TST Remove redundant max iter in sklearn/linear_model/tests #14622
Conversation
|
tests are failing which probably indicates that you changed behavior somewhere. |
|
still failing... |
|
@amueller I know it's still failing but I'm trying to find wich changes make it fail |
| @@ -1542,7 +1542,7 @@ def test_gradient_squared_epsilon_insensitive(): | ||
| def test_multi_thread_multi_class_and_early_stopping(): | ||
| # This is a non-regression test for a bad interaction between | ||
| # early stopping internal attribute and thread-based parallelism. | ||
| clf = SGDClassifier(alpha=1e-3, tol=1e-3, max_iter=1000, |
There was a problem hiding this comment.
This needs to be reverted because:
scikit-learn/sklearn/linear_model/tests/test_sgd.py
Lines 75 to 77 in 1a14920
Changes the max_iter=5 for all tests in this file.
| @@ -1560,7 +1560,7 @@ def test_multi_core_gridsearch_and_early_stopping(): | ||
| 'n_iter_no_change': [5, 10, 50], | ||
| } | ||
|
|
||
| clf = SGDClassifier(tol=1e-2, max_iter=1000, early_stopping=True, |
There was a problem hiding this comment.
This needs to be reverted because:
scikit-learn/sklearn/linear_model/tests/test_sgd.py
Lines 75 to 77 in 1a14920
Changes the max_iter=5 for all tests in this file.
|
Merging because this is a simple PR that removes redundant defaults. Thank you for working on this @qdeffense ! |
…earn#14622) Co-authored-by: Thomas J. Fan <thomasjpfan@gmail.com>
Reference Issues/PRs
Partially address #14351
What does this implement/fix? Explain your changes.
Remove redundant max_iter in examples and tests