Skip to content
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.

Sign up for GitHub

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

Jump to bottom

change_weak_segments_na_logic #2709

Merged
merged 8 commits into from
Dec 6, 2023
Merged

change_weak_segments_na_logic #2709

Show file tree
Hide file tree
Changes from 4 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
3b70a83
change_weak_segments_na_logic
JKL98ISR Dec 5, 2023
288f408
Merge branch 'main' into jonatan/llm-1167-insights-query-that-result-…
JKL98ISR Dec 5, 2023
fba9f4f
fix_tests
JKL98ISR Dec 5, 2023
b300ddd
Merge branch 'jonatan/llm-1167-insights-query-that-result-0-samples-w…
JKL98ISR Dec 5, 2023
3eb5fe9
lint
JKL98ISR Dec 5, 2023
655507a
fix_test
JKL98ISR Dec 5, 2023
aa61f9f
fix_test
JKL98ISR Dec 5, 2023
1510672
don't create heatmap display for too-small segments
noamzbr Dec 6, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
14 changes: 7 additions & 7 deletions deepchecks/utils/dataframes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,14 +43,14 @@ def default_fill_na_per_column_type(df: pd.DataFrame, cat_features: t.Optional[t

def default_fill_na_series(col: pd.Series, is_cat_column: t.Optional[bool] = None) -> t.Optional[pd.Series]:
"""Fill NaN values based on column type if possible otherwise returns None."""
if is_cat_column:
return col.astype('object').fillna('None')
if is_cat_column and 'None' not in col.astype('object').unique():
return col.astype('object').fillna('None')
elif is_numeric_dtype(col):
return col.astype('float64').fillna(col.mean())
else:
common_values_list = col.mode()
if isinstance(common_values_list, pd.Series) and len(common_values_list) > 0:
return col.fillna(common_values_list[0])
return col.astype('float64').fillna(np.nan)

common_values_list = col.mode()
if isinstance(common_values_list, pd.Series) and len(common_values_list) > 0:
return col.fillna(common_values_list[0])
return None


Expand Down
4 changes: 2 additions & 2 deletions tests/nlp/checks/model_evaluation/weak_segment_performance_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,7 @@ def test_column_with_nones(tweet_emotion_train_test_textdata, tweet_emotion_trai

# Assert
assert_that(result.value['avg_score'], close_to(0.707, 0.01))
assert_that(len(result.value['weak_segments_list']), equal_to(8))
assert_that(len(result.value['weak_segments_list']), equal_to(4))
assert_that(result.value['weak_segments_list'].iloc[0, 0], close_to(0.305, 0.01))


Expand Down Expand Up @@ -168,7 +168,7 @@ def test_multilabel_just_dance(just_dance_train_test_textdata, just_dance_train_

# Assert
assert_that(result.value['avg_score'], close_to(0.615, 0.001))
assert_that(len(result.value['weak_segments_list']), equal_to(5))
assert_that(len(result.value['weak_segments_list']), equal_to(3))
assert_that(result.value['weak_segments_list'].iloc[0, 0], close_to(0.433, 0.01))


Expand Down