10000 Revert "DOC Fixes whats new" · scikit-learn/scikit-learn@c4b7eab · GitHub
[go: up one dir, main page]
Skip to content

Commit c4b7eab

Browse files
thomasjpfanthomasjpfan
committed
Revert "DOC Fixes whats new"
This reverts commit 5b57944.
1 parent 1db6633 commit c4b7eab

File tree

3 files changed

+183
-0
lines changed

3 files changed

+183
-0
lines changed

‎doc/whats_new.rst

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -12,6 +12,7 @@ on libraries.io to be notified when new versions are released.
1212
.. toctree::
1313
:maxdepth: 1
1414

15+
Version 1.2 <whats_new/v1.2.rst>
1516
Version 1.1 <whats_new/v1.1.rst>
1617
Version 1.0 <whats_new/v1.0.rst>
1718
Version 0.24 <whats_new/v0.24.rst>

‎doc/whats_new/v1.1.rst

Lines changed: 23 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -66,6 +66,26 @@ Changelog
6666
- |Fix| Avoid spurious warning in :class:`decomposition.IncrementalPCA` when
6767
`n_samples == n_components`. :pr:`23264` by :user:`Lucy Liu <lucyleeow>`.
6868

69+
:mod:`sklearn.feature_selection`
70+
................................
71+
72+
- |Fix| The `partial_fit` method of :class:`feature_selection.SelectFromModel`
73+
now conducts validation for `max_features` and `feature_names_in` parameters.
74+
:pr:`23299` by :user:`Long Bao <lorentzbao>`.
75+
76+
:mod:`sklearn.decomposition`
77+
............................
78+
79+
- |Fix| Avoid spurious warning in :class:`decomposition.IncrementalPCA` when
80+
`n_samples == n_components`. :pr:`23264` by :user:`Lucy Liu <lucyleeow>`.
81+
82+
:mod:`sklearn.feature_selection`
83+
................................
84+
85+
- |Fix| The `partial_fit` method of :class:`feature_selection.SelectFromModel`
86+
now conducts validation for `max_features` and `feature_names_in` parameters.
87+
:pr:`23299` by :user:`Long Bao <lorentzbao>`.
88+
6989
:mod:`sklearn.metrics`
7090
......................
7191

@@ -633,6 +653,9 @@ Changelog
633653
:class:`ensemble.ExtraTreesClassifier`.
634654
:pr:`20803` by :user:`Brian Sun <bsun94>`.
635655

656+
- |Efficiency| Improve runtime performance of :class:`ensemble.IsolationForest`
657+
by skipping repetitive input checks. :pr:`23149` by :user:`Zhehao Liu <MaxwellLZH>`.
658+
636659
:mod:`sklearn.feature_extraction`
637660
.................................
638661

‎doc/whats_new/v1.2.rst

Lines changed: 159 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,159 @@
1+
.. include:: _contributors.rst
2+
3+
.. currentmodule:: sklearn
4+
5+
.. _changes_1_2:
6+
7+
Version 1.2.0
8+
=============
9+
10+
**In Development**
11+
12+
.. include:: changelog_legend.inc
13+
14+
Changed models
15+
--------------
16+
17+
The following estimators and functions, when fit with the same data and
18+
parameters, may produce different models from the previous version. This often
19+
occurs due to changes in the modelling logic (bug fixes or enhancements), or in
20+
random sampling procedures.
21+
22+
- |Fix| :class:`manifold.TSNE` now throws a `ValueError` when fit with
23+
`perplexity>=n_samples` to ensure mathematical correctness of the algorithm.
24+
:pr:`10805` by :user:`Mathias Andersen <MrMathias>` and
25+
:pr:`23471` by :user:`Meekail Zain <micky774>`
26+
27+
Changes impacting all modules
28+
-----------------------------
29+
30+
- |Enhancement| Finiteness checks (detection of NaN and infinite values) in all
31+
estimators are now significantly more efficient for float32 data by leveraging
32+
NumPy's SIMD optimized primitives.
33+
:pr:`23446` by :user:`Meekail Zain <micky774>`
34+
35+
Changelog
36+
---------
37+
38+
..
39+
Entries should be grouped by module (in alphabetic order) and prefixed with
40+
one of the labels: |MajorFeature|, |Feature|, |Efficiency|, |Enhancement|,
41+
|Fix| or |API| (see whats_new.rst for descriptions).
42+
Entries should be ordered by those labels (e.g. |Fix| after |Efficiency|).
43+
Changes not specific to a module should be listed under *Multiple Modules*
44+
or *Miscellaneous*.
45+
Entries should end with:
46+
:pr:`123456` by :user:`Joe Bloggs <joeongithub>`.
47+
where 123456 is the *pull request* number, not the issue number.
48+
49+
:mod:`sklearn.cluster`
50+
......................
51+
52+
- |Enhancement| The `predict` and `fit_predict` methods of :class:`cluster.OPTICS` now
53+
accept sparse data type for input data. :pr:`14736` by :user:`Hunt Zhan <huntzhan>`,
54+
:pr:`20802` by :user:`Brandon Pokorny <Clickedbigfoot>`,
55+
and :pr:`22965` by :user:`Meekail Zain <micky774>`.
56+
57+
- |Enhancement| :class:`cluster.Birch` now preserves dtype for `numpy.float32`
58+
inputs. :pr:`22968` by `Meekail Zain <micky774>`.
59+
60+
:mod:`sklearn.ensemble`
61+
.......................
62+
63+
- |Efficiency| Improve runtime performance of :class:`ensemble.IsolationForest`
64+
by avoiding data copies. :pr:`23252` by :user:`Zhehao Liu <MaxwellLZH>`.
65+
66+
:mod:`sklearn.decomposition`
67+
............................
68+
69+
- |Enhancement| :class:`decomposition.FastICA` now allows the user to select
70+
how whitening is performed through the new `whiten_solver` parameter, which
71+
supports `svd` and `eigh`. `whiten_solver` defaults to `svd` although `eigh`
72+
may be faster and more memory efficient in cases where
73+
`num_features > num_samples`. An additional `sign_flip` parameter is added.
74+
When `sign_flip=True`, then the output of both solvers will be reconciled
75+
during `fit` so that their outputs match. This may change the output of the
76+
default solver, and hence may not be backwards compatible.
77+
:pr:`11860` by :user:`Pierre Ablin <pierreablin>`,
78+
:pr:`22527` by :user:`Meekail Zain <micky774>` and `Thomas Fan`_.
79+
80+
:mod:`sklearn.feature_selection`
81+
................................
82+
- |Fix| :class:`feature_selection.SelectFromModel` defaults to selection
83+
threshold 1e-5 when the estimator is either :class:`linear_model.ElasticNet`
84+
or :class:`linear_model.ElasticNetCV` with `l1_ratio` equals 1 or
85+
:class:`linear_model.LassoCV`. :pr:`23636` by :user:`Hao Chun Chang
86+
<haochunchang>`
87+
88+
:mod:`sklearn.impute`
89+
.....................
90+
91+
- |Fix| :class:`impute.SimpleImputer` uses the dtype seen in `fit` for
92+
`transform` when the dtype is object. :pr:`22063` by `Thomas Fan`_.
93+
94+
:mod:`sklearn.linear_model`
95+
...........................
96+
97+
- |Fix| Use dtype-aware tolerances for the validation of gram matrices (passed by users
98+
or precomputed). :pr:`22059` by :user:`Malte S. Kurz <MalteKurz>`.
99+
100+
- |Fix| Fixed an error in :class:`linear_model.LogisticRegression` with
101+
`solver="newton-cg"`, `fit_intercept=True`, and a single feature. :pr:`23608`
102+
by `Tom Dupre la Tour`_.
103+
104+
:mod:`sklearn.metrics`
105+
......................
106+
107+
- |Feature| :func:`class_likelihood_ratios` is added to compute the positive and
108+
negative likelihood ratios derived from the confusion matrix
109+
of a binary classification problem. :pr:`22518` by
110+
:user:`Arturo Amor <ArturoAmorQ>`.
111+
112+
- |Fix| Fixed error message of :class:`metrics.coverage_error` for 1D array input.
113+
:pr:`23548` by :user:`Hao Chun Chang <haochunchang>`.
114+
115+
:mod:`sklearn.neighbors`
116+
........................
117+
118+
- |Enhancement| :class:`neighbors.KernelDensity` bandwidth parameter now accepts
119+
definition using Scott's and Silvermann's estimation methods.
120+
:pr:`10468` by :user:`Ruben <icfly2>` and :pr:`22993` by
121+
:user:`Jovan Stojanovic <jovan-stojanovic>`.
122+
123+
:mod:`sklearn.feature_selection`
124+
................................
125+
126+
- |Fix| The `partial_fit` method of :class:`feature_selection.SelectFromModel`
127+
now conducts validation for `max_features` and `feature_names_in` parameters.
128+
:pr:`23299` by :user:`Long Bao <lorentzbao>`.
129+
130+
:mod:`sklearn.tree`
131+
...................
132+
133+
- |Fix| Fixed invalid memory access bug during fit in
134+
:class:`tree.DecisionTreeRegressor` and :class:`tree.DecisionTreeClassifier`.
135+
:pr:`23273` by `Thomas Fan`_.
136+
137+
:mod:`sklearn.utils`
138+
....................
139+
140+
- |Enhancement| :func:`utils.extmath.randomized_svd` now accepts an argument,
141+
`lapack_svd_driver`, to specify the lapack driver used in the internal
142+
deterministic SVD used by the randomized SVD algorithm.
143+
:pr:`20617` by :user:`Srinath Kailasa <skailasa>`
144+
145+
:mod:`sklearn.manifold`
146+
.......................
147+
148+
- |Fix| :class:`manifold.TSNE` now throws a `ValueError` when fit with
149+
`perplexity>=n_samples` to ensure mathematical correctness of the algorithm.
150+
:pr:`10805` by :user:`Mathias Andersen <MrMathias>` and
151+
:pr:`23471` by :user:`Meekail Zain <micky774>`
152+
153+
Code and Documentation Contributors
154+
-----------------------------------
155+
156+
Thanks to everyone who has contributed to the maintenance and improvement of
157+
the project since version 1.1, including:
158+
159+
TODO: update at the time of the release.

0 commit comments

Comments
 (0)
0