EDA Toolkit 0.0.29

What's Changed

• fix(table1): handle NaN categories in groupby by @lshpaner in #130

Full Changelog: 0.0.28...0.0.29

EDA Toolkit 0.0.28

What's Changed

• feat: grouped_distributions, table1, box_violin by @lshpaner in #128

Full Changelog: 0.0.27...0.0.28

EDA Toolkit 0.0.27

Fix missing thousands separator in continuous "Overall" column of generate_table1()

Inside generate_table1, the make_overall helper was formatting mean and SD with :.{n}f instead of :,.{n}f, so values like
189664.13 appeared without commas. Added the , flag to both format strings. No behavioral change to categorical Overall or any
other column.

EDA Toolkit 0.0.26

What's Changed

• ENH: Add Overall column to generate_table1 by @lshpaner in #126
• refactor: remove duplicate _flag_iqr and _flag_zscore definitions from detect_outliers by @lshpaner in #127

Full Changelog: 0.0.25...0.0.26

EDA Toolkit 0.0.25

Fix: correct p-value precision loss in Bonferroni/BH adjustment

• Previously, p-values were rounded before being fed into the multiple comparison correction block, causing small but meaningful errors in adjusted values. Raw p-values are now stored as _raw_pval internally, the correction block reads from those, and a cleanup pass removes the key before DataFrame construction. All rounding is handled consistently via decimal_places at the formatting stage.

Added new normality_tests() function to eda_toolkit.

New Features

• Batch normality testing across numeric columns via Shapiro-Wilk,
  D'Agostino K², and Anderson-Darling
• alpha, tests, features, and decimal_places parameters
• Shapiro-Wilk auto-skipped for n > 5,000 with printed explanation
• Anderson-Darling uses method="interpolate" on scipy >= 1.17
  for real p-values; falls back to critical value comparison on
  older scipy
• Returns tidy summary DataFrame with Variable, Test,
  Statistic, P-value, Normal columns
• 40 pytest tests added

EDA Toolkit 0.0.24

Version 0.0.24

New Features

• detect_outliers(): New batch outlier detection function supporting
  IQR, Z-score, and Isolation Forest methods with groupby, flag_col,
  return_mask, return_bounds, and verbose parameters
• flex_corr_matrix(): Added corr_method (pearson/spearman/kendall),
  show_significance, significance_level, significance_method (stars/mask),
  significance_legend_x, filter_significance, and image_filename parameters

Enhancements

• read_csv_with_progress(): Added chunksize, low_memory, and **kwargs
  passthrough to pd.read_csv

Bug Fixes

• flex_corr_matrix(): Fixed title misalignment, -0.00 display artifact,
  diagonal now shows 1.00 in both stars and mask modes; fixed abs() stripping
  negative signs from significance annotations
• detect_outliers(): Added observed=True to groupby calls to silence
  pandas FutureWarning

Dependency Updates

• Removed upper caps on matplotlib, scikit-learn, and tqdm to prevent
  Colab downgrades

Infrastructure

• _data_manager_utils.py: Extracted _flag_iqr and _flag_zscore helpers
• new pytest tests added across detect_outliers, flex_corr_matrix,
  and read_csv_with_progress
• README stripped to lean landing page; full docs at
  datasciencedynamics.com/eda_toolkit_docs/

EDA Toolkit 0.0.23

What's Changed

• enhance stacked_crosstab_plot params & visuals by @lshpaner in #119
• fix deps, bump python floor to 3.8 by @lshpaner in #120
• fix: patch box_violin_plot bugs and add features by @lshpaner in #121
• (+) read_csv_with_progress by @lshpaner in #122
• enhance flex_corr_matrix with sig & filters by @lshpaner in #123
• Fix generate table1 ls by @lshpaner in #124
• add detect_outliers with iqr/zscore/isoforest by @lshpaner in #125

Full Changelog: 0.0.22...0.0.23

EDA Toolkit 0.0.22

• Renamed conditional_histograms to grouped_distributions
• Updated SciPy version requirement in README

EDA Toolkit 0.0.21

• Fixed dependency resolution for Python 3.11+ by updating SciPy version constraints.
• Improved compatibility with Google Colab and Python 3.12 environments.
• Updated package acknowledgements and metadata in __init__.py.
• No API changes.

EDA Toolkit 0.0.20

What's Changed

• flexible summarize_combinations sans excel by @lshpaner in #104
• Groupby imputer by @Oscar-Gil-Data in #109
• Refactor kde_distributions and extract density overlay logic into plot utilities by @lshpaner in #110
• Introduce Enhanced Q-Q Plot Support for Distribution GOF Diagnostics by @lshpaner in #111
• Add ECDF plot type to data_doctor; add SciPy to requirements by @Oscar-Gil-Data in #112
• Add conditional histograms and unify figure-saving utilities by @lshpaner in #114
• DataFrame memory management utility, del_inactive_dataframes by @Oscar-Gil-Data in #115

Full Changelog: 0.0.19...0.0.20