-
Notifications
You must be signed in to change notification settings - Fork 88
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
Integration tests fix #1171
Integration tests fix #1171
Conversation
predefined_model = Pipeline(PipelineNode('ridge', nodes_from=[PipelineNode('lagged')])) | ||
|
||
automl.fit(path, target='sea_height', predefined_model=predefined_model) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Эти строки не нарушают контракт юнит-теста?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Нарушают смысл, потому что тут идея в том чтобы протестировать оптимизацию структуры пайплайна с использованием mock-а для удаленной вычислительной среды. А с predefined_model оптимизатор вообще не стартует.
А падал именно этот тест?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Да, модель не всегда вычисляется, зачастую при обучении выдаётся ошибка "No models were found". Но когда срабатывает, то будет пайплайн со скриншота.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hm. Ну давай пока просто отключим этот тест, функциональность не критичная. Пометь плз как issue.
Codecov Report
@@ Coverage Diff @@
## master #1171 +/- ##
=======================================
Coverage 79.70% 79.70%
=======================================
Files 141 141
Lines 9851 9851
=======================================
Hits 7852 7852
Misses 1999 1999 |
@@ -76,13 +76,14 @@ def run_exogenous_experiment(path_to_file, len_forecast=250, with_exog=True, vis | |||
task_params=task.task_params, | |||
timeout=10, | |||
initial_assumption=pipeline, | |||
max_pipeline_fit_time=1, | |||
available_operations=['lagged', 'ridge', 'exog_ts'], |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Почему прищлось ограничить выбор моделей? Так по идее ничего не скомпозируется
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Сделал так, потому что без этого выбираться модели, которые не обучаются на данных теста. Например, "sparse_lagged" выбиралась и вызывала ошибку, но не только.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Стоит добавить сюда табличных моделек (хотя бы rf и knn)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Регрессионные версии, разумеется
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Вообще убрал эти правки, кажется, в новой версии master, это уже исправно работает.
Hello @IIaKyJIuH! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found: There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻 Comment last updated at 2023-09-25 07:41:11 UTC |
Hello @IIaKyJIuH! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found: There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻 Comment last updated at 2023-09-22 15:13:27 UTC |
predict = automl.predict(path) | ||
shutil.rmtree(os.path.join(fedot_project_root(), 'test', 'data', 'remote', 'fitted_pipeline')) # recursive deleting | ||
assert predict is not None | ||
# TODO: "No models were found" error as for 22.09.2023 appears. Fix it. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Вроде можно в pytest фикстурой отключать тесты
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Спасибо, не знал.
Так намного красивее.
Made all integration tests pass.