There test CI has a lot of warnings once the tests are run. I would like to clean most of these up (where possible). I have created a list of the general warnings we can begin cleaning up.
Deprecation Warnings
[ ] _base.py Future Warnings
[ ] Address the deprecation of SklearnClassifierPipeline class. It's advised to use ClassifierPipeline or the sklearn pipeline instead. See _base.py:569.
[ ] Update the initialization method for ClassifierPipeline to avoid breakage when the positions of the classifier and transformers arguments are swapped. See _base.py:569.
[ ] _reduce.py Deprecation Warnings
[ ] Correct the use of converting an array with ndim > 0 to a scalar to avoid future errors. This affects various test files including test_all_forecasters.py, test_reduce.py, test_autoensemble.py, test_reduce_global.py, and test_pipeline.py. See _reduce.py:629 and _reduce.py:937.
Future Warnings
[ ] Deprecated Function get_examples
[ ] Replace the deprecated get_examples function across various datatypes tests (test_check.py, test_convert.py, test_vectorize.py) with its newer version from the testing module.
[ ] Statsmodels Index Support
[ ] Address the future warning regarding the lack of supported index in test_all_forecasters.py. This warning originates from the Statsmodels library.
[ ] Joblib Deprecation Warnings
[ ] Update code to avoid using deprecated Joblib features related to ast.Num and n attribute, as seen in various tests (test_softdeps.py, test_window_summarizer.py, test_optionalpassthrough.py, test_reduce_global.py).
Class and Method Deprecations
[x] Deprecated Classes warnings
[x] Update or replace usage of deprecated classes such as MatrixProfileClassifier, BaseForecastingErrorMetric, HMM, and Hidalgo across various tests and modules.
[ ] Sklearn Pipeline and Classifier Warnings
[ ] Address the deprecation of SklearnClassifierPipeline and related warnings in test_all_estimators.py and classification/tests.
[ ] Numpy and Scipy Warnings
[ ] Address deprecation warnings related to numpy array operations and scipy optimizations.
TonyBagnall
commented
3 months ago
There test CI has a lot of warnings once the tests are run. I would like to clean most of these up (where possible). I have created a list of the general warnings we can begin cleaning up.
Deprecation Warnings
Future Warnings
Class and Method Deprecations
Here is a link to a CI run for more details: https://github.com/aeon-toolkit/aeon/actions/runs/8145354708/job/22261460345