaresh
Copy link
Contributor

@ParthivNaresh ParthivNaresh commented May 1, 2021

Fixes #2070

This is a refiling of the old PR with the previous (main) version of ARIMA. After this is merged, ARIMA will be updated in a different PR using sktime as opposed to statsmodels.

@codecov
Copy link

codecov bot commented May 1, 2021

Codecov Report

Merging #2217 (2a117c3) into main (48ab46a) will increase coverage by 0.1%.
The diff coverage is 100.0%.

Impacted file tree graph

@@            Coverage Diff            @@
##             main    #2217     +/-   ##
=========================================
+ Coverage   100.0%   100.0%   +0.1%     
=========================================
  Files         288      288             
  Lines       24449    24462     +13     
=========================================
+ Hits        24431    24444     +13     
  Misses         18       18             
Impacted Files Coverage Δ
evalml/automl/utils.py 100.0% <ø> (ø)
evalml/preprocessing/utils.py 100.0% <ø> (ø)
evalml/tests/component_tests/test_components.py 100.0% <ø> (ø)
...s/prediction_explanations_tests/test_explainers.py 100.0% <ø> (ø)
evalml/utils/gen_utils.py 99.6% <ø> (ø)
evalml/automl/automl_search.py 100.0% <100.0%> (ø)
...omponents/estimators/regressors/arima_regressor.py 100.0% <100.0%> (ø)
.../transformers/preprocessing/datetime_featurizer.py 100.0% <100.0%> (ø)
...rmers/preprocessing/delayed_feature_transformer.py 100.0% <100.0%> (ø)
.../pipelines/time_series_classification_pipelines.py 100.0% <100.0%> (ø)
... and 17 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 48ab46a...2a117c3. Read the comment docs.

@ParthivNaresh ParthivNaresh self-assigned this May 3, 2021
# Conflicts:
#	evalml/automl/automl_search.py
#	evalml/tests/automl_tests/test_automl_search_classification.py
#	evalml/tests/automl_tests/test_automl_search_regression.py
#	evalml/tests/pipeline_tests/test_time_series_baseline_pipeline.py
@ParthivNaresh ParthivNaresh marked this pull request as ready for review May 3, 2021 18:37
@ParthivNaresh ParthivNaresh requested review from dsherry and freddyaboulton and removed request for dsherry and freddyaboulton May 3, 2021 18:37
@@ -1,6 +1,8 @@
name: Build Conda Package

on:
pull_request:
types: [ opened, synchronize ]
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we want to keep this for feature builds moving forward?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We can try it out and if it's too slow we can undo the change?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sure I am a fan!

@dsherry
Copy link
Contributor

dsherry commented May 3, 2021

@ParthivNaresh I see this in the PR description:

After this is merged, ARIMA will be updated using sktime as opposed to statsmodels.

This is no longer true, correct? If so please delete :)

@ParthivNaresh
Copy link
Contributor Author

@dsherry Sorry I meant after this is merged, ARIMA will be updated in another PR :)

Copy link
Contributor

@freddyaboulton freddyaboulton left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@ParthivNaresh Looks great!

@@ -1,6 +1,8 @@
name: Build Conda Package

on:
pull_request:
types: [ opened, synchronize ]
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We can try it out and if it's too slow we can undo the change?

if not y_none:
y_train, y_test = y.iloc[train], y.iloc[test]
pd.testing.assert_index_equal(y_train.index, answer[i][0])
pd.testing.assert_index_equal(y_test.index, answer[i][1])


@pytest.mark.parametrize("max_delay,gap,date_index", [(0, 0, "Date"), (1, 0, "Date"), (0, 1, "Date")])
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we can delete this test? The ts data splitter doesn't sort the data.

@@ -318,7 +318,7 @@ def __init__(self, dummy_int_parameter=0, dummy_categorical_parameter='dummy', d

pipeline = BinaryClassificationPipeline([MockEstimator])
algo = IterativeAlgorithm(allowed_pipelines=[pipeline, pipeline, pipeline],
pipeline_params={'pipeline': {"gap": 2, "max_delay": 10}},
pipeline_params={'pipeline': {'date_index': "Date", "gap": 2, "max_delay": 10}},
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍

Copy link
Contributor

@dsherry dsherry left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Excellent!!

@@ -2,6 +2,7 @@ Release Notes
-------------
**Future Releases**
* Enhancements
* Added `date_index` as a required parameter for TimeSeries problems :pr:`2217`
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍

Copy link
Contributor

@bchen1116 bchen1116 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍

@ParthivNaresh ParthivNaresh merged commit 035510d into main May 4, 2021
This was referenced May 4, 2021
@freddyaboulton freddyaboulton deleted the 2070-User-Specified-Date-Feature branch May 13, 2022 15:34
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

Successfully merging this pull request may close these issues.

Time series: require user to specify a date feature
4 participants