8000 ENH: zero_division parameter for classification… (#14900) · crankycoder/scikit-learn@7f079e3 · GitHub
[go: up one dir, main page]
Skip to content

Commit 7f079e3

Browse files
marctorsocthomasjpfan
marctorsoc
authored and
thomasjpfan
committed
ENH: zero_division parameter for classification… (scikit-learn#14900)
1 parent 6d53d06 commit 7f079e3

File tree

3 files changed

+397
-128
lines changed

3 files changed

+397
-128
lines changed

‎doc/whats_new/v0.22.rst

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -402,6 +402,13 @@ Changelog
402402
:mod:`sklearn.metrics`
403403
......................
404404

405+
- |Feature| Added a new parameter ``zero_division`` to multiple classification
406+
metrics: :func:`precision_score`, :func:`recall_score`, :func:`f1_score`,
407+
:func:`fbeta_score`, :func:`precision_recall_fscore_support`,
408+
:func:`classification_report`. This allows to set returned value for
409+
ill-defined metrics.
410+
:pr:`14900` by :user:`Marc Torrellas Socastro <marctorrellas>`.
411+
405412
- |MajorFeature| :func:`metrics.plot_roc_curve` has been added to plot roc
406413
curves. This function introduces the visualization API described in
407414
the :ref:`User Guide <visualizations>`. :pr:`14357` by `Thomas Fan`_.

0 commit comments

Comments
 (0)
0