instance_id stringlengths 21 53 | repo stringclasses 188
values | language stringclasses 1
value | pull_number int64 20 148k | title stringlengths 6 144 | body stringlengths 0 83.4k | created_at stringdate 2015-09-25 03:17:17 2025-07-10 16:50:35 | problem_statement stringlengths 188 240k | hints_text stringlengths 0 145k | resolved_issues listlengths 1 6 | base_commit stringlengths 40 40 | commit_to_review dict | reference_review_comments listlengths 1 62 | merged_commit stringlengths 40 40 | merged_patch stringlengths 297 9.87M | metadata dict |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
psf__black-2526@b821ed8 | psf/black | Python | 2,526 | Enhance `--verbose` | ### Description
Black would now echo the location that it determined as the root path for the project if `--verbose` is enabled by the user, according to which it chooses the SRC paths, i.e. the absolute path of the project: `{root}/{src}`.
 -> Path:\n+ if not root:\n+ return path\n+ return Path(os.path.relpath(path, start=Path(f\"{root}/../\")))",
"line": null,
"original_line": 19,
"original_st... | c02122a48d42c3fc74c6ca5d762cacf48d4a99ce | diff --git a/CHANGES.md b/CHANGES.md
index bfee1b6f259..f6e8343ed00 100644
--- a/CHANGES.md
+++ b/CHANGES.md
@@ -22,6 +22,8 @@
`values: Tuple[int, ...] = 1, 2, 3`) now implies 3.8+ (#2708)
- For stubs, one blank line between class attributes and methods is now kept if there's
at least one pre-existing blank line... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
psf__black-2005@ab9345c | psf/black | Python | 2,005 | Add --extend-exclude parameter | Adding the `--extend-exclude` parameter to add additional exclude regexes on top of the `--exclude` regexes. This is especially useful if you want to specify _additional_ files/directories on top of the very useful default ones (as demonstrated by the change to our own `pyproject.toml`).
The parameter name also matc... | 2021-02-24T21:30:54Z | Please offer a way to exclude dirs *in addition to the default exclusions*
**Is your feature request related to a problem? Please describe.**
I needed to exclude one auto-generated file so I naively specified `--exclude thefile.py`. Unfortunately that disabled the useful standard exclusions, which I did not realize ... | For comparison, Flake8 does this, named `--extend-exclude=<patterns>`
https://flake8.pycqa.org/en/latest/user/options.html#cmdoption-flake8-extend-exclude
I'd be supportive of doing this, using the `--extend-exclude` name.
I can take a stab at it, with `--extend-exclude`. | [
{
"body": "**Is your feature request related to a problem? Please describe.**\r\n\r\nI needed to exclude one auto-generated file so I naively specified `--exclude thefile.py`. Unfortunately that disabled the useful standard exclusions, which I did not realize until it broke a unit test running on Jenkins.\r\n\r... | b06cd15666e5d766347cda0434dc6c828a96c74a | {
"head_commit": "ab9345ca5b1801073b2fb6a2f4a54523df25d0ec",
"head_commit_message": "Fix flake8 issues",
"patch_to_review": "diff --git a/README.md b/README.md\nindex 178f763c32d..df21d2b1f00 100644\n--- a/README.md\n+++ b/README.md\n@@ -135,11 +135,17 @@ Options:\n hg|\\.mypy_ca... | [
{
"diff_hunk": "@@ -1433,6 +1443,35 @@ def test_get_sources_with_stdin_filename_and_exclude(self) -> None:\n verbose=False,\n include=include,\n exclude=exclude,\n+ extend_exclude=None,\n+ force_exclude=None,\n+ report=... | 3a8f4beee1de321d12f7fdc4dd39dd808673bc6a | diff --git a/README.md b/README.md
index 178f763c32d..5f8b52cb823 100644
--- a/README.md
+++ b/README.md
@@ -135,11 +135,17 @@ Options:
hg|\.mypy_cache|\.nox|\.tox|\.venv|\.svn|_bu
ild|buck-out|build|dist)/]
+ --extend-exclude TEXT Like ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
psf__black-1915@0aa6927 | psf/black | Python | 1,915 | Update Contributing Docs | - Update docs with all new tox hotness
- Test running docs build:
- `sphinx-build -a -b html -W docs/ docs/_build/`
Fixes #1907 | 2021-01-10T23:18:50Z | Update contributing
The correct way to run tests seems to be `tox -e py` (as run on CI) and not `python -m unittest` (that doesn’t even work due to parametrized not being in the Pipfile!) as advised currently. We should also ask contributors to run `tox -e fuzz` too maybe. | [
{
"body": "The correct way to run tests seems to be `tox -e py` (as run on CI) and not `python -m unittest` (that doesn’t even work due to parametrized not being in the Pipfile!) as advised currently. We should also ask contributors to run `tox -e fuzz` too maybe.",
"number": 1907,
"title": "Update cont... | a81702ce027ef6792955d5db02aad60a34c7ec44 | {
"head_commit": "0aa69275c104e480989d518cfe34e6c5ef4bf214",
"head_commit_message": "Remove autogenerated copy pasta",
"patch_to_review": "diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md\nindex a4250f78718..c42d266327e 100644\n--- a/CONTRIBUTING.md\n+++ b/CONTRIBUTING.md\n@@ -34,14 +34,39 @@ If you haven't used `p... | [
{
"diff_hunk": "@@ -34,14 +34,39 @@ If you haven't used `pipenv` before but are comfortable with virtualenvs, just r\n `pip install pipenv` in the virtualenv you're already using and invoke the command above\n from the cloned _Black_ repo. It will do the correct thing.\n \n-Before submitting pull requests, run ... | 025c321d3ec292275b3e57f7cff69784220d985f | diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index a4250f78718..571d870452b 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -34,14 +34,40 @@ If you haven't used `pipenv` before but are comfortable with virtualenvs, just r
`pip install pipenv` in the virtualenv you're already using and invoke the command above
... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Documentation Updates"
} | |
psf__black-2208@e3bbe53 | psf/black | Python | 2,208 | Cover more in the usage docs | I had this laying around so why not use it :)
Resolves GH-2004. | 2021-05-08T19:52:31Z | Mention that "-" is how one specifies standard input
Mention somewhere in --help that "-" is how one specifies standard
input. (No, /dev/stdin etc. won't work.) Else one thinks --pyi is how one does it.
```
--pyi Format all input files like typing stubs
... | [
{
"body": "Mention somewhere in --help that \"-\" is how one specifies standard\r\ninput. (No, /dev/stdin etc. won't work.) Else one thinks --pyi is how one does it.\r\n```\r\n --pyi Format all input files like typing stubs\r\n regardless of file exten... | 1bedc176d19f0e71c24149c696c5a2a71f21e941 | {
"head_commit": "e3bbe53f767f6c50d871f097f58642faf974c52e",
"head_commit_message": "Minor fixes",
"patch_to_review": "diff --git a/docs/usage_and_configuration/the_basics.md b/docs/usage_and_configuration/the_basics.md\nindex 6fd8769b591..1bafb3fe0a8 100644\n--- a/docs/usage_and_configuration/the_basics.md\n+++ ... | [
{
"diff_hunk": "@@ -124,6 +124,148 @@ running `black --help`.\n \n </details>\n \n+### Code input alternatives\n+\n+#### Standard Input\n+\n+_Black_ supports formatting code via stdin, with the result being printed to stdin. Just",
"line": null,
"original_line": 131,
"original_start_line": null,
... | 9a8adab6c0d4b3743ce4e073f5e2256a58c712c3 | diff --git a/docs/usage_and_configuration/the_basics.md b/docs/usage_and_configuration/the_basics.md
index 6fd8769b591..4b7aaf30591 100644
--- a/docs/usage_and_configuration/the_basics.md
+++ b/docs/usage_and_configuration/the_basics.md
@@ -124,6 +124,149 @@ running `black --help`.
</details>
+### Code input alter... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Documentation Updates"
} | |
psf__black-2472@895695b | psf/black | Python | 2,472 | Treat functions/classes in blocks as if they're nested | ### Description
One curveball is that we still want two preceding newlines before blocks
that are probably logically disconnected. In other words:
if condition:
def foo():
return "hi"
# <- aside: this is the goal of this commit
else:
def f... | 2021-09-02T18:03:01Z | Inconsistent spacing around top level functions
Operating system: macOS 10.14.2
Python version: 3.7.1
*Black* version: 18.9b0
Does also happen on master: yes
---
I have a bit of OS-specific code that looks like this (when formatted by black):
```py
import os
if os.name == "posix":
import termios
... | This is indeed the case: https://black.now.sh/?version=stable&state=_Td6WFoAAATm1rRGAgAhARYAAAB0L-Wj4AFlAKtdAD2IimZxl1N_WlwxRhR7MSgE-ANVlfGy-Pf-urovcUul06DGIKtNJ_PI3I6sH4UsuvCCJK3cSn3O4qbqqEoVdC4RVidtOakKqE1C2PRTK-dEU4CIhv140rXetsz6FlHHNcogH0t3yNeHxqr6iAwuCmSu430fNaT3oxW598BtIbDYmvw9WdVHpVPmNBsa5a2N64u1JlxqaEp3vqccXdLe... | [
{
"body": "Operating system: macOS 10.14.2\r\nPython version: 3.7.1\r\n*Black* version: 18.9b0\r\nDoes also happen on master: yes\r\n\r\n---\r\n\r\nI have a bit of OS-specific code that looks like this (when formatted by black):\r\n\r\n```py\r\nimport os\r\n\r\nif os.name == \"posix\":\r\n import termios\r\n... | f1813e31b6deed0901c8a7fb1f102b9af53de351 | {
"head_commit": "895695b87cf0e953ffd42d715efba8a31e3ae4fc",
"head_commit_message": "Treat functions/classes in blocks as if they're nested\n\nOne curveball is that we still want two preceding newlines before blocks\nthat are probably logically disconnected. In other words:\n\n if condition:\n\n def foo()... | [
{
"diff_hunk": "@@ -447,11 +447,30 @@ def _maybe_empty_lines(self, current_line: Line) -> Tuple[int, int]:\n before = 0\n depth = current_line.depth\n while self.previous_defs and self.previous_defs[-1] >= depth:\n- self.previous_defs.pop()\n if self.is_pyi:\n ... | 4253c26094980962120e866607de3039c6045fe0 | diff --git a/CHANGES.md b/CHANGES.md
index 5b648225ad7..59042914174 100644
--- a/CHANGES.md
+++ b/CHANGES.md
@@ -17,6 +17,7 @@
- Add `flake8-simplify` and `flake8-comprehensions` plugins (#2653)
- Fix determination of f-string expression spans (#2654)
- Fix parser error location on invalid syntax in a `match` statem... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Code Style, Linting, Formatting Fixes"
} |
psf__black-1780@c2ddf0a | psf/black | Python | 1,780 | Provide a stdin-filename to allow stdin to respect force-exclude rules | Fix #1776 | 2020-10-22T20:10:34Z | Provide a --stdin-display-name option like flake8
I'm trying to use `black` as a fixer on ale on vim. It works fine and my config file is respected, except for the `exclude` and `force-exclude` files/folders.
The problem is that the tool calls black passing the code through stdin, and thus, black doesn't know the fi... | [
{
"body": "I'm trying to use `black` as a fixer on ale on vim. It works fine and my config file is respected, except for the `exclude` and `force-exclude` files/folders.\r\n\r\nThe problem is that the tool calls black passing the code through stdin, and thus, black doesn't know the filename that is being parsed... | 407052724fa1c97ee8bcd4e96de650def00be03e | {
"head_commit": "c2ddf0adaa23cd4f8ac154af97f1763cea44d604",
"head_commit_message": "Provide a stdin-filename to allow stdin to respect exclude/force-exclude rules\n\nThis will allow automatic tools to enforce the project's\nexclude/force-exclude rules even if they pass the file through stdin to\nupdate its buffer.... | [
{
"diff_hunk": "@@ -626,8 +644,6 @@ def get_sources(\n gitignore,\n )\n )\n- elif s == \"-\":\n- sources.add(p)\n elif p.is_file():",
"line": null,
"original_line": 647,
"original_start_line": null,
"path": "src/black/__in... | 7320a10ec942fd8256192f0714e6db9f8d304aa6 | diff --git a/README.md b/README.md
index 47472810114..0615ba6aa3a 100644
--- a/README.md
+++ b/README.md
@@ -135,6 +135,11 @@ Options:
matching this regex will be excluded even
when they are passed explicitly as arguments.
+ --stdin-filename TEXT ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} | |
psf__black-2343@fc40194 | psf/black | Python | 2,343 | Fix bad formatting of error messages about EOF in multi-line statements | Closes #2317 | 2021-06-20T14:57:09Z | Bad formatting of error messages about EOF in multi-line statements
**Describe the bug**
"EOF in multi-line statement" error is shown to the user as the raw tuple it was returned as, rather than in a nicely formatted message.
**To Reproduce**
For example:
1. Take this (invalid) file:
```py
print(
```
2.... | Many thanks for reporting!
It appears to be about `TokenError` (and potentially also `IndentationError`) exception leaking out from `lib2to3_parse()` call. Should the `lib2to3_parse()` function just be updated to handle these 2 errors or is there some other solution that would be preferred here?
Eh, it seems to be just... | [
{
"body": "**Describe the bug**\r\n\r\n\"EOF in multi-line statement\" error is shown to the user as the raw tuple it was returned as, rather than in a nicely formatted message.\r\n\r\n**To Reproduce**\r\n\r\nFor example:\r\n1. Take this (invalid) file:\r\n```py\r\nprint(\r\n\r\n```\r\n2. Run _Black_ on it with... | 136930fccb99320865622e12ffc21bdd45fd7501 | {
"head_commit": "fc40194c38aa540c020b3ae31e29167558333d5b",
"head_commit_message": "Changing message for TokenError",
"patch_to_review": "diff --git a/CHANGES.md b/CHANGES.md\nindex 6714a9d9eb2..15eac739e66 100644\n--- a/CHANGES.md\n+++ b/CHANGES.md\n@@ -44,6 +44,7 @@\n - Fix regression where `R` prefixes would ... | [
{
"diff_hunk": "@@ -92,6 +93,12 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -\n except IndexError:\n faulty_line = \"<line number missing in source>\"\n exc = InvalidInput(f\"Cannot parse: {lineno}:{column}: {faulty_line}\")\n+\n+ ... | 339a1738e22dbd7d36a04aaa0620a31700d6a9d4 | diff --git a/CHANGES.md b/CHANGES.md
index e5f4a1fdf82..5d89c71f580 100644
--- a/CHANGES.md
+++ b/CHANGES.md
@@ -19,6 +19,7 @@
- Add `flake8-simplify` and `flake8-comprehensions` plugins (#2653)
- Fix determination of f-string expression spans (#2654)
- Fix parser error location on invalid syntax in a `match` statem... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
psf__black-1940@55a46b8 | psf/black | Python | 1,940 | Add `version` to github action (and rewrite the whole thing while at it) | Allows setting the exact black version use with the github action.
This is useful to be able to avoid needing to deal with black formatting changes introduced in new black versions in unrelated pull-requests. | 2021-01-22T23:51:53Z | Allow picking a specific black version in the github action
Sometimes we want to stick with the formatting rules of a specific black version for a little while until we've concluded large PRs to avoid merge conflicts.
The Github action currently does not support picking a specific version of black though.
Suggested... | Sounds good, a PR would be welcome. | [
{
"body": "Sometimes we want to stick with the formatting rules of a specific black version for a little while until we've concluded large PRs to avoid merge conflicts.\r\nThe Github action currently does not support picking a specific version of black though.\r\n\r\nSuggested solution: Add a parameter for the ... | 519f807f879350704449e424c5c5b7e049e582b3 | {
"head_commit": "55a46b85e7aaccc8dd56c350f2875d3267831698",
"head_commit_message": "Add black_version to github action",
"patch_to_review": "diff --git a/CHANGES.md b/CHANGES.md\nindex 97a3be33c93..82903ef0a6b 100644\n--- a/CHANGES.md\n+++ b/CHANGES.md\n@@ -45,6 +45,8 @@\n - PR #2053: Python 2 support is now opt... | [
{
"diff_hunk": "@@ -6,9 +6,20 @@ inputs:\n description: \"Black input arguments.\"\n required: false\n default: \"\"\n+ black_version:\n+ description: 'Version of black to use. E.g \"==20.8b1\"'",
"line": null,
"original_line": 10,
"original_start_line": null,
"path": "action.yml... | 2e40c836bb3142718a3b42967fbce2c31fa5d9e6 | diff --git a/CHANGES.md b/CHANGES.md
index d9800a6979c..8569b5d487f 100644
--- a/CHANGES.md
+++ b/CHANGES.md
@@ -16,7 +16,12 @@
- Add a lower bound for the `aiohttp-cors` dependency. Only 0.4.0 or higher is
supported. (#2231)
-### _Packaging_
+### Integrations
+
+- The official Black action now supports choosing ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
psf__black-948@1084e90 | psf/black | Python | 948 | Fix unstable format involving backslash + whitespace at beginning of file | This PR fixes #922 - an unstable sort involve Python line continuation with `\`. The issue describes the problem pretty well.
To my knowledge, there should be no situation where Black outputs a file formatted with empty lines at the beginning. By performing this check in the format function, we fix the above issue... | 2019-07-28T21:59:41Z | Black produced different code on the second pass of the formatter: Explicit line join in the first line
Operating system: macOS Mojave 10.14.5 (18F132)
Python version: Python 3.7.3
*Black* version: 19.3b0 (f3bb22a828)
Does also happen on master: yes
Minified reproducer:
```python
\
pass
```
Black diff:... | [
{
"body": "Operating system: macOS Mojave 10.14.5 (18F132)\r\nPython version: Python 3.7.3\r\n*Black* version: 19.3b0 (f3bb22a828)\r\nDoes also happen on master: yes\r\n\r\nMinified reproducer:\r\n```python\r\n\\\r\n\r\n\r\npass\r\n```\r\n\r\nBlack diff:\r\n```diff\r\n--- source\r\n+++ first pass\r\n@@ -1,5 +1,... | e66451761fab864cce9077a3ee72999543b8ef65 | {
"head_commit": "1084e90ccf9cc63d62e09c14d73a3bce1fe6b69e",
"head_commit_message": "Add newline to end of test data file",
"patch_to_review": "diff --git a/black.py b/black.py\nindex 910a0ed1e37..c3e449aa96d 100644\n--- a/black.py\n+++ b/black.py\n@@ -736,8 +736,11 @@ def format_str(src_contents: str, *, mode: F... | [
{
"diff_hunk": "@@ -736,8 +736,11 @@ def format_str(src_contents: str, *, mode: FileMode) -> FileContent:\n for _ in range(after):\n dst_contents.append(str(empty_line))\n before, after = elt.maybe_empty_lines(current_line)\n- for _ in range(before):\n- dst_contents... | fbc4e60f8324d907afbc88086cc3491f1acbd574 | diff --git a/black.py b/black.py
index 910a0ed1e37..1a6e6ca3671 100644
--- a/black.py
+++ b/black.py
@@ -1480,7 +1480,13 @@ def maybe_empty_lines(self, current_line: Line) -> Tuple[int, int]:
lines (two on module-level).
"""
before, after = self._maybe_empty_lines(current_line)
- befor... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Code Style, Linting, Formatting Fixes"
} | |
spotify__luigi-2813@daeb295 | spotify/luigi | Python | 2,813 | Add pod_creation_wait_interal | <!--- This template is optional. Please use it as a starting point to help guide PRs -->
<!--- Provide a general summary of your changes in the Title above -->
## Description
<!--- Describe your changes -->
Add pod_creation_wait_interal property to give Kubernetes cluster some time to create pods for Job.
##... | 2019-10-28T21:07:42Z | KubernetesJobTask "No pod scheduled" error
I often see the following error when using `KubernetesJobTask`:
```
[28:luigi-interface:224@18:33] [ERROR] [pid 28] Worker Worker(salt=476402868, workers=20, host=something-host-1541289600-zntd7, username=root, pid=1) failed Something(local_execution=Falser)
Traceback (m... | I have same issues.
@keisuke-umezawa I've managed to fix it with the code above. Just create a custom `KubernetesJobTask` class with an overridden `__verify_job_has_started` method.
@StasDeep Thanks! Do you have example how to override `__verify_job_has_started`?
Yep. Something like that:
```
class MyKubernetesJobT... | [
{
"body": "I often see the following error when using `KubernetesJobTask`:\r\n```\r\n[28:luigi-interface:224@18:33] [ERROR] [pid 28] Worker Worker(salt=476402868, workers=20, host=something-host-1541289600-zntd7, username=root, pid=1) failed Something(local_execution=Falser)\r\nTraceback (most recent call la... | db8ff63f8cd4418be91fa6bfb75b23102a5fae2f | {
"head_commit": "daeb295ed115af02af9a256d53defa4e12c0726a",
"head_commit_message": "Add pod_creation_wait_interal",
"patch_to_review": "diff --git a/luigi/contrib/kubernetes.py b/luigi/contrib/kubernetes.py\nindex c1c7afce65..13db2dbe94 100644\n--- a/luigi/contrib/kubernetes.py\n+++ b/luigi/contrib/kubernetes.py... | [
{
"diff_hunk": "@@ -213,10 +213,15 @@ def poll_interval(self):\n \"\"\"How often to poll Kubernetes for job status, in seconds.\"\"\"\n return self.__DEFAULT_POLL_INTERVAL\n \n+ @property\n+ def pod_creation_wait_interal(self):\n+ \"\"\"Delay for initial pod creation for just submit... | 87d64d22dd2b992ee2d16b7864dba6edbf3a47de | diff --git a/luigi/contrib/kubernetes.py b/luigi/contrib/kubernetes.py
index c1c7afce65..c913a33cdd 100644
--- a/luigi/contrib/kubernetes.py
+++ b/luigi/contrib/kubernetes.py
@@ -66,6 +66,7 @@ class kubernetes(luigi.Config):
class KubernetesJobTask(luigi.Task):
__DEFAULT_POLL_INTERVAL = 5 # see __track_job
+ ... | {
"difficulty": "low",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
spotify__luigi-2779@ea97908 | spotify/luigi | Python | 2,779 | PySparkTask: handle special characters in name (#2778) | Fixes #2778
## Description
It cleans _name_ property before it is used to derive _run_path_ and _run_pickle_ properties.
## Motivation and Context
If name property contains some special characters PySparkTask fails: #2778
## Have you tested this? If so, how?
Unit test is included
| 2019-09-09T22:19:46Z | PySparkTask fails if name contains invalid characters
PySparkTask derives run_path and run_pickle from name property, but it do not escape special characters:
1. '/' in name causes FileNotFoundError raised by mkdtemp
2. ',' is separator of files in spark-submit '--files' arg, and run_pickle is passed to 'files' arg i... | [
{
"body": "PySparkTask derives run_path and run_pickle from name property, but it do not escape special characters:\r\n1. '/' in name causes FileNotFoundError raised by mkdtemp\r\n2. ',' is separator of files in spark-submit '--files' arg, and run_pickle is passed to 'files' arg in cluster mode.\r\n3. ':' cause... | 1e99e6b68a51653553fc5cb26a4009fa58444331 | {
"head_commit": "ea97908903497c29ff0f076fffbc9251827e0370",
"head_commit_message": "PySparkTask: special characters in name (#2778)",
"patch_to_review": "diff --git a/luigi/contrib/spark.py b/luigi/contrib/spark.py\nindex d84d08312c..bb9b0d400c 100644\n--- a/luigi/contrib/spark.py\n+++ b/luigi/contrib/spark.py\n... | [
{
"diff_hunk": "@@ -298,8 +299,9 @@ def app_command(self):\n return [self.app, pickle_loc] + self.app_options()\n \n def run(self):\n- self.run_path = tempfile.mkdtemp(prefix=self.name)\n- self.run_pickle = os.path.join(self.run_path, '.'.join([self.name.replace(' ', '_'), 'pickle']))\... | 962c12c3497b34f1f6b4a8c780e477926fc26853 | diff --git a/luigi/contrib/spark.py b/luigi/contrib/spark.py
index d84d08312c..3282b61bfa 100644
--- a/luigi/contrib/spark.py
+++ b/luigi/contrib/spark.py
@@ -17,6 +17,7 @@
import logging
import os
+import re
import sys
import tempfile
import shutil
@@ -298,8 +299,9 @@ def app_command(self):
return [sel... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} | |
psf__black-679@7360c01 | psf/black | Python | 679 | Wrap `loop.run_in_executor` up in `asyncio.ensure_future` for reliable cross-platform berhavior. | Closes #494
Task completion should also remove the task from `pending`.
Only replicates on some platforms. (eg. Can replicate on Python 3.7+, with either Windows or whatever default Linux distro Travis uses.) | 2019-01-23T10:26:09Z | Black is not formatting code on Python 3.7 on some platform configurations (Windows, Ubuntu)
Howdy! Sorry you're having trouble. To expedite your experience,
provide some basics for me:
Operating system: Windows 10 64bit
Python version: Python 3.7.0 (32Bit)
*Black* version: 18.6b4
Does also happen on master: Yes... | Interesting, some related tests indeed fail on a 32bit 3.7.0, but not 3.6.5. I'll look into this.
Sounds like an issue with asyncio.
Just confirming this issue here too. The traceback I see for it is...
```python
Traceback (most recent call last):
File "/home/travis/virtualenv/python3.7-dev/bin/black", line 11, ... | [
{
"body": "Howdy! Sorry you're having trouble. To expedite your experience,\r\nprovide some basics for me:\r\n\r\nOperating system: Windows 10 64bit\r\nPython version: Python 3.7.0 (32Bit)\r\n*Black* version: 18.6b4\r\nDoes also happen on master: Yes\r\n\r\nWhen running black for directory it fails without any ... | ea55ff28782f7e3b481c99faaf9f57e88597bdde | {
"head_commit": "7360c013875c66c897bf69062f431908d2c54a34",
"head_commit_message": "Ensure that 'tasks.keys()' are always Task instances",
"patch_to_review": "diff --git a/black.py b/black.py\nindex 32693227ed2..b0a9a32a40f 100644\n--- a/black.py\n+++ b/black.py\n@@ -464,15 +464,17 @@ async def schedule_formatti... | [
{
"diff_hunk": "@@ -464,15 +464,17 @@ def reformat_one(\n manager = Manager()\n lock = manager.Lock()\n tasks = {\n- loop.run_in_executor(\n- executor,\n- format_file_in_place,\n- src,\n- line_length,\n- fast,\n- write_... | 6406f35c9cf60abd5fa14f833ada0176b9d7d61d | diff --git a/black.py b/black.py
index 34bd59eacdd..84f1942a0be 100644
--- a/black.py
+++ b/black.py
@@ -519,12 +519,14 @@ async def schedule_formatting(
manager = Manager()
lock = manager.Lock()
tasks = {
- loop.run_in_executor(
- executor, format_file_in_place, src, fast, mode... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
spotify__luigi-2487@868a122 | spotify/luigi | Python | 2,487 | Added optional choice for hdfs clients | ## Description
Added optional choice for hdfs clients. Previously only InsecureClient was supported. This PR adds KerberosClient and TokenClient.
## Motivation and Context
Some hdfs work with kerberos authentication. Mine does... This adds support for those clusters.
This fixes #2481.
## Have you tested th... | 2018-08-10T18:52:03Z | Use of KerberosClient with webhdfs
Currently webhdfs uses InsecureClient by default. However, the hdfs package also has an KerberosClient which can be used to do authenticated requests.
Our hdfs system uses Kerberos and I would like it if KerberosClient is added as an option.
The KerberosClient does not need an... | I, personally, do not use the HDFS/Hadoop side of Luigi. But so long as your suggested PR doesn't change the default behavior, please feel free to submit for review!
> Are you willing to accept a pull request for this?
Yep, for sure. | [
{
"body": "Currently webhdfs uses InsecureClient by default. However, the hdfs package also has an KerberosClient which can be used to do authenticated requests. \r\n\r\nOur hdfs system uses Kerberos and I would like it if KerberosClient is added as an option. \r\n\r\nThe KerberosClient does not need any additi... | de4629a4c7c562b53fee2045aabd84189f897593 | {
"head_commit": "868a122c2b6f1c275beda8105b7b4e060f4359a8",
"head_commit_message": "Added documentation for the new client_type and token parameter of the webhdfs config documentation section",
"patch_to_review": "diff --git a/doc/configuration.rst b/doc/configuration.rst\nindex dec55c3679..7305d1756d 100644\n--... | [
{
"diff_hunk": "@@ -43,6 +43,9 @@ class webhdfs(luigi.Config):\n description='Port for webhdfs')\n user = luigi.Parameter(default='', description='Defaults to $USER envvar',\n config_path=dict(section='hdfs', name='user'))\n+ client_type = luigi.Pa... | 19e5d69145d6f51799e536418257ac597e6f9655 | diff --git a/doc/configuration.rst b/doc/configuration.rst
index dec55c3679..d152855ca9 100644
--- a/doc/configuration.rst
+++ b/doc/configuration.rst
@@ -922,11 +922,17 @@ summary-length
port
The port to use for webhdfs. The normal namenode port is probably on a
different port from this one.
+
user
Perform ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-31528@a2e89ea | scikit-learn/scikit-learn | Python | 31,528 | FIX fix comparison between array-like parameters when detecting non-default params for HTML representation | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2025-06-12T06:55:34Z | Issue with the `RidgeCV` diagram representation with non-default alphas
It seems that we introduced a regression in the HTML representation. The following code is failing:
```python
import numpy as np
from sklearn.linear_model import RidgeCV
RidgeCV(np.logspace(-3, 3, num=10)
```
leads to the following error:
```py... | My guess here is that `param_value != init_default_params[param_name]` compare 2 numpy arrays:
```python
np.array([1, 2, 3]) != np.array([1, 2, 3, 4])
```
```pytb
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
Cel... | [
{
"body": "It seems that we introduced a regression in the HTML representation. The following code is failing:\n\n```python\nimport numpy as np\nfrom sklearn.linear_model import RidgeCV\n\nRidgeCV(np.logspace(-3, 3, num=10)\n```\n\nleads to the following error:\n\n```pytb\n--------------------------------------... | dab084231912cb8bbb70bae52872b06bf6d2ee0f | {
"head_commit": "a2e89eaf5311723373fabfbe17bb12de549495c1",
"head_commit_message": "added array vs scalar",
"patch_to_review": "diff --git a/doc/whats_new/upcoming_changes/sklearn.base/31528.fix.rst b/doc/whats_new/upcoming_changes/sklearn.base/31528.fix.rst\nnew file mode 100644\nindex 0000000000000..25bf0404c6... | [
{
"diff_hunk": "@@ -1000,3 +1001,46 @@ def test_get_params_html():\n \n assert est._get_params_html() == {\"l1\": 0, \"empty\": \"test\"}\n assert est._get_params_html().non_default == (\"empty\",)\n+\n+\n+# non regression test see github issue #31525\n+def test_get_params_html_ridgecv():\n+ est = Ri... | c774f2c6a2e96055c05beccf0e0d0244b87f56bd | diff --git a/doc/whats_new/upcoming_changes/sklearn.base/31528.fix.rst b/doc/whats_new/upcoming_changes/sklearn.base/31528.fix.rst
new file mode 100644
index 0000000000000..312c8318eadcd
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.base/31528.fix.rst
@@ -0,0 +1,3 @@
+- Fix regression in HTML representatio... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-5022@3dc1033 | reflex-dev/reflex | Python | 5,022 | expose socket constants | fixes #3081 | 2025-03-25T01:34:11Z | Entering more than 999775 characters breaks text_area
**Describe the bug**
Entering more than 999775 characters breaks rx.text_area. If you enter more than 999775 characters into a text area you will have to refresh the page for the text_area to work again.
**To Reproduce**
Steps to reproduce the behavior:
```
... | Hi!
This error happens because the size of a message sent from browser to backend by default is limited to ~1Mb, you can see this limitation [here](https://github.com/reflex-dev/reflex/blob/0a8aaea599f95bb6c4730c743f6c4b636134ca6a/reflex/app.py#L234) in code. In `reflex.App` you can pass your custom socket engine ... | [
{
"body": "**Describe the bug**\r\nEntering more than 999775 characters breaks rx.text_area. If you enter more than 999775 characters into a text area you will have to refresh the page for the text_area to work again.\r\n\r\n**To Reproduce**\r\nSteps to reproduce the behavior:\r\n\r\n```\r\nimport reflex as rx\... | c6e2707b343b07f6a33ab63d22e9ea91de8aefdf | {
"head_commit": "3dc1033e98b7dac9e813d2ebc7dc165ddee6adf6",
"head_commit_message": "expose socket constants",
"patch_to_review": "diff --git a/reflex/app.py b/reflex/app.py\nindex 2c823ec1e4b..3c05b1b1496 100644\n--- a/reflex/app.py\n+++ b/reflex/app.py\n@@ -498,9 +498,9 @@ def _setup_state(self) -> None:\n ... | [
{
"diff_hunk": "@@ -498,9 +498,9 @@ def _setup_state(self) -> None:\n else config.cors_allowed_origins\n ),\n cors_credentials=True,\n- max_http_buffer_size=constants.POLLING_MAX_HTTP_BUFFER_SIZE,\n- ping_interval=constants.Ping.I... | 5443e85dbe1d7c4f14523ea68cd868bd08b4f5b8 | diff --git a/reflex/app.py b/reflex/app.py
index 2c823ec1e4b..a1e98500df0 100644
--- a/reflex/app.py
+++ b/reflex/app.py
@@ -498,9 +498,9 @@ def _setup_state(self) -> None:
else config.cors_allowed_origins
),
cors_credentials=True,
- max_http_buffer_... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-4862@d3c22c9 | reflex-dev/reflex | Python | 4,862 | opt for powershell as opposed to wmic | Fixes #4198
| 2025-02-22T17:55:23Z | wmic cpu get addresswidth,caption,manufacturer /FORMAT:csv
**Describe the bug**
The command `wmic cpu get addresswidth,caption,manufacturer /FORMAT:csv` failed with error: Command 'wmic cpu get
addresswidth,caption,manufacturer /FORMAT:csv' returned non-zero exit status 1.. This will return None.
AND infinity relo... | <p><a href="https://linear.app/reflex-dev/issue/ENG-3962/wmic-cpu-get-addresswidthcaptionmanufacturer-formatcsv">ENG-3962 wmic cpu get addresswidth,caption,manufacturer /FORMAT:csv</a></p>
is this a regression from 0.6.2?
> [ENG-3962 wmic cpu get addresswidth,caption,manufacturer /FORMAT:csv](https://linear.app/reflex-... | [
{
"body": "**Describe the bug**\r\nThe command `wmic cpu get addresswidth,caption,manufacturer /FORMAT:csv` failed with error: Command 'wmic cpu get\r\naddresswidth,caption,manufacturer /FORMAT:csv' returned non-zero exit status 1.. This will return None.\r\n\r\nAND infinity reloading local server!\r\n\r\n\r\n*... | 98f50811f97639bbf2015bacf0f4b0b3b843fbc8 | {
"head_commit": "d3c22c95570fd025421fc455426a242ee5fc39f6",
"head_commit_message": "mkdir of parents",
"patch_to_review": "diff --git a/reflex/utils/prerequisites.py b/reflex/utils/prerequisites.py\nindex 8afa706be5b..d3d8bd6ab49 100644\n--- a/reflex/utils/prerequisites.py\n+++ b/reflex/utils/prerequisites.py\n@... | [
{
"diff_hunk": "@@ -1925,12 +1925,11 @@ def format_address_width(address_width: str | None) -> int | None:\n return None\n \n \n-@functools.lru_cache(maxsize=None)\n-def get_cpu_info() -> CpuInfo | None:\n- \"\"\"Get the CPU info of the underlining host.\n+def _load_cpu_info() -> CpuInfo | None:\n+ ... | 7137ef522374853f16c704bac2c047bde998f39b | diff --git a/reflex/utils/exec.py b/reflex/utils/exec.py
index 5474ae82a0c..e3fe315b2cc 100644
--- a/reflex/utils/exec.py
+++ b/reflex/utils/exec.py
@@ -524,17 +524,12 @@ def output_system_info():
fnm_info = f"[FNM {prerequisites.get_fnm_version()} (Expected: {constants.Fnm.VERSION}) (PATH: {constants.Fnm.EXE})]... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-30819@f872efb | scikit-learn/scikit-learn | Python | 30,819 | FIX Add input array check to `randomized_range_finder` | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2025-02-12T20:23:02Z | `randomized_svd` incorrect for complex valued matrices
### Describe the bug
The `randomized_svd` utility function accepts complex valued inputs without error, but the result is inconsistent with `scipy.linalg.svd`.
### Steps/Code to Reproduce
```python
import numpy as np
from scipy import linalg
from sklearn.utils.e... | [
{
"body": "### Describe the bug\n\nThe `randomized_svd` utility function accepts complex valued inputs without error, but the result is inconsistent with `scipy.linalg.svd`.\n\n### Steps/Code to Reproduce\n\n```python\nimport numpy as np\nfrom scipy import linalg\nfrom sklearn.utils.extmath import randomized_sv... | e47b7c0f49450cbbcdb5abcc198cf07719567f34 | {
"head_commit": "f872efbde4c25b437dc1bdfe113bc622f67eaf2f",
"head_commit_message": "iter",
"patch_to_review": "diff --git a/doc/whats_new/upcoming_changes/array-api/30819.feature.rst b/doc/whats_new/upcoming_changes/array-api/30819.feature.rst\nnew file mode 100644\nindex 0000000000000..40c4fc306a90c\n--- /dev/n... | [
{
"diff_hunk": "@@ -274,6 +274,7 @@ def randomized_range_finder(\n [-0.82..., -0.38...]])\n \"\"\"\n xp, is_array_api_compliant = get_namespace(A)\n+ A = check_array(A, accept_sparse=True)",
"line": 276,
"original_line": 277,
"original_start_line": null,
"path": "sklearn/ut... | d0324464fc3ace1c72d76d9d93e1d041aa21db75 | diff --git a/doc/whats_new/upcoming_changes/array-api/30819.feature.rst b/doc/whats_new/upcoming_changes/array-api/30819.feature.rst
new file mode 100644
index 0000000000000..fac6d32b00375
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/array-api/30819.feature.rst
@@ -0,0 +1,2 @@
+- :func:`sklearn.utils.extmath.rand... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} | |
reflex-dev__reflex-4406@5ec2c11 | reflex-dev/reflex | Python | 4,406 | add datetime var comparison operations | Closes #4304 | 2024-11-20T22:15:24Z | Unable to compare `date` objects
**Describe the bug**
I'm getting an error when trying to compare two `date` objects and one of them is a `State var` (see code below)
**To Reproduce**
Here's a code example
```
rx.flex(
rx.cond(
datetime.now(pytz.utc).date() <= competition.end_date,
... | <p><a href="https://linear.app/reflex-dev/issue/ENG-4048/unable-to-compare-date-objects">ENG-4048 Unable to compare `date` objects</a></p> | [
{
"body": "**Describe the bug**\r\nI'm getting an error when trying to compare two `date` objects and one of them is a `State var` (see code below)\r\n\r\n**To Reproduce**\r\nHere's a code example\r\n ```\r\nrx.flex(\r\n rx.cond(\r\n datetime.now(pytz.utc).date() <= competition.end_date,\r\n ... | 681b61600025c67e1438c2252917ff078ff0db2c | {
"head_commit": "5ec2c119365d5fe5353cc7ca30fd96527e1e995a",
"head_commit_message": "add future annotations",
"patch_to_review": "diff --git a/reflex/vars/__init__.py b/reflex/vars/__init__.py\nindex 1a4cebe19a3..cb02319bc6a 100644\n--- a/reflex/vars/__init__.py\n+++ b/reflex/vars/__init__.py\n@@ -9,6 +9,7 @@\n f... | [
{
"diff_hunk": "@@ -0,0 +1,184 @@\n+\"\"\"Immutable datetime and date vars.\"\"\"\n+\n+from __future__ import annotations\n+\n+from datetime import date, datetime\n+from typing import Any, NoReturn, TypeVar, Union, overload\n+\n+from reflex.utils.exceptions import VarTypeError\n+from reflex.vars.number import B... | c757788c0b2a5bce3c6ef9333784e74dbc4c7f71 | diff --git a/reflex/vars/__init__.py b/reflex/vars/__init__.py
index 1a4cebe19a3..cb02319bc6a 100644
--- a/reflex/vars/__init__.py
+++ b/reflex/vars/__init__.py
@@ -9,6 +9,7 @@
from .base import get_uuid_string_var as get_uuid_string_var
from .base import var_operation as var_operation
from .base import var_operatio... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-4347@a91f89a | reflex-dev/reflex | Python | 4,347 | fix: do not allow instantiation of State mixins | Closes #4343 | 2024-11-10T19:55:13Z | Testing `rx.ComponentState` by direct initialization broken in `v0.6.2+` (SetUndefinedStateVarError)
**Describe the bug**
Before v0.6.2, it was possible to directly initialize `ComponentState` for testing purposes (same as directly initializing `State` instances), but since `v0.6.2` I get an error:
```
reflex.utils.... | <p><a href="https://linear.app/reflex-dev/issue/ENG-4086/testing-rxcomponentstate-by-direct-initialization-broken-in-v062">ENG-4086 Testing `rx.ComponentState` by direct initialization broken in `v0.6.2+` (SetUndefinedStateVarError)</a></p>
@TimChild i suggest adjusting the test to smth like this:
```python
from ty... | [
{
"body": "**Describe the bug**\r\nBefore v0.6.2, it was possible to directly initialize `ComponentState` for testing purposes (same as directly initializing `State` instances), but since `v0.6.2` I get an error:\r\n```\r\nreflex.utils.exceptions.SetUndefinedStateVarError: The state variable 'some_var' has not ... | 34c11fdf108395fcacbe2488655969fef5c7958c | {
"head_commit": "a91f89aade92d8a67956ea7aa49b019b327f7f51",
"head_commit_message": "improve error message for ComponentState mixins",
"patch_to_review": "diff --git a/reflex/state.py b/reflex/state.py\nindex 66b1e3cabfa..1cbac53ce2c 100644\n--- a/reflex/state.py\n+++ b/reflex/state.py\n@@ -86,6 +86,7 @@\n Im... | [
{
"diff_hunk": "@@ -2361,6 +2366,23 @@ def get_component(cls, **props):\n # The number of components created from this class.\n _per_component_state_instance_count: ClassVar[int] = 0\n \n+ def __init__(self, *args, **kwargs):\n+ \"\"\"Do not allow direct initialization of the ComponentState.\n... | 99dd4c6e6ba8b648a64edfee6aac8e420d4c3841 | diff --git a/reflex/state.py b/reflex/state.py
index 719ff43b3d1..9ff6f0ea8fb 100644
--- a/reflex/state.py
+++ b/reflex/state.py
@@ -87,6 +87,7 @@
ImmutableStateError,
InvalidStateManagerMode,
LockExpiredError,
+ ReflexRuntimeError,
SetUndefinedStateVarError,
StateSchemaMismatchError,
)
@@ ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-3548@a6cc8ec | reflex-dev/reflex | Python | 3,548 | [REF-2588]Clear color mode local storage for dev mode | Sometimes an app's `appearance` value(`rx.App(theme=rx.theme(appearance="dark"))`) does not persist correctly in the local storage(as `chakra-ui-color-mode`). Eg, if you manipulate the local storage value and restart the app, the app's theme/color-mode is not respected. This is typically annoying in dev mode.
We want... | 2024-06-24T16:59:15Z | Dev Mode Clear Dark Cookie
During dev mode sometimes when you are on another app and running dark mode stop that app and go to a new one the cookie for local host is cached, and your other app has dark mode. On every run dev, we should clear this cookie | [
{
"body": "During dev mode sometimes when you are on another app and running dark mode stop that app and go to a new one the cookie for local host is cached, and your other app has dark mode. On every run dev, we should clear this cookie",
"number": 2047,
"title": "Dev Mode Clear Dark Cookie"
}
] | 44233c9cd111f282f574b890d51ccc226a47e019 | {
"head_commit": "a6cc8ec9526e2143def89535659ebc7d511224ea",
"head_commit_message": "fix CI",
"patch_to_review": "diff --git a/integration/test_client_storage.py b/integration/test_client_storage.py\nindex 393ea104410..03dbd168147 100644\n--- a/integration/test_client_storage.py\n+++ b/integration/test_client_sto... | [
{
"diff_hunk": "@@ -1,13 +1,31 @@\n import { useTheme } from \"next-themes\";\n import { useEffect, useState } from \"react\";\n-import { ColorModeContext, defaultColorMode } from \"/utils/context.js\";\n+import {\n+ ColorModeContext,\n+ defaultColorMode,\n+ isDevMode,\n+} from \"/utils/context.js\";\n+impor... | 8b041a5a4d48e589a3fe8157e5bf3b067cb6a8f3 | diff --git a/integration/test_client_storage.py b/integration/test_client_storage.py
index 393ea104410..03dbd168147 100644
--- a/integration/test_client_storage.py
+++ b/integration/test_client_storage.py
@@ -311,6 +311,7 @@ def set_sub_sub(var: str, value: str):
assert not driver.get_cookies()
local_storage_... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} | |
scikit-learn__scikit-learn-31065@03f618d | scikit-learn/scikit-learn | Python | 31,065 | FIX Fix adjusted_mutual_info_score numerical issue | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2025-03-24T21:27:12Z | Potential Problem in the Computation of Adjusted Mutual Info Score
### Describe the bug
It seems to me that for clusters of size 2 and 4, the AMI yields unexpected results of 0 instead of 1, if all items belong to different clusters.
### Steps/Code to Reproduce
Sample Code to Reproduce:
```python
>>> from sklear... | Weird, it seems like a real bug, although probably not that important in practice? I guess someone needs to investigate in more details?
Yes, it occurs only if you stubbornly try to determine how these methods react on different clusters. That's what I did. But it leaves some general suspicion to the method, right? Spe... | [
{
"body": "### Describe the bug\n\nIt seems to me that for clusters of size 2 and 4, the AMI yields unexpected results of 0 instead of 1, if all items belong to different clusters. \n\n\n\n### Steps/Code to Reproduce\n\nSample Code to Reproduce:\n\n```python\n>>> from sklearn.metrics.cluster import adjusted_mut... | 812ff67e6725a8ca207a37f5ed4bfeafc5d1265d | {
"head_commit": "03f618d940173d6653a6ee9b61ce4833b209bbbd",
"head_commit_message": "Merge branch 'main' into adj-mi-bugfix",
"patch_to_review": "diff --git a/doc/whats_new/upcoming_changes/sklearn.metrics/31065.fix.rst b/doc/whats_new/upcoming_changes/sklearn.metrics/31065.fix.rst\nnew file mode 100644\nindex 00... | [
{
"diff_hunk": "@@ -1033,12 +1033,16 @@ def adjusted_mutual_info_score(\n or classes.shape[0] == clusters.shape[0] == 0\n ):\n return 1.0\n+ # if there is only one class or one cluster return 0.0.\n+ elif classes.shape[0] == 1 or clusters.shape[0] == 1:\n+ return 0.0\n \n co... | 0b408299147c1506f0dc89e85f0f723b8cadb27b | diff --git a/doc/whats_new/upcoming_changes/sklearn.metrics/31065.fix.rst b/doc/whats_new/upcoming_changes/sklearn.metrics/31065.fix.rst
new file mode 100644
index 0000000000000..82126da7852cc
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.metrics/31065.fix.rst
@@ -0,0 +1,3 @@
+- Fix :func:`metrics.adjusted... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-30516@c27046d | scikit-learn/scikit-learn | Python | 30,516 | FIX warn if an estimator does have a concrete __sklearn_tags__ implementation | closes #30479
Catch the error in the case that there is no `__sklearn_tags__` implementation in the MRO. We return the default tags and request to inherit from `BaseEstimator` that implements it. | 2024-12-20T18:25:08Z | Version 1.6.X: ClassifierMixIn failing with new __sklearn_tags__ function
### Describe the bug
Hi,
we are using Sklearn in our projects for different classification training methods on production level. In the dev stage we upgraded to the latest release and our Training failed due to changes in the ClassifierMixI... | @DaMuBo According to [the official documentation](https://scikit-learn.org/1.5/developers/develop.html#rolling-your-own-estimator), I think your `MyEstimator` should inherit from `BaseEstimator` as well.
```python
import numpy as np
from sklearn.base import BaseEstimator, ClassifierMixin
class MyEstimator(Class... | [
{
"body": "### Describe the bug\r\n\r\nHi,\r\n\r\nwe are using Sklearn in our projects for different classification training methods on production level. In the dev stage we upgraded to the latest release and our Training failed due to changes in the ClassifierMixIn Class. We use it in combination with a sklear... | 72b35a46684c0ecf4182500d3320836607d1f17c | {
"head_commit": "c27046db4f4da2c0edb2d466d993393a224481e4",
"head_commit_message": "Update doc/whats_new/upcoming_changes/sklearn.utils/30516.fix.rst\n\nCo-authored-by: Adrin Jalali <adrin.jalali@gmail.com>",
"patch_to_review": "diff --git a/doc/whats_new/upcoming_changes/sklearn.utils/30516.fix.rst b/doc/whats_... | [
{
"diff_hunk": "@@ -0,0 +1,4 @@\n+- Raise a `FutureWarning` when there is no concrete implementation of `__sklearn_tags__`",
"line": null,
"original_line": 1,
"original_start_line": null,
"path": "doc/whats_new/upcoming_changes/sklearn.utils/30516.fix.rst",
"start_line": null,
"text": "@... | 1e0a0a6dc30526df1e3b9eda4357632cdcfd916b | diff --git a/doc/whats_new/upcoming_changes/sklearn.utils/30516.fix.rst b/doc/whats_new/upcoming_changes/sklearn.utils/30516.fix.rst
new file mode 100644
index 0000000000000..6e008f3beeb3c
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.utils/30516.fix.rst
@@ -0,0 +1,4 @@
+- Raise a `DeprecationWarning` when... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-3895@603bf40 | reflex-dev/reflex | Python | 3,895 | Fix type propagation in ToStringOperation | ### Type of change
- [X] Bug fix (non-breaking change which fixes an issue)
- [X] Does your submission pass the tests?
closes #3890
| 2024-09-05T20:24:01Z | `ImmutableVar.guess_type()` hides the original var type on fields derived from integral types
**Describe the bug**
In the example below you can see my problem, currently the bottom assertion fails.
**To Reproduce**
```python
class Email(str):
... # validation and stuff ...
class State(rx.State):
em... | Or maybe
```python
var_type = (
original_var._var_type
if original_var is not None
and isclass(original_var._var_type)
and issubclass(original_var._var_type, str)
else str
)
Failing test:
```python
def test_to_string_operation():... | [
{
"body": "**Describe the bug**\r\nIn the example below you can see my problem, currently the bottom assertion fails.\r\n\r\n**To Reproduce**\r\n```python\r\nclass Email(str):\r\n ... # validation and stuff ...\r\n \r\nclass State(rx.State):\r\n email: Email = Email(\"someone@reflex.dev\")\r\n\r\nassert ... | 0810bd843cf8baf5e84a03ef588dbb339301c787 | {
"head_commit": "603bf406a137cfb5a531fc7a8de6fb32a92cfb23",
"head_commit_message": "fix: Added test that fails without the fix.",
"patch_to_review": "diff --git a/reflex/ivars/base.py b/reflex/ivars/base.py\nindex 4cd3550ddf7..1b65449d269 100644\n--- a/reflex/ivars/base.py\n+++ b/reflex/ivars/base.py\n@@ -1043,6... | [
{
"diff_hunk": "@@ -1043,6 +1043,27 @@ def unionize(*args: Type) -> Type:\n return Union[first, unionize(*rest)]\n \n \n+def original_var_type_or_target_type(original_var: Var, target_type: type) -> type:\n+ \"\"\"Inherit the original var type if possible, i.e. the original var is a subclass of the targe... | e8602bd6cfae828ffdf21d59e19553952c744fa8 | diff --git a/reflex/ivars/base.py b/reflex/ivars/base.py
index 27c4828afb8..8bb7957c494 100644
--- a/reflex/ivars/base.py
+++ b/reflex/ivars/base.py
@@ -416,7 +416,7 @@ def to(
return ToArrayOperation.create(self, var_type or list)
if issubclass(output, StringVar):
- return ToStringOp... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-2837@c5b8a90 | reflex-dev/reflex | Python | 2,837 | ruff-format: unify Black with Ruff `v0.1` | ### All Submissions:
- [x] Have you followed the guidelines stated in [CONTRIBUTING.md](https://github.com/reflex-dev/reflex/blob/main/CONTRIBUTING.md) file?
- [x] Have you checked to ensure there aren't any other open [Pull Requests](https://github.com/reflex-dev/reflex/pulls ) for the desired changed?
<!-- You... | 2024-03-12T00:07:02Z | Fully Migrate off Black to Ruff
Reflex is already using Ruff for linting and formatting. Yet their is still a legacy dependency on the black formatter. At this point, Ruff seems to be on course to completely take over as industry standard and the sooner Reflex standardizes the better. I see that the file `scripts/pyi_g... | Hi @dapomeranz thanks for pointing this out. We do have plans of fully migrating to Ruff at some point. The team is currently focused on other high-priority issues at the moment. However, we're open to receiving a PR regarding this from the community.
Noting that there is a related PR #2837
Relevant Discord Conversat... | [
{
"body": "Reflex is already using Ruff for linting and formatting. Yet their is still a legacy dependency on the black formatter. At this point, Ruff seems to be on course to completely take over as industry standard and the sooner Reflex standardizes the better. I see that the file `scripts/pyi_generator.py` ... | 25b84107da014ab16c9b615f884fe5c8d66c8a57 | {
"head_commit": "c5b8a904fc2ba2ec301fb1d5264b04b7fefaf431",
"head_commit_message": "Fix or\n\nCo-authored-by: Thomas Brandého <thomas.brandeho@gmail.com>",
"patch_to_review": "diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml\nindex 560598e9e06..e36cbdfe631 100644\n--- a/.pre-commit-config.yaml\n+++... | [
{
"diff_hunk": "@@ -1082,7 +1082,7 @@ class CodeBlock(Component):\n on_unmount: Optional[\n Union[EventHandler, EventSpec, list, function, BaseVar]\n ] = None,\n- **props\n+ **props,",
"line": null,
"original_line": 1085,
"original_start_line": null,
"pa... | 9fe484ae5b2a5ee42c31443b5e55b9aad1bba101 | diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 560598e9e06..e36cbdfe631 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -1,17 +1,14 @@
fail_fast: true
repos:
- - repo: https://github.com/psf/black
- rev: 22.10.0
- hooks:
- - id: black
- args: [integrati... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Code Style, Linting, Formatting Fixes"
} |
reflex-dev__reflex-3170@cdafb49 | reflex-dev/reflex | Python | 3,170 | Fixed unused kwargs in reflex.App | ### Type of change
- [x] Bug fix (non-breaking change which fixes an issue)
Fixed kwargs `sio` and `socket_app` are not used during reflex.App initialization
Closes #3081
| 2024-04-26T08:34:12Z | Entering more than 999775 characters breaks text_area
**Describe the bug**
Entering more than 999775 characters breaks rx.text_area. If you enter more than 999775 characters into a text area you will have to refresh the page for the text_area to work again.
**To Reproduce**
Steps to reproduce the behavior:
```
... | [
{
"body": "**Describe the bug**\r\nEntering more than 999775 characters breaks rx.text_area. If you enter more than 999775 characters into a text area you will have to refresh the page for the text_area to work again.\r\n\r\n**To Reproduce**\r\nSteps to reproduce the behavior:\r\n\r\n```\r\nimport reflex as rx\... | 0a8aaea599f95bb6c4730c743f6c4b636134ca6a | {
"head_commit": "cdafb4942d6c56dfa6fd618b61b0ef032ca0003d",
"head_commit_message": "Fixed unused kwargs in reflex.App",
"patch_to_review": "diff --git a/reflex/app.py b/reflex/app.py\nindex c47077d0ef8..dcd7ef742a1 100644\n--- a/reflex/app.py\n+++ b/reflex/app.py\n@@ -223,21 +223,24 @@ def setup_state(self) -> N... | [
{
"diff_hunk": "@@ -223,21 +223,24 @@ def setup_state(self) -> None:\n self._state_manager = StateManager.create(state=self.state)\n \n # Set up the Socket.IO AsyncServer.\n- self.sio = AsyncServer(\n- async_mode=\"asgi\",\n- cors_allowed_origins=(\n- ... | fbb527056591284174dc56ca442d6b2aaed5a5d3 | diff --git a/reflex/app.py b/reflex/app.py
index c47077d0ef8..95c9de2c615 100644
--- a/reflex/app.py
+++ b/reflex/app.py
@@ -116,9 +116,6 @@ class App(Base):
# The Socket.IO AsyncServer.
sio: Optional[AsyncServer] = None
- # The socket app.
- socket_app: Optional[ASGIApp] = None
-
# The state cla... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} | |
scikit-learn__scikit-learn-30268@02b8599 | scikit-learn/scikit-learn | Python | 30,268 | MNT Tags: quality of life improvements | Fixes https://github.com/scikit-learn/scikit-learn/issues/30011
This adds tests, and makes it more clear on how to extend tags.
It also does the right check on whether `_more_tags` is implemented but `__sklearn_tags__` is missing.
cc @glemaitre @thomasjpfan | 2024-11-12T15:53:50Z | Extending tags infrastructure is not easy anymore
In the context of bringing `imbalanced-learn` compatible with changes done in `scikit-learn` 1.6, I could see that now this is quite difficult to extend the tags infrastructure.
For instance, we added a `"dataframe"` entry to the previous `X_type` to check whether or... | The idea is that you need to subclass and add an attribute that way.
But we haven't tested this
OK so I'll give a try.
So I'm playing with the tag in `imbalanced-learn`. Right now, I would like to extend `Tags` such that it contains a `SamplerTags`. There is two problems:
- I cannot subclass `Tags` and add `Sampler... | [
{
"body": "In the context of bringing `imbalanced-learn` compatible with changes done in `scikit-learn` 1.6, I could see that now this is quite difficult to extend the tags infrastructure.\r\n\r\nFor instance, we added a `\"dataframe\"` entry to the previous `X_type` to check whether or not we should run our ow... | b21452450c5e679568882eacd78c2ec48e5f354b | {
"head_commit": "02b8599c5bc2819b9016175e59f32b5c042a1808",
"head_commit_message": "add imports to doc",
"patch_to_review": "diff --git a/doc/developers/develop.rst b/doc/developers/develop.rst\nindex ace3fbbcfa9c6..1e4d82c8d0425 100644\n--- a/doc/developers/develop.rst\n+++ b/doc/developers/develop.rst\n@@ -519... | [
{
"diff_hunk": "@@ -1,11 +1,18 @@\n+from dataclasses import asdict, dataclass\n+\n import pytest\n \n from sklearn.base import (\n BaseEstimator,\n RegressorMixin,\n TransformerMixin,\n )\n+from sklearn.utils import Tags\n from sklearn.utils._tags import get_tags",
"line": null,
"original_li... | ee76a507b653651149ee1fe08ee64d2aa844f30c | diff --git a/doc/developers/develop.rst b/doc/developers/develop.rst
index ace3fbbcfa9c6..3b8a455c75228 100644
--- a/doc/developers/develop.rst
+++ b/doc/developers/develop.rst
@@ -519,7 +519,26 @@ which returns the new values for your estimator's tags. For example::
return tags
You can create a new sub... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Code Refactoring / Architectural Improvement"
} |
reflex-dev__reflex-2827@769323f | reflex-dev/reflex | Python | 2,827 | added name of Var to error message | - Added the name of the Var to the error message for ```TypeError(f"No JSON serializer found for var {value} of type {type(value)}.")``` closes #2826
### All Submissions:
- [x] Have you followed the guidelines stated in [CONTRIBUTING.md](https://github.com/reflex-dev/reflex/blob/main/CONTRIBUTING.md) fi... | 2024-03-10T17:33:11Z | include the name of the Var so it's easier to track down `No JSON serializer` error
It will be nice to also include the name of the Var causing the ```TypeError(f"No JSON serializer found for var {value} of type {type(value)}.")``` so it's easier for a user to track down and maybe convert the Var to a string | [
{
"body": "It will be nice to also include the name of the Var causing the ```TypeError(f\"No JSON serializer found for var {value} of type {type(value)}.\")``` so it's easier for a user to track down and maybe convert the Var to a string",
"number": 2826,
"title": "include the name of the Var so it's ... | 7b43b923b81cf2b59ac9f3f19973459b1dd79f22 | {
"head_commit": "769323f9aa0e4c0c95a4aef4066e1094e7121a38",
"head_commit_message": "added name of Var to error message",
"patch_to_review": "diff --git a/reflex/utils/format.py b/reflex/utils/format.py\nindex 53b55d0ebdf..dd79be132d0 100644\n--- a/reflex/utils/format.py\n+++ b/reflex/utils/format.py\n@@ -2,9 +2,... | [
{
"diff_hunk": "@@ -598,7 +600,14 @@ def format_state(value: Any) -> Any:\n \"\"\"\n # Handle dicts.\n if isinstance(value, dict):\n- return {k: format_state(v) for k, v in value.items()}\n+ result = {}\n+ for k, v in value.items():\n+ if isinstance(v, (pathlib.PosixP... | 289d68e696cdfe0cf42aba2ad634b8a2f5a1cf7f | diff --git a/reflex/utils/format.py b/reflex/utils/format.py
index 0e57ca9efcb..5f39349c85d 100644
--- a/reflex/utils/format.py
+++ b/reflex/utils/format.py
@@ -6,7 +6,7 @@
import json
import os
import re
-from typing import TYPE_CHECKING, Any, List, Union
+from typing import TYPE_CHECKING, Any, List, Optional, Unio... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} | |
reflex-dev__reflex-2291@202dba8 | reflex-dev/reflex | Python | 2,291 | apps should no longer call "app.compile()" | Closes #2293.
Two main tricks:
* To handle backend running, `gunicorn` or `uvicorn` to load a special shim `app_module_for_backend.py`, instead of the real app (e.g. `counter.py`).
* A new `get_compiled_app` function, which essentially does `get_app() + compile` - to account for the fact that `get_app()` will now ... | 2023-12-14T05:45:00Z | Remove `app.compile()` boilerplate line in apps.
**Describe the bug**
Reflex apps require `app.compile()` line. It would be better to not require that.
**To Reproduce**
Any existing reflex app in the wild.
**Expected behavior**
All existing reflex functionality works as usual, without `app.compile()` in apps. | [
{
"body": "**Describe the bug**\r\nReflex apps require `app.compile()` line. It would be better to not require that.\r\n\r\n**To Reproduce**\r\nAny existing reflex app in the wild.\r\n\r\n**Expected behavior**\r\nAll existing reflex functionality works as usual, without `app.compile()` in apps.\r\n",
"numb... | e245b6b6a0f90f71c8aba6e0299f4bcc31638330 | {
"head_commit": "202dba88ddc56d40bc8ada56c9339e4cfdaefd8e",
"head_commit_message": "get_compiled_app",
"patch_to_review": "diff --git a/README.md b/README.md\nindex 0b301ea0d01..874c2de9216 100644\n--- a/README.md\n+++ b/README.md\n@@ -123,7 +123,6 @@ def index():\n # Add state and page to the app.\n app = rx.Ap... | [
{
"diff_hunk": "@@ -619,6 +619,24 @@ def _should_compile(self) -> bool:\n return True\n \n def compile(self):\n+ \"\"\"App modules no longer need to call this.\n+\n+ Reflex framework will perform compile_() as needed.\n+\n+ Existing call sites will get a deprecation message.\n+\... | e896b43cc21e64c943af40ca9afe601f7dd45663 | diff --git a/README.md b/README.md
index 0b301ea0d01..874c2de9216 100644
--- a/README.md
+++ b/README.md
@@ -123,7 +123,6 @@ def index():
# Add state and page to the app.
app = rx.App()
app.add_page(index, title="reflex:DALL·E")
-app.compile()
```
## Let's break this down.
diff --git a/docs/es/README.md b/docs/e... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} | |
scikit-learn__scikit-learn-30122@0e41007 | scikit-learn/scikit-learn | Python | 30,122 | ENH move estimator type to tags | Closes https://github.com/scikit-learn/scikit-learn/issues/16469
Moving `_estimator_tpye` to tags.
Right now this doesn't work. | 2024-10-21T07:42:21Z | Deprecate _estimator_type, replace by estimator tag
I think we should deprecate ``_estimator_type`` which predates ``estimator_tags``. It's private but we should probably still deprecate it as it can lead to weird errors downstream. If we change it and downstream is not updating it, it would lead to classifiers using K... | Should we do the same with _pairwise?
I think so.
Not against it.
Another not directly related question: since we are integrating such tag, do we intend (or allow third-party) to use it outside of the common tests?
That was my intention. Right now both the ``_pairwise`` and ``_estimator_type`` are already used in t... | [
{
"body": "I think we should deprecate ``_estimator_type`` which predates ``estimator_tags``. It's private but we should probably still deprecate it as it can lead to weird errors downstream. If we change it and downstream is not updating it, it would lead to classifiers using KFold cross-validation, which can ... | e887583b933a50063081c2c012661ff13968adc3 | {
"head_commit": "0e410077b925a606bd8ab13787a037124cccad46",
"head_commit_message": "use method instead of operator",
"patch_to_review": "diff --git a/doc/api_reference.py b/doc/api_reference.py\nindex 86fa072d3ed25..b3e658bd22120 100644\n--- a/doc/api_reference.py\n+++ b/doc/api_reference.py\n@@ -1176,7 +1176,6 ... | [
{
"diff_hunk": "@@ -197,6 +198,52 @@ def score(self, X, y, *args, **kwargs):\n )\n \n \n+def _get_instance_with_pipeline(meta_estimator, init_params):\n+ \"\"\"Given a single meta-estimator instance, generate an instance with a pipeline\"\"\"\n+ if {\"estimator\", \"base_estimator\", \"regress... | 216f442fbc8bf0ab043a857fcec489c252648c36 | diff --git a/doc/api_reference.py b/doc/api_reference.py
index 86fa072d3ed25..b3e658bd22120 100644
--- a/doc/api_reference.py
+++ b/doc/api_reference.py
@@ -1176,7 +1176,6 @@ def _get_submodule(module_name, submodule_name):
"ClassifierTags",
"RegressorTags",
... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Code Refactoring / Architectural Improvement"
} |
scikit-learn__scikit-learn-29874@f45d62f | scikit-learn/scikit-learn | Python | 29,874 | TST check that binary only classifiers fail on multiclass data | Fixes https://github.com/scikit-learn/scikit-learn/issues/18005
Checks that if the estimator has `tags.classifier_tags.multi_class=False`, then it actually fails.
I would be happier if we had a better error message telling people how to fix their issue though. Not sure what to put there.
cc @chkoar @glemaitre | 2024-09-17T15:02:22Z | check_estimator and binary_only tag
Shouldn't the `check_estimator` check if you pass a non binary problem for a binary classifier? I mean, if you have an untagged binary estimator that raises an exception the `check_estimator` will fail. On the other hand, if you have a tagged estimator as `binary_only` the `check_es... | `check_estimator` does not take data as input. It will internally generate data based on the estimator tags.
Yes. I know that. I would expect that it will test a tagged binary classifier that will raise an exception on a non-binary test case.
Currently, if a classifier is tagged binary, it will only be passed binary pr... | [
{
"body": "Shouldn't the `check_estimator` check if you pass a non binary problem for a binary classifier? I mean, if you have an untagged binary estimator that raises an exception the `check_estimator` will fail. On the other hand, if you have a tagged estimator as `binary_only` the `check_estimator` will pas... | 325930e0151b67f6ddcd155d47a7b2a320f30ea7 | {
"head_commit": "f45d62f794820bdf3ed76217da4808aebd9f1ad6",
"head_commit_message": "Merge branch 'tests/tag-multiclass-false' of github.com:adrinjalali/scikit-learn into tests/tag-multiclass-false",
"patch_to_review": "diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst\nindex 0c4bf127c0b43..0cb7ac296d2... | [
{
"diff_hunk": "@@ -298,6 +304,17 @@ def type_of_target(y, input_name=\"\"):\n 'multilabel-indicator'\n \"\"\"\n xp, is_array_api_compliant = get_namespace(y)\n+\n+ def _raise_or_return():\n+ \"\"\"Depending on the value of raise_unknown, either raise an error or return\n+ 'unknown'... | 0f4289b0a8a0d3eff8d463e7bf76b9ab85232f4e | diff --git a/doc/whats_new/upcoming_changes/sklearn.utils/29874.enhancement.rst b/doc/whats_new/upcoming_changes/sklearn.utils/29874.enhancement.rst
new file mode 100644
index 0000000000000..58f3919af7c2c
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.utils/29874.enhancement.rst
@@ -0,0 +1,5 @@
+- :func:`~s... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-1984@313058a | reflex-dev/reflex | Python | 1,984 | fix serialization as a whole for list/dict/Base containing custom items to serialize | Improve the serialization logic to handle the case where we want to serialize a list of custom objects.
Before PR, a datetime can be serialized, but not a list of datetime or a dict containing datetime as value.
After PR, they can all be serialized.
Include the changes suggested by @ElijahAhianyo in #1977.
C... | 2023-10-17T12:42:26Z | Need for custom json serializer
During compile time, any non-serializable object encountered during the JSON serialization process is treated as a list, This means the code below:
```python
import reflex as rx
from datetime import datetime
from sqlmodel import Field
class User(rx.Model, table=True):
fir... | [
{
"body": "During compile time, any non-serializable object encountered during the JSON serialization process is treated as a list, This means the code below:\r\n\r\n```python\r\nimport reflex as rx\r\nfrom datetime import datetime\r\nfrom sqlmodel import Field\r\n\r\nclass User(rx.Model, table=True):\r\n ... | 864a66b20d9c26bbb4fbf674171857e288ab455a | {
"head_commit": "313058a370f4736912bc12e981f63d0538b02d83",
"head_commit_message": "serialization is fixed",
"patch_to_review": "diff --git a/reflex/base.py b/reflex/base.py\nindex 53510786b36..6a8d9cac158 100644\n--- a/reflex/base.py\n+++ b/reflex/base.py\n@@ -30,7 +30,9 @@ def json(self) -> str:\n Retu... | [
{
"diff_hunk": "@@ -93,6 +95,31 @@ def serialize_foo(value: Foo) -> str:\n assert not serializers.has_serializer(Foo)\n \n \n+class TestEnum(str, Enum):\n+ \"\"\"TestEnum.\"\"\"\n+\n+ FOO = \"foo\"\n+ BAR = \"bar\"\n+\n+\n+class TestEnum2(Enum):",
"line": null,
"original_line": 105,
"or... | 364b9def0219a6e389d605c65d010f8c10d05997 | diff --git a/reflex/base.py b/reflex/base.py
index 53510786b36..6a8d9cac158 100644
--- a/reflex/base.py
+++ b/reflex/base.py
@@ -30,7 +30,9 @@ def json(self) -> str:
Returns:
The object as a json string.
"""
- return self.__config__.json_dumps(self.dict(), default=list)
+ fr... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} | |
scikit-learn__scikit-learn-29842@af1847d | scikit-learn/scikit-learn | Python | 29,842 | FIX scoring != None for RidgeCV should used unscaled y for evaluation | closes https://github.com/scikit-learn/scikit-learn/issues/13998
closes https://github.com/scikit-learn/scikit-learn/pull/15648
While discussing with @jeromedockes, we recall to have observed something weird in the `RidgeCV` code. I check a bit closer and I open this PR to highlight what is the potential problem.
... | 2024-09-13T16:59:59Z | `RidgeCV` `cv_values_` are for preprocessed data: centered and scaled by sample weights.
when `store_cv_values=True`, `RidgeCV` stores the leave-one-out squared errors,
when `scoring=None`, or the leave-one-out predictions, when `scoring` is
provided by the user, in its `cv_values_` attribute.
However, when `scori... | ping @agramfort
yes `cv_values_` should be put in native space and fixed when using a custom scoring. | [
{
"body": "when `store_cv_values=True`, `RidgeCV` stores the leave-one-out squared errors,\r\nwhen `scoring=None`, or the leave-one-out predictions, when `scoring` is\r\nprovided by the user, in its `cv_values_` attribute.\r\n\r\nHowever, when `scoring` is not `None`, it stores the predictions for the\r\nprepro... | 25bc29ab9703e1c7dba2af29943c9cf616124290 | {
"head_commit": "af1847dc8fcc1308f7b268e5172743dcb789bb8b",
"head_commit_message": "address thomas review + n_target test",
"patch_to_review": "diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst\nindex 22a0d7acfd24e..bf141b582b365 100644\n--- a/doc/whats_new/v1.6.rst\n+++ b/doc/whats_new/v1.6.rst\n@@ -... | [
{
"diff_hunk": "@@ -2178,13 +2179,20 @@ def fit(self, X, y, sample_weight=None, score_params=None):\n self.cv_results_[:, i] = squared_errors.ravel()\n else:\n predictions = y - (c / G_inverse_diag)\n+ # represent the predictions in the original sca... | e9bd7784b041a59c8f862b3e8883489316b3ce65 | diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst
index 22a0d7acfd24e..2a7211ca368b5 100644
--- a/doc/whats_new/v1.6.rst
+++ b/doc/whats_new/v1.6.rst
@@ -257,6 +257,15 @@ Changelog
for the calculation of test scores.
:pr:`29419` by :user:`Shruti Nath <snath-xoc>`.
+- |Fix| :class:`linear_model.RidgeC... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Code Refactoring / Architectural Improvement"
} |
scikit-learn__scikit-learn-30616@586d89a | scikit-learn/scikit-learn | Python | 30,616 | Fixes API Deprecate n_alphas in LinearModelCV | Fixes #30467
Updated LinearModelCV and derived classes LassoCV, ElasticNetCV, MultiTaskElasticNetCV, MultiTaskLassoCV to remove `n_alphas` parameter from the constructor. `alphas` parameter is updated to support an integer or array-like argument. Functionality of `n_alphas` is preserved by passing an integer to `alp... | 2025-01-09T17:48:13Z | API Deprecate n_alphas in LinearModelCV
In LassoCV, ElasticNetCV, ... we have two parameters, `alphas` and `n_alphas`, that have the same purpose, i.e. determine the alpha values to test.
I'd be in favor of deprecating `n_alphas` and make `alphas` accept either an int or an array-like, filling both roles.
I chose... | Hi, can I take a shot at refactoring?
sure ! You can take a look at the [contributing guide](https://scikit-learn.org/dev/developers/contributing.html) and especially the [deprecation section](https://scikit-learn.org/dev/developers/contributing.html#deprecation) to get started. | [
{
"body": "In LassoCV, ElasticNetCV, ... we have two parameters, `alphas` and `n_alphas`, that have the same purpose, i.e. determine the alpha values to test.\r\n\r\nI'd be in favor of deprecating `n_alphas` and make `alphas` accept either an int or an array-like, filling both roles.\r\n\r\nI chose to keep `alp... | eec4449e2b36f051baabfb6d20c097fc9e83bf65 | {
"head_commit": "586d89a8c47a7e9fe96c6af78aa6f8aba46ddec6",
"head_commit_message": "Merge branch 'main' into LinearModelCV_api_refactor",
"patch_to_review": "diff --git a/doc/whats_new/upcoming_changes/sklearn.linear_model/30616.api.rst b/doc/whats_new/upcoming_changes/sklearn.linear_model/30616.api.rst\nnew fil... | [
{
"diff_hunk": "@@ -1691,39 +1700,61 @@ def fit(self, X, y, sample_weight=None, **params):\n l1_ratios = [\n 1,\n ]\n+\n path_params.pop(\"cv\", None)\n path_params.pop(\"n_jobs\", None)\n \n- alphas = self.alphas\n+ alphas = None\n n... | 40077003563784b90ea2505ee03d479218177237 | diff --git a/doc/whats_new/upcoming_changes/sklearn.linear_model/30616.api.rst b/doc/whats_new/upcoming_changes/sklearn.linear_model/30616.api.rst
new file mode 100644
index 0000000000000..8d0a032fd284f
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.linear_model/30616.api.rst
@@ -0,0 +1,9 @@
+The parameter ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-30187@f73913f | scikit-learn/scikit-learn | Python | 30,187 | FIX Check and correct the input_tags.sparse flag | #### Reference Issues/PRs
Fixes #30139
#### What does this implement/fix? Explain your changes.
The test `check_estimator_sparse_tag` is added.
It checks that the `input_tags.sparse` tag of an estimator is consistent:
- if `input_tags.sparse=True` the estimator can be fitted on sparse data
- if `input_ta... | 2024-10-31T16:58:20Z | The input_tags.sparse flag is often incorrect
### Describe the bug
If I understood correctly the developer API for tags, `input_tags.sparse` tells us whether an estimator can accept sparse data or not. For many estimators it seems that `input_tags.sparse` is False but should be True.
### Steps/Code to Reproduce
```p... | The current estimator check `_check_estimator_sparse_container` does not rely on the presence of the tag: it just calls the estimator with sparse input and checks that the word "sparse" appears somewhere in the exception message if a `ValueError` is raised.
This is quite a low requirement. We could make this a stron... | [
{
"body": "### Describe the bug\n\nIf I understood correctly the developer API for tags, `input_tags.sparse` tells us whether an estimator can accept sparse data or not. For many estimators it seems that `input_tags.sparse` is False but should be True.\n\n### Steps/Code to Reproduce\n\n```python\r\nfrom sklearn... | 8481e681128822050e996c3ef6b65d3801d102f5 | {
"head_commit": "f73913fe3a81691c5b879ae4a5c6e89aabffe325",
"head_commit_message": "Merge remote-tracking branch 'upstream/main' into fix_input_sparse_tag",
"patch_to_review": "diff --git a/doc/whats_new/upcoming_changes/changed-models/30187.fix.rst b/doc/whats_new/upcoming_changes/changed-models/30187.fix.rst\n... | [
{
"diff_hunk": "@@ -0,0 +1,4 @@\n+- :func:`utils.estimator_checks.check_estimator_sparse_tag` ensures that\n+ the estimator tag `input_tags.sparse` is consistent with its `fit`\n+ method (acceting sparse input `X` or raising the appropriate error).",
"line": null,
"original_line": 3,
"original_sta... | b54a408c0bad6dd4eaa5329cf29acbec3e43bdb9 | diff --git a/doc/whats_new/upcoming_changes/changed-models/30187.fix.rst b/doc/whats_new/upcoming_changes/changed-models/30187.fix.rst
new file mode 100644
index 0000000000000..001b8840d9a7b
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/changed-models/30187.fix.rst
@@ -0,0 +1,2 @@
+- The `tags.input_tags.sparse` f... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-2010@120e5d8 | reflex-dev/reflex | Python | 2,010 | Improve Var type handling for better rx.Model attribute access | Support accessing relationship attributes on Models
Support accessing all possible attributes when Var is annotated as a Union (fix #1814)
Use null coalescing when the var is Optional or has a default value of None
`reflex.utils.types.is_union` supports new style `UnionType` on py3.10 | 2023-10-20T22:12:59Z | rx.foreach Union support
In my app I'm trying to use an union to hold slightly different classes (common base members but extra members) and using rx.foreach to try to display those entities. If I specify the union, the foreach component won't run, citing :
```AttributeError: The State var `_` has no attribute 'MY_A... | > `if isinstance(entry, Reaction2):`
This line is problematic because `entry` is always going to be a `BaseVar`, not an actual instance of the class it originated from.
The code that comprises the frontend is compiled into javascript code that runs in the browser, so any python conditions and type differentiation... | [
{
"body": "In my app I'm trying to use an union to hold slightly different classes (common base members but extra members) and using rx.foreach to try to display those entities. If I specify the union, the foreach component won't run, citing :\r\n\r\n```AttributeError: The State var `_` has no attribute 'MY_ATT... | f404205c3fa11ff854ada60fd97e4d1725fb2eb8 | {
"head_commit": "120e5d81c37f027c0c679b626df9b99835d081f6",
"head_commit_message": "types: Only check for UnionType if it is present",
"patch_to_review": "diff --git a/reflex/model.py b/reflex/model.py\nindex 12703e7be2b..69a3bbf5369 100644\n--- a/reflex/model.py\n+++ b/reflex/model.py\n@@ -15,6 +15,7 @@\n impor... | [
{
"diff_hunk": "@@ -85,7 +87,58 @@ def is_literal(cls: GenericType) -> bool:\n Returns:\n Whether the class is a literal.\n \"\"\"\n- return hasattr(cls, \"__origin__\") and cls.__origin__ is Literal\n+ return get_origin(cls) is Literal\n+\n+\n+def is_optional(cls: GenericType) -> bool:\n+... | fcc12c9d33b55887e8963da21fbc61ebf09ecb1f | diff --git a/reflex/model.py b/reflex/model.py
index 12703e7be2b..3167782e5f8 100644
--- a/reflex/model.py
+++ b/reflex/model.py
@@ -15,6 +15,7 @@
import alembic.script
import alembic.util
import sqlalchemy
+import sqlalchemy.orm
import sqlmodel
from reflex import constants
@@ -68,6 +69,22 @@ def __init_subclass... | {
"difficulty": "high",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} |
reflex-dev__reflex-1842@cca8296 | reflex-dev/reflex | Python | 1,842 | add option to disable Tailwind | This PR provides an option to disable the out-of-the-box supported Tailwind. To do that, simply set the `tailwind` config to `None` :
`rxconfig.py`
---------------------------------
```python
import reflex as rx
config = ExampleConfig(
app_name="example",
tailwind= None
)
```
fixes #1829
Doc... | 2023-09-20T11:13:57Z | Don't have Tailwind defaulty included in reflex app
Currently Tailwind is added to each fresh initialization of Reflex app. This may interfere with some other css libraries.
The `tailwindcss` module is installed defaulty. Also the is a default style style file in `.web\styles\tailwind.css`.
It would be nice, if i... | [
{
"body": "Currently Tailwind is added to each fresh initialization of Reflex app. This may interfere with some other css libraries.\r\n\r\nThe `tailwindcss` module is installed defaulty. Also the is a default style style file in `.web\\styles\\tailwind.css`.\r\n\r\nIt would be nice, if it was included only opt... | 33ee0675d44b1e193ab207f4a1e27c89dee58a5d | {
"head_commit": "cca8296f4931e1e14e5e37d8f91be0d619496218",
"head_commit_message": "fix precommit",
"patch_to_review": "diff --git a/reflex/.templates/web/package.json b/reflex/.templates/web/package.json\nindex 191e7a4e50d..09c44220fa4 100644\n--- a/reflex/.templates/web/package.json\n+++ b/reflex/.templates/we... | [
{
"diff_hunk": "@@ -432,13 +432,23 @@ def install_frontend_packages(packages: set[str]):\n processes.show_status(\"Installing base frontend packages\", process)\n \n config = get_config()\n- if config.tailwind is not None and \"plugins\" in config.tailwind:\n+ if config.tailwind is not None:\n+ ... | 2721d8e41639be89009805fd3c03fd19268f4ca6 | diff --git a/reflex/.templates/web/package.json b/reflex/.templates/web/package.json
index 191e7a4e50d..09c44220fa4 100644
--- a/reflex/.templates/web/package.json
+++ b/reflex/.templates/web/package.json
@@ -24,7 +24,6 @@
},
"devDependencies": {
"autoprefixer": "^10.4.14",
- "postcss": "^8.4.24",
- "t... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Dependency Updates & Env Compatibility"
} | |
reflex-dev__reflex-2012@4ab1245 | reflex-dev/reflex | Python | 2,012 | Support Form controls via name attribute (no ID or ref) | Dynamic refs / ids don't really work... but dynamic `name` is not a problem!
```python
import reflex as rx
class State(rx.State):
n_fields: int = 0
grid: int = 1
def handle_submit(self, form_data: dict):
print(form_data)
def add_field(self):
self.n_fields += 1
d... | 2023-10-21T10:11:31Z | Short-hand for clearing a form after submitting
Please create an alternative method that can set all the elements inside a form to "" after the on_submit takes place.
We used reflex in CalHacks 10.0, and this is one of the enhancements we think would be great! | [
{
"body": "Please create an alternative method that can set all the elements inside a form to \"\" after the on_submit takes place.\r\n\r\nWe used reflex in CalHacks 10.0, and this is one of the enhancements we think would be great!",
"number": 2074,
"title": "Short-hand for clearing a form after submit... | 96eca4f690240accaab3ee45e1e366219a4a46b5 | {
"head_commit": "4ab12457d24b66f6dfe9b2df858bb9600ca7b595",
"head_commit_message": "form: useCallback to memoize handleSubmit",
"patch_to_review": "diff --git a/integration/test_form_submit.py b/integration/test_form_submit.py\nindex a9611ecc175..b8d0fd0bde3 100644\n--- a/integration/test_form_submit.py\n+++ b/i... | [
{
"diff_hunk": "@@ -625,3 +625,25 @@ def unescape_double_quotes_in_var(m: re.Match) -> str:\n string=value,\n flags=re.VERBOSE,\n )\n+\n+\n+def collect_form_dict_names(form_dict: dict[str, Any]) -> dict[str, Any]:\n+ \"\"\"Collapse keys with consecutive suffixes into a single list value.\... | 505f1e833aa82f3d6f78f07971c0f33b7a6edae9 | diff --git a/integration/test_form_submit.py b/integration/test_form_submit.py
index 25296b3d4eb..4a9f6c2d10a 100644
--- a/integration/test_form_submit.py
+++ b/integration/test_form_submit.py
@@ -7,6 +7,7 @@
from selenium.webdriver.common.keys import Keys
from reflex.testing import AppHarness
+from reflex.utils im... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} | |
scikit-learn__scikit-learn-29738@9ec3865 | scikit-learn/scikit-learn | Python | 29,738 | PERF speedup classification_report by attaching unique values to dtype.metadata | Fixes #26808
Closes #26820
This is alternative to https://github.com/scikit-learn/scikit-learn/pull/26820 where we attach unique values to the `dtype.metadata` of a view on `y`.
This gets the same speedup as reported in #26820 but is a lot cleaner IMO.
WDYT @ogrisel @glemaitre @thomasjpfan
(I'm still wor... | 2024-08-29T11:50:54Z | Speed up classification_report
### Describe the workflow you want to enable
I'm concerned with slow execution speed of the classification_report procedure which makes it barely suitable for production-grade workloads.
On a 8M sample it already takes 15 seconds whereas simple POC numba implementation takes 20 to 40... | I have actually improved it slightly to have more concise variable names & handle multiclasses. The only things of sklearn's classification report it does not support are micro,sample averages and sample weights (not hard to add them as well). As an extra it has balanced accuracy though.
```python
@njit()
def fast... | [
{
"body": "### Describe the workflow you want to enable\r\n\r\nI'm concerned with slow execution speed of the classification_report procedure which makes it barely suitable for production-grade workloads.\r\nOn a 8M sample it already takes 15 seconds whereas simple POC numba implementation takes 20 to 40 **mil... | a5ea01c838aef012d54c5859a251a954105a5015 | {
"head_commit": "9ec38652fe3e84a4174a7bd9402282e4398da738",
"head_commit_message": "add another test",
"patch_to_review": "diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst\nindex d4d373e9bf280..347fa429e12d3 100644\n--- a/doc/whats_new/v1.6.rst\n+++ b/doc/whats_new/v1.6.rst\n@@ -271,6 +271,9 @@ Chang... | [
{
"diff_hunk": "@@ -0,0 +1,96 @@\n+# Authors: The scikit-learn developers\n+# SPDX-License-Identifier: BSD-3-Clause\n+\n+import numpy as np\n+\n+from sklearn.utils._array_api import get_namespace\n+\n+\n+def _attach_unique(y):\n+ \"\"\"Attach unique values of y to y and return the result.\n+\n+ The result... | c6ced7cdda8d92a0dd40e38e4f468177686993ed | diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst
index d4d373e9bf280..6bb6eab32464e 100644
--- a/doc/whats_new/v1.6.rst
+++ b/doc/whats_new/v1.6.rst
@@ -271,6 +271,10 @@ Changelog
:pr:`29210` by :user:`Marc Torrellas Socastro <marctorsoc>` and
:user:`Stefanie Senger <StefanieSenger>`.
+- |Efficiency... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Performance Optimizations"
} |
reflex-dev__reflex-3624@d1f537f | reflex-dev/reflex | Python | 3,624 | rx._x.asset improvements | Closes #3623 | 2024-07-04T18:53:07Z | Multiple rx._x.asset issues
- Assets should not be checked and copied in backend only mode. Even if you do not put them in globals, they will for example be executed if you use them in a `@rx.page` which will be loaded. Even in backend only mode.
- Local/Normal asset support - Assets which are in the local `assets/` d... | [
{
"body": "- Assets should not be checked and copied in backend only mode. Even if you do not put them in globals, they will for example be executed if you use them in a `@rx.page` which will be loaded. Even in backend only mode.\r\n- Local/Normal asset support - Assets which are in the local `assets/` director... | 34c11fdf108395fcacbe2488655969fef5c7958c | {
"head_commit": "d1f537f50cb6be9e80024c85887b0d457d6c2618",
"head_commit_message": "fix darglint",
"patch_to_review": "diff --git a/reflex/constants/__init__.py b/reflex/constants/__init__.py\nindex e974ab91583..185c2ee3e00 100644\n--- a/reflex/constants/__init__.py\n+++ b/reflex/constants/__init__.py\n@@ -2,6 +... | [
{
"diff_hunk": "@@ -1,57 +1,90 @@\n \"\"\"Helper functions for adding assets to the app.\"\"\"\n \n import inspect\n+import os\n from pathlib import Path\n from typing import Optional\n \n from reflex import constants\n \n \n-def asset(relative_filename: str, subfolder: Optional[str] = None) -> str:\n- \"\"\... | 45e2852f52ad5f00ed6e79a5c9a17106fd558aaa | diff --git a/reflex/__init__.py b/reflex/__init__.py
index 3941542f2d6..56252441601 100644
--- a/reflex/__init__.py
+++ b/reflex/__init__.py
@@ -264,6 +264,7 @@
"experimental": ["_x"],
"admin": ["AdminDash"],
"app": ["App", "UploadFile"],
+ "assets": ["asset"],
"base": ["Base"],
"components.... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} | |
scikit-learn__scikit-learn-29693@5bddd8c | scikit-learn/scikit-learn | Python | 29,693 | DOC clarify that n_jobs argument uses threading in pairwise_distances and pairwise_kernels | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2024-08-20T08:57:36Z | Truly parallel execution of pairwise_kernels and pairwise_distances
### Describe the workflow you want to enable
Both `pairwise_kernels` and `pairwise_distances` functions call `_parallel_pairwise` function, which is (contrary to its name) not parallel as it enforces the threading backend. Therefore, these functions... | > which is (contrary to its name) not parallel as it enforces the threading backend.
> ... Therefore, these functions are terribly slow, especially for computationally expensive user-defined metrics.
Some CPU intensive operations in numpy or pandas do release the GIL, so the threading backend might be a good choice... | [
{
"body": "### Describe the workflow you want to enable\r\n\r\nBoth `pairwise_kernels` and `pairwise_distances` functions call `_parallel_pairwise` function, which is (contrary to its name) not parallel as it enforces the threading backend. Therefore, these functions are terribly slow, especially for computatio... | f38aa4923330b593c1e5319ab8aa043844582fc5 | {
"head_commit": "5bddd8cada8d3f97122ea8f7d9a5416a2a0efc29",
"head_commit_message": "clarification of the documentation for n_jobs argument",
"patch_to_review": "diff --git a/sklearn/metrics/pairwise.py b/sklearn/metrics/pairwise.py\nindex fd08850ce47b3..771da7e1a1d9a 100644\n--- a/sklearn/metrics/pairwise.py\n++... | [
{
"diff_hunk": "@@ -2597,8 +2597,8 @@ def pairwise_kernels(\n \n n_jobs : int, default=None\n The number of jobs to use for the computation. This works by breaking\n- down the pairwise matrix into n_jobs even slices and computing them in\n- parallel.\n+ down the pairwise matrix ... | dfcd469d2d46f182ea759227110852583137d912 | diff --git a/sklearn/metrics/pairwise.py b/sklearn/metrics/pairwise.py
index fd08850ce47b3..c27bc22583660 100644
--- a/sklearn/metrics/pairwise.py
+++ b/sklearn/metrics/pairwise.py
@@ -1938,7 +1938,7 @@ def _dist_wrapper(dist_func, dist_matrix, slice_, *args, **kwargs):
def _parallel_pairwise(X, Y, func, n_jobs, **k... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Performance Optimizations"
} |
scikit-learn__scikit-learn-29708@98bce9b | scikit-learn/scikit-learn | Python | 29,708 | FIX #13117: implement fit methods in subclasses of LinearModelCV | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2024-08-23T14:42:26Z | Possible error in documentation of ElasticNetCV()
The documentation for class `ElasticNetCV()` reads:
y : array-like, shape (n_samples,) or (n_samples, n_targets)
However, when I try to pass a 2D array I get the error:
```
---------------------------------------------------------------------------
ValueE... | would you like to submit a PR to fix it @albertotb ?
Sure, I'll get to it
On Sat, Feb 9, 2019 at 2:51 PM Adrin Jalali <notifications@github.com>
wrote:
> would you like to submit a PR to fix it @albertotb
> <https://github.com/albertotb> ?
>
> —
> You are receiving this because you were mentioned.
> Reply to this ema... | [
{
"body": "The documentation for class `ElasticNetCV()` reads:\r\n\r\n y : array-like, shape (n_samples,) or (n_samples, n_targets)\r\n\r\nHowever, when I try to pass a 2D array I get the error:\r\n\r\n```\r\n---------------------------------------------------------------------------\r\nValueError ... | ede02b4d000d3f9334940cf5c81155f9dc070e77 | {
"head_commit": "98bce9bd4d8f5cf1f3029e66b7f8a8fe201f35be",
"head_commit_message": "remove duplicate fits",
"patch_to_review": "diff --git a/sklearn/linear_model/_coordinate_descent.py b/sklearn/linear_model/_coordinate_descent.py\nindex 4ede6c04b462b..544cdb1412cc6 100644\n--- a/sklearn/linear_model/_coordinate... | [
{
"diff_hunk": "@@ -2322,6 +2362,46 @@ def _is_multitask(self):\n def _more_tags(self):\n return {\"multioutput\": False}\n \n+ def fit(self, X, y, sample_weight=None, **params):\n+ \"\"\"Fit ElasticNet model with coordinate descent.\n+\n+ Fit is on grid of alphas and best alpha est... | 0037db53d67c03e20f1ec45ed124054a716de9cb | diff --git a/sklearn/linear_model/_coordinate_descent.py b/sklearn/linear_model/_coordinate_descent.py
index 4ede6c04b462b..60ffa0b8f468c 100644
--- a/sklearn/linear_model/_coordinate_descent.py
+++ b/sklearn/linear_model/_coordinate_descent.py
@@ -2083,6 +2083,46 @@ def _is_multitask(self):
def _more_tags(self):
... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Documentation Updates"
} |
reflex-dev__reflex-1458@f2c96ae | reflex-dev/reflex | Python | 1,458 | Remove curl and parallelize node/bun install | Use `httpx` to install nvm and bun instead of curl, removing one dependency. Also parallelize the process to install nvm/bun during `reflex init` | 2023-07-28T22:15:11Z | Download node and bun in parallel
Currently we download node and bun sequentially during the initial `reflex init`. We can parallelize this.
Curl dependency in Bun and NVM installs
It would be good to remove this curl dependency. Instead we can use Python to download the relevant installation script and then run them... | [
{
"body": "Currently we download node and bun sequentially during the initial `reflex init`. We can parallelize this.",
"number": 1434,
"title": "Download node and bun in parallel"
},
{
"body": "It would be good to remove this curl dependency. Instead we can use Python to download the relevant ... | 0ebe1529a67e6395307415fa0fbf69b761ac69d4 | {
"head_commit": "f2c96ae32d4d3c07867dd823703c9f2ebcc26314",
"head_commit_message": "Parallelize node/bun install",
"patch_to_review": "diff --git a/reflex/constants.py b/reflex/constants.py\nindex d4f7725ae63..7fdd6a248bc 100644\n--- a/reflex/constants.py\n+++ b/reflex/constants.py\n@@ -45,54 +45,57 @@ def get_v... | [
{
"diff_hunk": "@@ -319,27 +336,35 @@ def install_bun():\n \"\"\"\n # Bun is not supported on Windows.\n if IS_WINDOWS:\n- console.log(\"Skipping bun installation on Windows.\")\n return\n \n- # Only install if bun is not already installed.\n- if not os.path.exists(constants.BUN... | 08ca4530d44668f86c65c96ce5895f9e35b276f7 | diff --git a/reflex/constants.py b/reflex/constants.py
index d4f7725ae63..7fdd6a248bc 100644
--- a/reflex/constants.py
+++ b/reflex/constants.py
@@ -45,54 +45,57 @@ def get_value(key: str, default: Any = None, type_: Type = str) -> Type:
# The current version of Reflex.
VERSION = metadata.version(MODULE_NAME)
-# Pr... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Code Refactoring / Architectural Improvement"
} | |
reflex-dev__reflex-1361@ef37bd9 | reflex-dev/reflex | Python | 1,361 | rx.App `state` arg should not be required | With this PR, the `state` arg for `rx.App` will be optional, which means you dont necessarily have to specify a state as it will be inferred. However an exception is thrown if there are multiple state classes that subclass `rx.State`.
Allowed:
```python
import reflex as rx
class AppState(rx.State):
pass... | 2023-07-18T12:18:28Z | rx.App doesn't need to specify state
Currently we require the user specify their state when creating the app i.e. `app = rx.App(state=State)`
We can remove this because we can infer the base state by checking what has subclassed `rx.State`. This will also let us throw an error when `rx.State` has multiple subclasses... | [
{
"body": "Currently we require the user specify their state when creating the app i.e. `app = rx.App(state=State)`\r\n\r\nWe can remove this because we can infer the base state by checking what has subclassed `rx.State`. This will also let us throw an error when `rx.State` has multiple subclasses, which is a c... | 915685c07374b376c001334e1629026c2c42d5a2 | {
"head_commit": "ef37bd998e0711754fd572b9ce7f8de846f614a0",
"head_commit_message": "refactor",
"patch_to_review": "diff --git a/reflex/app.py b/reflex/app.py\nindex 235c0d87ac3..6aa379de033 100644\n--- a/reflex/app.py\n+++ b/reflex/app.py\n@@ -2,6 +2,7 @@\n \n import asyncio\n import inspect\n+import os\n from m... | [
{
"diff_hunk": "@@ -100,8 +101,25 @@ def __init__(self, *args, **kwargs):\n \n Raises:\n ValueError: If the event namespace is not provided in the config.\n+ Also, if there are multiple client subclasses of rx.State(Subclasses of rx.State should consist\n+ ... | 3751d0ece33634f8eb413216c5a81954c19dba91 | diff --git a/reflex/app.py b/reflex/app.py
index 235c0d87ac3..b5731b24f2f 100644
--- a/reflex/app.py
+++ b/reflex/app.py
@@ -2,6 +2,7 @@
import asyncio
import inspect
+import os
from multiprocessing.pool import ThreadPool
from typing import (
Any,
@@ -42,7 +43,7 @@
verify_route_validity,
)
from reflex... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} | |
reflex-dev__reflex-1111@9648173 | reflex-dev/reflex | Python | 1,111 | UI connection warning: | This Pr adds a UI warning when server is unreachable
<img width="1512" alt="Screenshot 2023-05-31 at 3 34 45 PM" src="https://github.com/pynecone-io/pynecone/assets/19895635/6bf310bd-351d-4ceb-b08b-2473bfe457a2">
You also have the option of modifying the component that gets displayed.
```python
# Add state an... | 2023-05-31T15:40:05Z | Show warning if can't connect to api_url
When deploying an app if you don't update the `api_url` in your `pcconfig.py` currently the app doesn't work and you have to go in the Network tab to see the error. We should have a way to throw a UI warning when the backend isn't connected. | Hi @picklelo @Alek99, I would like to take up this issue. I have gone through the deploy function in the pc.py file. I got some understanding. Please brief this issue so that I can take this and I work on this. | [
{
"body": "When deploying an app if you don't update the `api_url` in your `pcconfig.py` currently the app doesn't work and you have to go in the Network tab to see the error. We should have a way to throw a UI warning when the backend isn't connected.",
"number": 971,
"title": "Show warning if can't co... | d4b5c78406a90496b561ff1b66d54d9b6db43eed | {
"head_commit": "9648173ec539789356480efa9232cab4f0b2a551",
"head_commit_message": "added ommited libraries to default",
"patch_to_review": "diff --git a/pynecone/.templates/jinja/web/pages/index.js.jinja2 b/pynecone/.templates/jinja/web/pages/index.js.jinja2\nindex 17686aebe5a..e1293616c96 100644\n--- a/pynecon... | [
{
"diff_hunk": "@@ -70,7 +71,21 @@ export default function Component() {\n {% endfor %}\n \n return (\n+ <Fragment>\n+ <Fragment>\n+ {notConnected ? (\n+ <Box sx={% raw %}{{\"textAlign\": \"center\"}}{% endraw %}>",
"line": null,
"original_line": 77,
"original_start_line": ... | 3b239b190876fc353dcf3017eb147a24fd2f2e0b | diff --git a/pynecone/.templates/jinja/web/pages/index.js.jinja2 b/pynecone/.templates/jinja/web/pages/index.js.jinja2
index 17686aebe5a..32dcc11279f 100644
--- a/pynecone/.templates/jinja/web/pages/index.js.jinja2
+++ b/pynecone/.templates/jinja/web/pages/index.js.jinja2
@@ -11,6 +11,7 @@
export default function Comp... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-29436@5074d55 | scikit-learn/scikit-learn | Python | 29,436 | Fix tests for numpy 2 and array api compat | Fixes #29396.
Update the array API tests and the `_average` private helper for compatibility with NumPy 2.0.0. | 2024-07-08T15:53:02Z | Array API tests fail on main
### Describe the bug
I ran the Array API tests on main and got 10 failing tests.
(Last week, with an older main and everything else the same, I had 4 failing tests.)
array_api_compat==1.7.1
I only ran the cpu tests.
### Steps/Code to Reproduce
`pytest sklearn/utils/tests/te... | I do have the same issues locally, it seems like the problem disappear when using `numpy<2` (tried with numpy 1.26.4) for some reason ...
Maybe @betatim or @OmarManzoor have some suggestions about this and whether this is expected?
@lesteve Might be because of some updates in the new version of numpy. Will have to c... | [
{
"body": "### Describe the bug\r\n\r\nI ran the Array API tests on main and got 10 failing tests. \r\n(Last week, with an older main and everything else the same, I had 4 failing tests.)\r\n\r\narray_api_compat==1.7.1\r\n\r\nI only ran the cpu tests.\r\n\r\n### Steps/Code to Reproduce\r\n\r\n`pytest sklearn/ut... | 0dba98f1eb7edc893e26c36d54519386a265467d | {
"head_commit": "5074d5562e48142b571ef434ce960454eec0c179",
"head_commit_message": "TST remove numpy deprecation warnings in _atol_for_type by assuming NumPy's default floating point precision level",
"patch_to_review": "diff --git a/build_tools/azure/pylatest_pip_openblas_pandas_environment.yml b/build_tools/az... | [
{
"diff_hunk": "@@ -179,7 +184,10 @@ def test_average(\n with config_context(array_api_dispatch=True):\n result = _average(array_in, axis=axis, weights=weights, normalize=normalize)\n \n- assert getattr(array_in, \"device\", None) == getattr(result, \"device\", None)\n+ if np_version < parse_v... | 0ee1a213fa2355f4ee0407108e114a0976d1f394 | diff --git a/build_tools/azure/pylatest_pip_openblas_pandas_environment.yml b/build_tools/azure/pylatest_pip_openblas_pandas_environment.yml
index adb7add7622e1..c0d6aeaa717c0 100644
--- a/build_tools/azure/pylatest_pip_openblas_pandas_environment.yml
+++ b/build_tools/azure/pylatest_pip_openblas_pandas_environment.yml... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-700@d497918 | reflex-dev/reflex | Python | 700 | [fix issue 632] textarea rendering issue | ### All Submissions:
- [x] Have you followed the guidelines stated in [CONTRIBUTING.md](https://github.com/pynecone-io/pynecone/blob/main/CONTRIBUTING.md) file?
- [x] Have you checked to ensure there aren't any other open [Pull Requests](https://github.com/pynecone-io/pynecone/pulls ) for the desired changed?
##... | 2023-03-17T09:16:07Z | How to solve `pc.text_area` state render problem?
**Describe the bug**
Hi,
I am trying a create a fully controlled `pc.text_area`, so that can be cleared by a button.
Using `pc.text_area(value=State.area_value, on_change=State.set_area_value)`, when I move the cursor inside the text and after key up, the cusor will... | According to [React document](https://reactjs.org/docs/forms.html) this is the right way.
I think thera are two reasons the cursor jumps to end of text:
1. reset the input value with state's value.
Textare with only `value=State.area_value` always update input.
3. update the input value with on change eve... | [
{
"body": "**Describe the bug**\r\nHi, \r\nI am trying a create a fully controlled `pc.text_area`, so that can be cleared by a button.\r\nUsing `pc.text_area(value=State.area_value, on_change=State.set_area_value)`, when I move the cursor inside the text and after key up, the cusor will be auto focus at the end... | e96f1c4d39e8ffd05a2373bf82967111a1232083 | {
"head_commit": "d49791861100140d031bc162b7abec52b64c1404",
"head_commit_message": "fix fully controled textarea inssue.",
"patch_to_review": "diff --git a/pynecone/compiler/templates.py b/pynecone/compiler/templates.py\nindex c8731f963ea..39fb89c9bdc 100644\n--- a/pynecone/compiler/templates.py\n+++ b/pynecone/... | [
{
"diff_hunk": "@@ -466,6 +479,27 @@ def get_custom_components(\n custom_components |= child.get_custom_components(seen=seen)\n return custom_components\n \n+ def is_full_control(self, kwargs: dict) -> bool:\n+ \"\"\"Return if the component is fully controled input.\n+\n+ Ar... | 7141eb798c2ceebabdfe9abd659d0f274580d337 | diff --git a/pynecone/compiler/templates.py b/pynecone/compiler/templates.py
index c8731f963ea..39fb89c9bdc 100644
--- a/pynecone/compiler/templates.py
+++ b/pynecone/compiler/templates.py
@@ -154,7 +154,15 @@ def format_state(
"}})",
]
).format
-
+FULL_CONTROL = path_ops.join(
+ [
+ "{{setStat... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-29401@819e167 | scikit-learn/scikit-learn | Python | 29,401 | FIX `TransformedTargetRegressor` warns when `set_output` expects dataframe | #### Reference Issues/PRs
closes #29361
#### What does this implement/fix? Explain your changes.
Prevents warnings on output type to be raised on `TransformedTargetRegressor().fit()` when global output is set to "pandas" or "polars".
#### Any other comments?
Did this together with @lesteve: we figured that `Tr... | 2024-07-03T13:00:00Z | TransformedTargetRegressor warns about set_output set to pandas
### Describe the bug
If `set_output` is set to `"pandas"`, `TransformedTargetRegressor` warns unnecessarily.
### Steps/Code to Reproduce
```python
import numpy as np
import pandas as pd
from sklearn import set_config
from sklearn.compose import Tran... | Thanks for the reproducer, this seems like an issue indeed, I'll try to take a closer look soonish. | [
{
"body": "### Describe the bug\n\nIf `set_output` is set to `\"pandas\"`, `TransformedTargetRegressor` warns unnecessarily.\n\n### Steps/Code to Reproduce\n\n```python\r\nimport numpy as np\r\nimport pandas as pd\r\nfrom sklearn import set_config\r\nfrom sklearn.compose import TransformedTargetRegressor\r\nfro... | ecdc9570953f21d702e3e96394133abdad8e3b6a | {
"head_commit": "819e167df9020453ac2f26e8d0230e97f0de6b20",
"head_commit_message": "clearer documentation for args None and 'default'",
"patch_to_review": "diff --git a/sklearn/_config.py b/sklearn/_config.py\nindex fc9392de68df6..e8da4bcaa3381 100644\n--- a/sklearn/_config.py\n+++ b/sklearn/_config.py\n@@ -142,... | [
{
"diff_hunk": "@@ -393,3 +393,17 @@ def test_transform_target_regressor_pass_extra_predict_parameters():\n regr.fit(X, y)\n regr.predict(X, check_input=False)\n assert regr.regressor_.predict_called\n+\n+\n+@pytest.mark.filterwarnings(\"error\")",
"line": null,
"original_line": 398,
"or... | 792bd4abe657f94a2c7d4297db35e47afdeaa061 | diff --git a/doc/whats_new/v1.5.rst b/doc/whats_new/v1.5.rst
index 20184bbd2a551..059875eec12d6 100644
--- a/doc/whats_new/v1.5.rst
+++ b/doc/whats_new/v1.5.rst
@@ -13,6 +13,23 @@ For a short description of the main highlights of the release, please refer to
.. include:: changelog_legend.inc
+.. _changes_1_5_2:
+
... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-773@529fc8c | reflex-dev/reflex | Python | 773 | Fix Event chaining in the on_load event handler return not working | This PR fixes a couple of issues with the way on-load events are handled. In this PR, the following should work now:
- event-chaining( handlers that return other handlers)
- list of on-load events (eg. `app.add_page('index', on_load = [State.first_handler, State.second_handler]`)
- Async handlers
fixes #610 , #... | 2023-04-04T15:55:48Z | Event chaining in the on_load event handler return is not working.
**Describe the bug**
A clear and concise description of what the bug is.
Event chaining in the on_load event handler return is not working.
**To Reproduce**
Steps to reproduce the behavior:
- Code/Link to Repo:
I implemented it like this
... | Currently event chains don't work in `on_load` since `on_load` events are handled in [HydrateMiddleware](https://github.com/pynecone-io/pynecone/blob/387bacff734078267183e72d05496fc3b271efc8/pynecone/middleware/hydrate_middleware.py).
A different method for handling `on_load` events has been proposed to handle chai... | [
{
"body": "**Describe the bug**\r\nA clear and concise description of what the bug is.\r\n\r\nEvent chaining in the on_load event handler return is not working.\r\n\r\n**To Reproduce**\r\nSteps to reproduce the behavior:\r\n - Code/Link to Repo:\r\n\r\nI implemented it like this\r\n\r\n```python\r\nclass BaseSt... | bd6ea9d977427c2590343ec1438096a87ac60e3c | {
"head_commit": "529fc8c41f18923c4ea1d4616f8d7dd5cecff97c",
"head_commit_message": "added async tests",
"patch_to_review": "diff --git a/pynecone/app.py b/pynecone/app.py\nindex 8c359e30da1..0a3ef91d49d 100644\n--- a/pynecone/app.py\n+++ b/pynecone/app.py\n@@ -148,7 +148,7 @@ def add_cors(self):\n al... | [
{
"diff_hunk": "@@ -37,20 +39,42 @@ def preprocess(self, app: App, state: State, event: Event) -> Optional[Delta]:\n load_event = None\n \n if load_event:\n- if isinstance(load_event, list):\n- for single_event in load_event:\n- ... | e69e819e3580bcf0f5a5ac3f6230ca8b3face9c0 | diff --git a/pynecone/app.py b/pynecone/app.py
index 8c359e30da1..c4868e8d9e9 100644
--- a/pynecone/app.py
+++ b/pynecone/app.py
@@ -148,7 +148,9 @@ def add_cors(self):
allow_origins=["*"],
)
- def preprocess(self, state: State, event: Event) -> Optional[Delta]:
+ async def preprocess(
+ ... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-915@997af8f | reflex-dev/reflex | Python | 915 | [Fix 477] Use jinja2 for templating | ### All Submissions:
- [x] Have you followed the guidelines stated in [CONTRIBUTING.md](https://github.com/pynecone-io/pynecone/blob/main/CONTRIBUTING.md) file?
- [x] Have you checked to ensure there aren't any other open [Pull Requests](https://github.com/pynecone-io/pynecone/pulls ) for the desired changed?
<!... | 2023-04-30T11:53:15Z | Use Jinja for templating
Currently all the JS templating is done manually in Python and f-string formatting. It would be nice to use something like Jinja that was designed for this very task. As the templates get bigger, this will help us stay more understandable and maintainable. | Waiting for a follow up PR to fix the build, then we will merge this | [
{
"body": "Currently all the JS templating is done manually in Python and f-string formatting. It would be nice to use something like Jinja that was designed for this very task. As the templates get bigger, this will help us stay more understandable and maintainable. ",
"number": 477,
"title": "Use Jinj... | dc2dff9323a7013824c3a73dfea02578b0ecea69 | {
"head_commit": "997af8fff71eb283dc08b0fc7670f2052746bd6b",
"head_commit_message": "fix test fails of python 3.8",
"patch_to_review": "diff --git a/poetry.lock b/poetry.lock\nindex 2012042a11d..6839da93192 100644\n--- a/poetry.lock\n+++ b/poetry.lock\n@@ -1,4 +1,4 @@\n-# This file is automatically @generated by ... | [
{
"diff_hunk": "@@ -117,55 +119,22 @@ def format_prop(\n assert isinstance(prop, str), \"The prop must be a string.\"\n return format.wrap(prop, \"{\", check_first=False)\n \n- def format_props(self) -> str:\n+ def format_props(self) -> list:",
"line": null,
"original_line": 122,
... | 631a4447a034de19681614875edaa55ff714d49a | diff --git a/poetry.lock b/poetry.lock
index 55cd70c4982..efdf81a702e 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -463,6 +463,84 @@ files = [
{file = "iniconfig-2.0.0.tar.gz", hash = "sha256:2d91e135bf72d31a410b17c16da610a82cb55f6b0477d1a902134b24a455b8b3"},
]
+[[package]]
+name = "jinja2"
+version = "3.1.2"
... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} |
reflex-dev__reflex-550@47a38ed | reflex-dev/reflex | Python | 550 | Anonymous Telemetry | ### All Submissions:
- [ ] Have you followed the guidelines stated in [CONTRIBUTING.md](https://github.com/pynecone-io/pynecone/blob/main/CONTRIBUTING.md) file?
- [ ] Have you checked to ensure there aren't any other open [Pull Requests](https://github.com/pynecone-io/pynecone/pulls ) for the desired changed?
<!... | 2023-02-16T06:59:35Z | Add anonymous telemetry
Currently it's hard to measure how much Pynecone is being used once it's installed. We want to measure some very basic usage statistics.
Key Measurements to Log:
* Operating System
* Python Version
* Node Version
* Pynecone Version
* CPU / Memory
Key Times to Log:
* pc init (to under... | [
{
"body": "Currently it's hard to measure how much Pynecone is being used once it's installed. We want to measure some very basic usage statistics.\r\n\r\nKey Measurements to Log:\r\n* Operating System\r\n* Python Version\r\n* Node Version\r\n* Pynecone Version\r\n* CPU / Memory\r\n\r\nKey Times to Log:\r\n* pc... | d0b47e1c2329ef134ac7c82d5ff8d98b73cef73d | {
"head_commit": "47a38ede12f9220854ad5606d8df0c4fe933df3b",
"head_commit_message": "Ruff",
"patch_to_review": "diff --git a/poetry.lock b/poetry.lock\nindex 7471e7e4751..20aadc3dff7 100644\n--- a/poetry.lock\n+++ b/poetry.lock\n@@ -122,6 +122,104 @@ files = [\n {file = \"certifi-2022.12.7.tar.gz\", hash = \"... | [
{
"diff_hunk": "@@ -100,6 +107,9 @@ def run(\n frontend_cmd, backend_cmd = utils.run_frontend_prod, utils.run_backend_prod\n assert frontend_cmd and backend_cmd, \"Invalid env\"\n \n+ # Post a telemetry event.\n+ pynecone_telemetry(f\"run-{env}\", utils.get_config().telemetry_enabled)",
"l... | 81175ee5010ce0faf44cd0bbecaba74fe03482a3 | diff --git a/poetry.lock b/poetry.lock
index 7471e7e4751..a9e23043e5a 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -628,14 +628,14 @@ plugins = ["importlib-metadata"]
[[package]]
name = "pyright"
-version = "1.1.293"
+version = "1.1.294"
description = "Command line wrapper for pyright"
category = "dev"
optional... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} | |
reflex-dev__reflex-513@a1e427a | reflex-dev/reflex | Python | 513 | add pc.route decorator | ### All Submissions:
- [x] Have you followed the guidelines stated in [CONTRIBUTING.md](https://github.com/pynecone-io/pynecone/blob/main/CONTRIBUTING.md) file?
- [x] Have you checked to ensure there aren't any other open [Pull Requests](https://github.com/pynecone-io/pynecone/pulls ) for the desired changed?
##... | 2023-02-11T23:48:08Z | Add wrapper for adding a page
Adding a function wrapper for adding a page is more pythonic and would be cleaner instead of dumping all your add_page's at the bottom.
Also don't remove the actual add_page function, just add an alternate wrapper method. | [
{
"body": "Adding a function wrapper for adding a page is more pythonic and would be cleaner instead of dumping all your add_page's at the bottom.\r\nAlso don't remove the actual add_page function, just add an alternate wrapper method.",
"number": 462,
"title": "Add wrapper for adding a page"
}
] | d3c3174ba4139bcbbecfaf9f3f675fcf9ecd59cd | {
"head_commit": "a1e427a06b5566d5f7167baa9547b2de1e798db1",
"head_commit_message": "remove unused import",
"patch_to_review": "diff --git a/pynecone/__init__.py b/pynecone/__init__.py\nindex b0fcb441f5a..56e66b33f52 100644\n--- a/pynecone/__init__.py\n+++ b/pynecone/__init__.py\n@@ -13,6 +13,7 @@\n from .event i... | [
{
"diff_hunk": "@@ -0,0 +1,42 @@\n+\"\"\"The route decorator and associated variables.\"\"\"\n+\n+DECORATED_ROUTES = []\n+\n+\n+def route(route=None, title=None, image=None, description=None, on_load=None):\n+ \"\"\"Decorate a function as a page.\n+\n+ pc.App() will automatically call add_page() for any m... | 9ee99503abaa9a205ee935467354b68748ef7a14 | diff --git a/pynecone/__init__.py b/pynecone/__init__.py
index b0fcb441f5a..56e66b33f52 100644
--- a/pynecone/__init__.py
+++ b/pynecone/__init__.py
@@ -13,6 +13,7 @@
from .event import EventChain, console_log, redirect, window_alert
from .middleware import Middleware
from .model import Model, session
+from .route i... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} | |
scikit-learn__scikit-learn-29442@3a19575 | scikit-learn/scikit-learn | Python | 29,442 | Fix elasticnect cv sample weight | Fixes #22914
What does this implement/fix? Explain your changes.
Adapted from the pull request #23045 by s-banach and #29308 from snath-xoc
Modifies _alpha_grid function in linear_model._coordinate_descent to accept a sample_weight argument and implements changes to be compatible with _preprocess_data
TODO
... | 2024-07-09T20:09:54Z | Calculation of alphas in ElasticNetCV doesn't use sample_weight
### Describe the bug
In ElasticNetCV, the first and largest value of `alpha`, call it `alpha_max`, should be just large enough to force all of the coefficients to become zero. The existing code works correctly when `sample_weight` is not specified. Howe... | I agree, the max `alpha` should give zero coefficients.
Do you want to give it a try and open a PR?
The *change policy* might be tricky as it corresponds to changing a model (once fixed, `ElasticNetCV.fit` will give different results than before) and there is no meaningful deprecation path, IMO.
I consider it a bu... | [
{
"body": "### Describe the bug\r\n\r\nIn ElasticNetCV, the first and largest value of `alpha`, call it `alpha_max`, should be just large enough to force all of the coefficients to become zero. The existing code works correctly when `sample_weight` is not specified. However, the computation of `alpha_max` does ... | f9776619f1144075289813634c5d450d5b31fa71 | {
"head_commit": "3a1957589f7dd56713ea9040b70228124dbd5fc7",
"head_commit_message": "add global_random_seed",
"patch_to_review": "diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst\nindex c98314d5ca1de..0153de68662c4 100644\n--- a/doc/whats_new/v1.6.rst\n+++ b/doc/whats_new/v1.6.rst\n@@ -103,6 +103,12 @... | [
{
"diff_hunk": "@@ -1306,55 +1307,60 @@ def test_enet_sample_weight_consistency(\n \n @pytest.mark.parametrize(\"fit_intercept\", [True, False])\n @pytest.mark.parametrize(\"sparse_container\", [None] + CSC_CONTAINERS)\n-def test_enet_cv_sample_weight_correctness(fit_intercept, sparse_container):\n+def test_ene... | 080c8469d2c14766b739da4bcdfd26eddd611c25 | diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst
index a45765cbd04a3..5d56e534364fd 100644
--- a/doc/whats_new/v1.6.rst
+++ b/doc/whats_new/v1.6.rst
@@ -233,6 +233,11 @@ Changelog
has no effect. `copy_X` will be removed in 1.8.
:pr:`29105` by :user:`Adam Li <adam2392>`.
+- |Fix| :class:`linear_model... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-503@f84b37b | reflex-dev/reflex | Python | 503 | 334: create PCDict | This PR adds a PCDict feature to handle mutable dicts
### All Submissions:
- [x] Have you followed the guidelines stated in [CONTRIBUTING.md](https://github.com/pynecone-io/pynecone/blob/main/CONTRIBUTING.md) file?
- [x] Have you checked to ensure there aren't any other open [Pull Requests](https://github.com/py... | 2023-02-11T03:15:58Z | Create pc.dict and pc.list
Currently Pynecone doesn't detect mutations in list/dict state vars which is annoying and can lead to bugs if one is not careful.
Maybe we could create classes `pc.dict` and `pc.list` that wrap any dict/list State vars, and modify them such that they detect mutations. | I can work on this. Can you provide an example how `pc.dict` & `pc.list` would look like? My guess would be sth like this:
```py
def index():
colors = pc.list(['red', 'oragen'])
return pc.vstack(
pc.text('Number of colors:' + len(colors)),
pc.button('+', on_click=lambda : colors.append('another colo... | [
{
"body": "Currently Pynecone doesn't detect mutations in list/dict state vars which is annoying and can lead to bugs if one is not careful.\r\n\r\nMaybe we could create classes `pc.dict` and `pc.list` that wrap any dict/list State vars, and modify them such that they detect mutations.",
"number": 334,
... | 5e0d5b9e2e2aec9bdf19dd75629bfcef26a33f6a | {
"head_commit": "f84b37b89044689993473d44ea6fdc441783aea7",
"head_commit_message": "334: create PCDict\n\nThis PR adds a PCDict feature to handle mutable dicts",
"patch_to_review": "diff --git a/pynecone/state.py b/pynecone/state.py\nindex e5c452e0585..6841bf40cd8 100644\n--- a/pynecone/state.py\n+++ b/pynecone/... | [
{
"diff_hunk": "@@ -79,7 +79,9 @@ def _init_mutable_fields(self):\n value, self._reassign_field, field.name\n )\n \n- if utils._issubclass(field.type_, List):\n+ if utils._issubclass(field.type_, List) or utils._issubclass(",
"line": null,
"original_line... | db42280fc088673c21bed6c28a99b061521a7df9 | diff --git a/pynecone/state.py b/pynecone/state.py
index e5c452e0585..0417b73806b 100644
--- a/pynecone/state.py
+++ b/pynecone/state.py
@@ -5,7 +5,18 @@
import functools
import traceback
from abc import ABC
-from typing import Any, Callable, ClassVar, Dict, List, Optional, Sequence, Set, Type
+from typing import (
... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-29243@4cc3995 | scikit-learn/scikit-learn | Python | 29,243 | DOC add example with ``set_params``. | #### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Atte... | 2024-06-12T19:17:59Z | When a Pipeline step is changed via set_params, the set_output state is cleared
### Describe the bug
When a Pipeline step is set via `set_params`, the subsequent output of `fit_transform` is a numpy ndarray even if previously the pipeline's output was set to be of type pandas.DataFrame via a call to `set_output(tran... | I think this happens because `pipe.set_output` will call `.set_output` to all its underlying estimators in particular `pipe['scale']` i.e. its `MinMaxScaler` estimator. When you do `pipe.set_params(scale=MinMaxScaler())` you replace `pipe['scale']` with an estimator that does not have `set_output` set.
The easiest w... | [
{
"body": "### Describe the bug\r\n\r\nWhen a Pipeline step is set via `set_params`, the subsequent output of `fit_transform` is a numpy ndarray even if previously the pipeline's output was set to be of type pandas.DataFrame via a call to `set_output(transform= 'pandas')`.\r\n\r\nThis only happens when the enti... | 7e8ad632ff7aec09e99f0c4b8e493e13e19aef14 | {
"head_commit": "4cc39952bc6a5c4512489b40ac9bb7700069727e",
"head_commit_message": "DOC Linting",
"patch_to_review": "diff --git a/examples/miscellaneous/plot_set_output.py b/examples/miscellaneous/plot_set_output.py\nindex 9baa71a1b3648..2e8a65cb82d6e 100644\n--- a/examples/miscellaneous/plot_set_output.py\n+++... | [
{
"diff_hunk": "@@ -63,6 +63,12 @@\n # means that the final logistic regression step contains the feature names of the input.\n clf[-1].feature_names_in_\n \n+# %%\n+# As a side note, if one uses the method `set_params`, the transformer will be\n+# replaced by a new one which does not have `set_output` set:\n+c... | 0d2cbae569b189655f691bae3aaafa74d67f318e | diff --git a/examples/miscellaneous/plot_set_output.py b/examples/miscellaneous/plot_set_output.py
index 9baa71a1b3648..e74d94957c685 100644
--- a/examples/miscellaneous/plot_set_output.py
+++ b/examples/miscellaneous/plot_set_output.py
@@ -63,6 +63,21 @@
# means that the final logistic regression step contains the fe... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Bug Fixes"
} |
reflex-dev__reflex-339@7ed7ca5 | reflex-dev/reflex | Python | 339 | Add and use `pc.list` to have list mutation detection | To add and use `pc.list`/`PcList` to have list mutation detection.
It's not as straightforward as I thought. Or is my approach too complicated?
For #334. `pc.dict` would be in another PR. | 2023-01-25T15:21:22Z | Create pc.dict and pc.list
Currently Pynecone doesn't detect mutations in list/dict state vars which is annoying and can lead to bugs if one is not careful.
Maybe we could create classes `pc.dict` and `pc.list` that wrap any dict/list State vars, and modify them such that they detect mutations. | I can work on this. Can you provide an example how `pc.dict` & `pc.list` would look like? My guess would be sth like this:
```py
def index():
colors = pc.list(['red', 'oragen'])
return pc.vstack(
pc.text('Number of colors:' + len(colors)),
pc.button('+', on_click=lambda : colors.append('another colo... | [
{
"body": "Currently Pynecone doesn't detect mutations in list/dict state vars which is annoying and can lead to bugs if one is not careful.\r\n\r\nMaybe we could create classes `pc.dict` and `pc.list` that wrap any dict/list State vars, and modify them such that they detect mutations.",
"number": 334,
... | bfe9ad807e29ada85486001efcae1e605630fb76 | {
"head_commit": "7ed7ca5ea311936d5946a08e00bfe4e7f3de6cc9",
"head_commit_message": "simply PcList",
"patch_to_review": "diff --git a/pynecone/state.py b/pynecone/state.py\nindex 0957001285d..378bc1696df 100644\n--- a/pynecone/state.py\n+++ b/pynecone/state.py\n@@ -13,7 +13,7 @@\n from pynecone import constants, ... | [
{
"diff_hunk": "@@ -748,3 +748,54 @@ def type_(self):\n if \"return\" in self.fget.__annotations__:\n return self.fget.__annotations__[\"return\"]\n return Any\n+\n+\n+class PcList(list):",
"line": null,
"original_line": 753,
"original_start_line": null,
"path": "pyne... | fd7c722759bc0e45631f10030e5446935b2f5df3 | diff --git a/pynecone/state.py b/pynecone/state.py
index 0957001285d..a5901e9f1fd 100644
--- a/pynecone/state.py
+++ b/pynecone/state.py
@@ -13,7 +13,7 @@
from pynecone import constants, utils
from pynecone.base import Base
from pynecone.event import Event, EventHandler, window_alert
-from pynecone.var import BaseVa... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} |
reflex-dev__reflex-323@f8afb87 | reflex-dev/reflex | Python | 323 | Resolve #304: Relax type requirement on pc.cond | Resolve #304: Relax type requirement on pc.cond | 2023-01-23T19:11:43Z | Relax type requirement on pc.cond
Currently the first argument to `pc.cond` must be a `pc.Var` with type `bool`. However since JS is lax about converting anything to a bool, perhaps we can relax this requirement and just allow anything in.
There's a tradeoff here when we do this stuff implicitly as it may lead to un... | [
{
"body": "Currently the first argument to `pc.cond` must be a `pc.Var` with type `bool`. However since JS is lax about converting anything to a bool, perhaps we can relax this requirement and just allow anything in.\r\n\r\nThere's a tradeoff here when we do this stuff implicitly as it may lead to unexpected er... | dd5fa88050aa9d3158b6a3185e07fbdc616c951a | {
"head_commit": "f8afb871e8802727c30b5e551a809e8b6499b7ba",
"head_commit_message": "unit test for pc.cond validation",
"patch_to_review": "diff --git a/pynecone/components/layout/cond.py b/pynecone/components/layout/cond.py\nindex 8146e2d5a33..d80f306f253 100644\n--- a/pynecone/components/layout/cond.py\n+++ b/p... | [
{
"diff_hunk": "@@ -0,0 +1,37 @@\n+import pytest\n+\n+import pynecone as pc\n+\n+\n+@pytest.fixture\n+def cond_state(request):\n+ class CondState(pc.State):\n+ value: request.param[\"value_type\"] = request.param[\"value\"]\n+\n+ return CondState\n+\n+\n+@pytest.mark.parametrize(\n+ \"cond_state... | 015b5a050a11388cea2cd0560831d2b1fe8b1890 | diff --git a/pynecone/components/layout/cond.py b/pynecone/components/layout/cond.py
index 8146e2d5a33..d80f306f253 100644
--- a/pynecone/components/layout/cond.py
+++ b/pynecone/components/layout/cond.py
@@ -1,9 +1,7 @@
"""Create a list of components from an iterable."""
from __future__ import annotations
-from ty... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} | |
reflex-dev__reflex-269@697d86c | reflex-dev/reflex | Python | 269 | Empty .web/pages directory on recompile | Empty the .web/pages directory on recompile so that old pages that don't exist anymore in the python code get removed. | 2023-01-16T20:32:52Z | Old pages aren't deleted on recompile
**Describe the bug**
Old pages stay in the web directory
**To Reproduce**
Steps to reproduce the behavior:
* compile the app with pages ["page1", "page2"]
* compile the app with just ["page1"]
* the file `/.web/page2` still exists
**Expected behavior**
Only the specifie... | [
{
"body": "**Describe the bug**\r\nOld pages stay in the web directory\r\n\r\n**To Reproduce**\r\nSteps to reproduce the behavior:\r\n* compile the app with pages [\"page1\", \"page2\"]\r\n* compile the app with just [\"page1\"]\r\n* the file `/.web/page2` still exists\r\n\r\n**Expected behavior**\r\nOnly the s... | 8858754d409f597e59f19841487ffe85881f162c | {
"head_commit": "697d86cde93172269dc0e3f9455e5536c94687c5",
"head_commit_message": "added ignored files 404.js and _app.js when .web/pages is purged",
"patch_to_review": "diff --git a/pynecone/app.py b/pynecone/app.py\nindex d6a7c4f74bf..2df8b13bc26 100644\n--- a/pynecone/app.py\n+++ b/pynecone/app.py\n@@ -284,6... | [
{
"diff_hunk": "@@ -208,3 +208,10 @@ def compile_components(components: Set[CustomComponent]):\n # Compile the components.\n code = _compile_components(components)\n return output_path, code\n+\n+\n+def purge_web_pages_dir():\n+ \"\"\"Empty out .web directory.\n+ \"\"\"\n+ template_files = ... | 1c67ab0b44afd79caa4b51eb5685d8e9e5cb7b7c | diff --git a/pynecone/app.py b/pynecone/app.py
index d6a7c4f74bf..3979d246fc3 100644
--- a/pynecone/app.py
+++ b/pynecone/app.py
@@ -284,6 +284,9 @@ def compile(self, force_compile: bool = False):
# Create the database models.
Model.create_all()
+ # Empty the .web pages directory
+ com... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} | |
reflex-dev__reflex-228@bb5c9f0 | reflex-dev/reflex | Python | 228 | Add path event | Closes #138
I needed this feature so I figured I would try to add it, turn out it didn't require too much effort.
Notes:
- There is a modification in .templates/web/utils/state.js , if I'm getting it right, this file is copied during `pc init`, so I'm not sure how to update it for existing project? I haven't see... | 2023-01-08T19:50:40Z | Access current page within state
Currently within the state event handlers (and computed var functions), we don't have access to the current page. We can add a method to the `pc.State` class called `get_current_page` that returns this info (potentially along with route params, query params, etc.) | [
{
"body": "Currently within the state event handlers (and computed var functions), we don't have access to the current page. We can add a method to the `pc.State` class called `get_current_page` that returns this info (potentially along with route params, query params, etc.)",
"number": 138,
"title": "A... | f583c4f9421e909e9c2ed5dc6440ad8f12d65580 | {
"head_commit": "bb5c9f04ac09f00e7c354daefb699c588ce14f68",
"head_commit_message": "switch to using get_current_page() method instead",
"patch_to_review": "diff --git a/pynecone/.templates/web/utils/state.js b/pynecone/.templates/web/utils/state.js\nindex b0612df83e3..256f07e79ad 100644\n--- a/pynecone/.template... | [
{
"diff_hunk": "@@ -70,6 +70,9 @@ export const applyDelta = (state, delta) => {\n */\n export const applyEvent = async (event, router, socket) => {\n // Handle special events\n+\n+ event.router_data = router;",
"line": null,
"original_line": 74,
"original_start_line": null,
"path": "pynecone... | 619a9a3a2c508331c2e7a84c4444221a4ec6c1ad | diff --git a/pynecone/.templates/web/utils/state.js b/pynecone/.templates/web/utils/state.js
index b0612df83e3..b81e5e674ea 100644
--- a/pynecone/.templates/web/utils/state.js
+++ b/pynecone/.templates/web/utils/state.js
@@ -70,6 +70,9 @@ export const applyDelta = (state, delta) => {
*/
export const applyEvent = asy... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} | |
scikit-learn__scikit-learn-29312@78cc39f | scikit-learn/scikit-learn | Python | 29,312 | FEA Add metadata routing for RFE and RFECV | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2024-06-20T11:52:44Z | RFE/RFECV doesn't work with sample weights
As far as I can tell, `sklearn.feature_selection.RFE` has no way to pass sample weights to the estimator alongside the data.
I have fixed this in my code with:
``` diff
index bbe0cda..f5072b2 100644
--- a/sklearn/feature_selection/rfe.py
+++ b/sklearn/feature_selectio... | yeah I think you can open a PR for that. You need to add tests, though.
Note that testing for sample weights usually involves checking that weights
correspond to repeating samples.
On 1 September 2016 at 03:33, Andreas Mueller notifications@github.com
wrote:
> yeah I think you can open a PR for that. You need to add... | [
{
"body": "As far as I can tell, `sklearn.feature_selection.RFE` has no way to pass sample weights to the estimator alongside the data.\r\n\r\nI have fixed this in my code with:\r\n\r\n``` diff\r\nindex bbe0cda..f5072b2 100644\r\n--- a/sklearn/feature_selection/rfe.py\r\n+++ b/sklearn/feature_selection/rfe.py\r... | 5600faab452b0adb1f4090219012aa3ae6208c66 | {
"head_commit": "78cc39f92ffe3e2a9056d1a2f0ceaddc17c49d2e",
"head_commit_message": "Add _deprecate_positional_args for groups",
"patch_to_review": "diff --git a/doc/metadata_routing.rst b/doc/metadata_routing.rst\nindex 31dae6813bda5..a753269e1903a 100644\n--- a/doc/metadata_routing.rst\n+++ b/doc/metadata_routi... | [
{
"diff_hunk": "@@ -811,3 +925,42 @@ def fit(self, X, y, groups=None):\n \"n_features\": step_n_features_rev,\n }\n return self\n+\n+ def get_metadata_routing(self):\n+ \"\"\"Get metadata routing of this object.\n+\n+ Please check :ref:`User Guide <metadata_routing>`... | 2b860e213176f1c1badece8cee62115f709d2828 | diff --git a/doc/metadata_routing.rst b/doc/metadata_routing.rst
index fec2cf610c02f..9b21b74032562 100644
--- a/doc/metadata_routing.rst
+++ b/doc/metadata_routing.rst
@@ -284,6 +284,8 @@ Meta-estimators and functions supporting metadata routing:
- :class:`sklearn.ensemble.VotingRegressor`
- :class:`sklearn.ensemble... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-28860@6579b83 | scikit-learn/scikit-learn | Python | 28,860 | fixed error that prevented parametrizing custom estimators that implements `__call__` | #### Reference Issues/PRs
Fixes #28859
#### What does this implement/fix? Explain your changes.
Replaced `if callable(obj)` with two separate checks `if isisfunction(obj)` and `if isinstance(obj, partial)`.
| 2024-04-19T10:10:37Z | `parametrize_with_checks` fails if custom estimator implements `__call__`
### Describe the bug
Title.
### Steps/Code to Reproduce
```python
from sklearn.utils.estimator_checks import parametrize_with_checks
class MyEstimator:
"""Dummy estimator."""
def get_params(self, *, deep=True):
retur... | [
{
"body": "### Describe the bug\n\nTitle.\n\n### Steps/Code to Reproduce\n\n```python\r\nfrom sklearn.utils.estimator_checks import parametrize_with_checks\r\n\r\n\r\nclass MyEstimator:\r\n \"\"\"Dummy estimator.\"\"\"\r\n\r\n def get_params(self, *, deep=True):\r\n return {}\r\n\r\n def set_par... | 05ee9dd4b5d925c6172d2f21c095051587a84083 | {
"head_commit": "6579b83ee0b5ad40133478bdb7dc83b8106a86dd",
"head_commit_message": "fixed error that prevented parametrizing custom estimators that implement __call__",
"patch_to_review": "diff --git a/sklearn/tests/test_common.py b/sklearn/tests/test_common.py\nindex 4564eddee410c..ab9be06f4b579 100644\n--- a/s... | [
{
"diff_hunk": "@@ -103,6 +103,31 @@ def _sample_func(x, y=1):\n pass\n \n \n+class EstimatorStub:",
"line": null,
"original_line": 106,
"original_start_line": null,
"path": "sklearn/tests/test_common.py",
"start_line": null,
"text": "@user1:\n```suggestion\r\nclass CallableEstimator... | f0514afda3934be2181b7945ccb18978ac2e3aa7 | diff --git a/sklearn/tests/test_common.py b/sklearn/tests/test_common.py
index 4564eddee410c..9ff83953f4b0e 100644
--- a/sklearn/tests/test_common.py
+++ b/sklearn/tests/test_common.py
@@ -20,6 +20,7 @@
import pytest
import sklearn
+from sklearn.base import BaseEstimator
from sklearn.cluster import (
OPTICS,
... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} | |
scikit-learn__scikit-learn-28764@cab58f3 | scikit-learn/scikit-learn | Python | 28,764 | RFE and RFECV allow features_to_select to be larger than available | fixes #28725
Implemented a warning message for when the RFE or the RFECV objects are initialized with a n_features_to_select or min_features_to_select (respectively) attribute larger than the number of features present in the X variable that is passed to the fit method.
Additionally, a new test function has been ... | 2024-04-03T17:06:03Z | RFE and RFECV allow features_to_select to be larger than available features
### Describe the bug
If the `RFE` or the `RFECV` objects are initialized with a `n_features_to_select` or a `min_features_to_select` (respectively) attribute larger than the number of features present in the `X` variable that is passed to th... | Hi @miguelcsilva, I agree that the behavior is unclear. I'm not sure if we want an error or a warning. I sligthly tend towards a warning. We also need to explain the behavior in the parameter descriptions.
ping @glemaitre, wdyt ?
Can I work on this issue? @ogrisel @jeremiedbb
> Can I work on this issue? @ogrisel @j... | [
{
"body": "### Describe the bug\r\n\r\nIf the `RFE` or the `RFECV` objects are initialized with a `n_features_to_select` or a `min_features_to_select` (respectively) attribute larger than the number of features present in the `X` variable that is passed to the `fit` method, I would expect an error to be raised.... | beb71774647fe1e7154c546a69c7ce412e5a2027 | {
"head_commit": "cab58f3dbaef17ae52f2d417c2fd825718827044",
"head_commit_message": "RFE and RFECV allow features_to_select to be larger than available features (scikit-learn#28725)\n\nImplemented a warning message for when the RFE or the RFECV objects are initialized with a n_features_to_select or min_features_to_... | [
{
"diff_hunk": "@@ -289,6 +290,15 @@ def _fit(self, X, y, step_score=None, **fit_params):\n else: # float\n n_features_to_select = int(n_features * self.n_features_to_select)\n \n+ if n_features_to_select > n_features:\n+ warnings.warn(\n+ (\n+ ... | 81992c672d893e509f3f4704a99de4e0d3e219f7 | diff --git a/sklearn/feature_selection/_rfe.py b/sklearn/feature_selection/_rfe.py
index 44764655e988d..7c5cd8d45b8d1 100644
--- a/sklearn/feature_selection/_rfe.py
+++ b/sklearn/feature_selection/_rfe.py
@@ -6,6 +6,7 @@
"""Recursive feature elimination for feature ranking"""
+import warnings
from numbers import ... | {
"difficulty": "low",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-28564@e23af64 | scikit-learn/scikit-learn | Python | 28,564 | DOC add example docstring for make_low_rank_matrix | closes #27982
This the remaining docstring that needs to be addressed in #27982.
We can close the issue then. | 2024-03-02T14:41:49Z | Ensure that we have an example in the docstring of each public function or class
We should make sure that we have a small example for all public functions or classes. Most of the missing examples are linked to functions.
I could list the following classes and functions for which `numpydoc` did not find any example:
... | I am new to GitHub, and I want to work on this!
Hi @AdarshWase, to start working on this, you can fork the scikit-learn repository, clone it to your local machine, make your changes, and then send a pull request. The pull request should mention this issue (#27982).
I've already send a PR for the clustering functions... | [
{
"body": "We should make sure that we have a small example for all public functions or classes. Most of the missing examples are linked to functions.\r\n\r\nI could list the following classes and functions for which `numpydoc` did not find any example:\r\n\r\n- [x] sklearn.base.BaseEstimator\r\n- [x] sklearn.b... | d967cfe8124902181892411b18b50dce9921a32d | {
"head_commit": "e23af648573d69ade443be71a2a903757b6f3fe1",
"head_commit_message": "DOC add example docstring for make_low_rank_matrix",
"patch_to_review": "diff --git a/sklearn/datasets/_samples_generator.py b/sklearn/datasets/_samples_generator.py\nindex 9d689da99fa45..527c87641403d 100644\n--- a/sklearn/datas... | [
{
"diff_hunk": "@@ -1411,6 +1411,22 @@ def make_low_rank_matrix(\n -------\n X : ndarray of shape (n_samples, n_features)\n The matrix.\n+\n+ Examples\n+ --------\n+ >>> from numpy.linalg import svd\n+ >>> from sklearn.datasets import make_low_rank_matrix\n+ >>> X = make_low_rank_... | 55b773d034c87d6a5a75391423e0e05ec7033cb7 | diff --git a/sklearn/datasets/_samples_generator.py b/sklearn/datasets/_samples_generator.py
index 9d689da99fa45..21326bc3c630b 100644
--- a/sklearn/datasets/_samples_generator.py
+++ b/sklearn/datasets/_samples_generator.py
@@ -1411,6 +1411,20 @@ def make_low_rank_matrix(
-------
X : ndarray of shape (n_samp... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Documentation Updates"
} |
scikit-learn__scikit-learn-28365@6ad9733 | scikit-learn/scikit-learn | Python | 28,365 | FIX handle inconsistence between `fill_value` and `X` dtype in `SimpleImputer` | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2024-02-05T18:50:18Z | SimpleImputer silently cast fill values to integer when the input is of integer type
### Describe the bug
Fitting the SimpleImputer on an integer array silently cast the float `fill_value` values to integer. If `fill_value` is nan, nothing is imputed but a warning is raise:
`RuntimeWarning: invalid value encountered ... | Indeed this is bad. We probably introduced the bug here: https://github.com/scikit-learn/scikit-learn/pull/22063
I assume that we need to make a casting rule between the dtype of `X` seen at `fit` and the `fill_value` parameter.
@LeoGrin do you fill like making a fix. Basically we need a the casting rule to apply... | [
{
"body": "### Describe the bug\n\nFitting the SimpleImputer on an integer array silently cast the float `fill_value` values to integer. If `fill_value` is nan, nothing is imputed but a warning is raise:\r\n`RuntimeWarning: invalid value encountered in cast\r\n multiarray.copyto(a, fill_value, casting='unsafe'... | be35d8c58a1797c7243c25977d22d355dfdf8f1c | {
"head_commit": "6ad97336bffd2a53676a500193f9536bee0a3745",
"head_commit_message": "update changelog",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex 2baa45e88a115..5fa6b934f6f3e 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -152,6 +152,14 @@ Chan... | [
{
"diff_hunk": "@@ -152,6 +152,14 @@ Changelog\n works with pandas copy-on-write mode enabled.\n :pr:`28233` by :user:`John Hopfensperger <s-banach>`.\n \n+\n+:mod:`sklearn.impute`",
"line": null,
"original_line": 156,
"original_start_line": null,
"path": "doc/whats_new/v1.4.rst",
"start... | 92a1371b8ea3b3f7e7ecff78475d26f8582a45c1 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index df18adf3401ec..11051c727c239 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -115,6 +115,14 @@ Changelog
target is multilabel or multiclass-multioutput in a DataFrame format.
:pr:`27702` by :user:`Guillaume Lemaitre <glemaitre>... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-28165@1662a8f | scikit-learn/scikit-learn | Python | 28,165 | Fix prevent infinite loop in KMeans | Fixes https://github.com/scikit-learn/scikit-learn/issues/28055
When there are more centers than non-duplicate samples, the relocation of empty cluster will trigger at every iteration preventing the algorithm to stop. | 2024-01-18T11:17:24Z | Infinite Loop in K-means when relocating empty clusters
### Describe the bug
Relocating empty clusters in Kmeans is not working as expected in this edge case, where :
- There is duplicate entries.
- The number of clusters is equal to the number of entries.
- Very particular initial positions.
Kmeans is stuck i... | > Kmeans is stuck in a infinite loop, and the only way to end it is the max number of iterations.
Hi @brmel, I used the code example you provided, except I omitted the `max_iterations` parameter:
```
import numpy as np
from sklearn.cluster import KMeans
X = np.array([[100, 100], [100, 102], [106, 100], [107,... | [
{
"body": "### Describe the bug\r\n\r\nRelocating empty clusters in Kmeans is not working as expected in this edge case, where :\r\n- There is duplicate entries.\r\n- The number of clusters is equal to the number of entries.\r\n- Very particular initial positions.\r\n\r\nKmeans is stuck in a infinite loop, and ... | d418e79a41d5d4de167114d88465a0b47b0d6a2c | {
"head_commit": "1662a8fa6cd3082539356683ed634a7c56a3849c",
"head_commit_message": "Merge branch 'fix-infinite-loop-kmeans' of github.com:jeremiedbb/scikit-learn into fix-infinite-loop-kmeans",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex c674a8619e076..fa5e6c094796c 10... | [
{
"diff_hunk": "@@ -20,6 +20,9 @@ Changelog\n :pr:`28121` by :user:`Pietro Peterlongo <pietroppeter>` and\n :user:`Yao Xiao <Charlie-XIAO>`.\n \n+- |Fix| Avoid inifinte loop in :class:`cluster.KMeans` when the number of clusters is",
"line": null,
"original_line": 23,
"original_start_line": null... | 1f3d9e3296b34af10bc5f1512be26d86813d8733 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index c674a8619e076..ee47bae7b1f5b 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -20,6 +20,9 @@ Changelog
:pr:`28121` by :user:`Pietro Peterlongo <pietroppeter>` and
:user:`Yao Xiao <Charlie-XIAO>`.
+- |Fix| Avoid infinite loop i... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-28262@d642067 | scikit-learn/scikit-learn | Python | 28,262 | FIX handle column names renaming in ColumnTransformer | closes #28260
The adapters in charge of renaming columns with `set_output` to a DataFrame-like object was failing in case that the original DataFrame-like as duplicated columns (that is possible in `pandas`.
We inverse the stacking and and renaming to avoid potential column duplicate that should not happen. | 2024-01-25T15:24:13Z | ColumnTransformer output unexpected prefixed feature names from FunctionTransformer() step
### Describe the bug
The following code demonstrates that when `FunctionTransformer` is present as a step in `ColumnTransformer`, the feature names output are all prefixed with the name from the last step '**C__**'. For examp... | I can reproduce the issue. I'll investigate the reason.
As a temporary workaround, setting `feature_names_out="one-to-one"` provide the expected results.
We indeed refactor and use a dictionary to rename the columns: https://github.com/scikit-learn/scikit-learn/blob/main/sklearn/utils/_set_output.py#L125
But we over... | [
{
"body": "### Describe the bug\r\n\r\nThe following code demonstrates that when `FunctionTransformer` is present as a step in `ColumnTransformer`, the feature names output are all prefixed with the name from the last step '**C__**'. For example, column '**x1**' is output as '**C__x1**' for 3 times. \r\nWhen '... | bb8776874410d65f510717c97b8027331bb0f3ff | {
"head_commit": "d642067be790f9f155a52b95abe9eae10f093852",
"head_commit_message": "TST check that we raise the proper error message",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex 4b5a822108f4b..855c64ba7e053 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.... | [
{
"diff_hunk": "@@ -1083,13 +1067,69 @@ def _hstack(self, Xs):\n skip_empty_columns=True,\n )\n ]\n- # Selection of columns might be empty.\n- # Hence feature names are filtered for non-emptiness.\n feature... | 4c1634594bc093b884652754d68f79b4f5af7350 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index c68c03a12ee10..f255e3abbcab8 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -44,6 +44,11 @@ Changes impacting all modules
:pr:`28235` by :user:`Julien Jerphanion <jjerphan>`.
+- |Fix| Fixes a bug for all scikit-learn transfor... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-28119@2d96398 | scikit-learn/scikit-learn | Python | 28,119 | DOC Provide info regarding impact of sample weights scale | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2024-01-13T18:35:33Z | Weighted ridge regression regularization variable is dependent on sample weight magnitude
### Describe the issue linked to the documentation
When doing weighted ridge regression, the value of the regularization parameter for a particular solution is dependent on the sample weight vector due to scaling in the implement... | It is somehow connected to https://github.com/scikit-learn/scikit-learn/issues/26848
Maybe there should be a discussion about some general accepted way of dealing with sample weights in different algorithms ( scaling them to sum up to 1, etc)
I recall some discussion here regarding the scaling (or not) of sample-we... | [
{
"body": "### Describe the issue linked to the documentation\n\nWhen doing weighted ridge regression, the value of the regularization parameter for a particular solution is dependent on the sample weight vector due to scaling in the implementation. Scaling should either be according to the weighted average in... | 8635d96e49f460cffcea82285408d297a534bacf | {
"head_commit": "2d963980c80195d308951ef77bc72ed9d17145e1",
"head_commit_message": "linting",
"patch_to_review": "diff --git a/sklearn/linear_model/_ridge.py b/sklearn/linear_model/_ridge.py\nindex e39af10053c34..2b77e45bf26f7 100644\n--- a/sklearn/linear_model/_ridge.py\n+++ b/sklearn/linear_model/_ridge.py\n@@... | [
{
"diff_hunk": "@@ -1969,7 +1969,8 @@ def fit(self, X, y, sample_weight=None):\n \n sample_weight : float or ndarray of shape (n_samples,), default=None\n Individual weights for each sample. If given a float, every sample\n- will have the same weight.\n+ will have the s... | 62a63e842d1bb366ef3f164fb989567066c3e5af | diff --git a/sklearn/linear_model/_ridge.py b/sklearn/linear_model/_ridge.py
index e39af10053c34..1fdefee5c174a 100644
--- a/sklearn/linear_model/_ridge.py
+++ b/sklearn/linear_model/_ridge.py
@@ -1969,7 +1969,9 @@ def fit(self, X, y, sample_weight=None):
sample_weight : float or ndarray of shape (n_samples,... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-28048@ba58f6f | scikit-learn/scikit-learn | Python | 28,048 | FIX more precise log loss gradient and hessian | #### Reference Issues/PRs
Fixes #28046.
#### What does this implement/fix? Explain your changes.
This PR improves gradient and hessian of `HalfBinomialLoss` thereby preventing overflow of exp(large number) resulting in inf/nan return values.
The implemented change is very carefully designed and tested for minim... | 2024-01-02T16:21:07Z | Log Loss gradient and hessian returns NaN for large negative values
### Describe the bug
The private `HalfBinomialLoss` gradient and hessian returns `np.NaN` for large negative values of `raw_prediction`:
- `gradient`
- `gradient_hessian`
Only the `loss_gradient` returns the correct gradient.
### Steps/Code to Rep... | [
{
"body": "### Describe the bug\n\nThe private `HalfBinomialLoss` gradient and hessian returns `np.NaN` for large negative values of `raw_prediction`:\r\n- `gradient`\r\n- `gradient_hessian`\r\nOnly the `loss_gradient` returns the correct gradient.\n\n### Steps/Code to Reproduce\n\n```python\r\nimport numpy as ... | ffe4e789dab9237230530d49dd210aeb84e8608d | {
"head_commit": "ba58f6f3ff2e450c00e386c60b8a2b2c298b0947",
"head_commit_message": "ENH more precise gradient and hessian for CyHalfBinomialLoss",
"patch_to_review": "diff --git a/sklearn/_loss/_loss.pyx.tp b/sklearn/_loss/_loss.pyx.tp\nindex 0ce653de84310..ff7d62cf1b53a 100644\n--- a/sklearn/_loss/_loss.pyx.tp\... | [
{
"diff_hunk": "@@ -224,48 +224,123 @@ def test_loss_boundary_y_pred(loss, y_pred_success, y_pred_fail):\n \n \n @pytest.mark.parametrize(\n- \"loss, y_true, raw_prediction, loss_true\",\n+ \"loss, y_true, raw_prediction, loss_true, gradient_true, hessian_true\",\n [\n- (HalfSquaredError(), 1.0... | 0293fad165e9c964e51789e92ddead75d84af307 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index d2de5ee433f94..a932391b732cd 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -220,6 +220,17 @@ See :ref:`array_api` for more details.
- :class:`preprocessing.MinMaxScaler` in :pr:`26243` by `Tim Head`_;
- :class:`preprocessing.Norm... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} | |
scikit-learn__scikit-learn-27898@11752ba | scikit-learn/scikit-learn | Python | 27,898 | FIX raise error in HDBSCAN with precomputed X and request centers | closes #27893
Raise a proper `ValueError` when requesting storing center and `metric="precomputed"`. | 2023-12-04T10:09:45Z | sklearn.cluster.HDBSCAN shape error when making medoids with precomputed metric
### Describe the bug
When fitting with HDBSCAN with metric="precomputed" and store_centers='medoid', it would raise the ValueError
`ValueError: Precomputed metric requires shape (n_queries, n_indexed). Got (11, 300) for 11 indexed.`
Cl... | Reading just the documentation of the private functions, it seems that `_weighted_cluster_center` expect to get the raw data and not the precomputed array. Therefore, we should raise an error much earlier that you cannot store the centroids and medoids in combination with `metric="precomputed"`. | [
{
"body": "### Describe the bug\r\n\r\nWhen fitting with HDBSCAN with metric=\"precomputed\" and store_centers='medoid', it would raise the ValueError\r\n`ValueError: Precomputed metric requires shape (n_queries, n_indexed). Got (11, 300) for 11 indexed.`\r\nClaiming the shape of input distance matrix not squar... | 0027893288dd58806c731de7efc52b6c8e6c131a | {
"head_commit": "11752badb434c9a6a04814172c54cbebe2939133",
"head_commit_message": "update changelog",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex 204cbb9e9c890..51d16ec284193 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -259,6 +259,10 @@ Chan... | [
{
"diff_hunk": "@@ -699,6 +699,11 @@ def fit(self, X, y=None):\n Returns self.\n \"\"\"\n self._validate_params()",
"line": null,
"original_line": 701,
"original_start_line": null,
"path": "sklearn/cluster/_hdbscan/hdbscan.py",
"start_line": null,
"text": "@us... | 5164e2489032a12f953558acda0939455526b00c | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index 204cbb9e9c890..51d16ec284193 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -259,6 +259,10 @@ Changelog
value instead.
:pr:`27828` by :user:`Guillaume Lemaitre <glemaitre>`.
+- |Fix| Raises a proper `ValueError` when `metric... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27817@bbb702b | scikit-learn/scikit-learn | Python | 27,817 | FIX RandomForestRegressor OOB fails with integer-values multiple targets | Fixes #27814
Allow to compute the OOB score with `RandomForestRegressor` and an integral multi-target regression. | 2023-11-20T23:15:18Z | RandomForestRegressor having problem with integer-values targets: The type of target cannot be used to compute OOB estimates
### Describe the bug
When having:
- RandomForestRegressor
- Multiple targets
- integer values only (e.g., 1.0, 2.0, 3.0, ...) in the targets
- oob_score=True
The check in `BaseForest` wil... | Indeed, we should be able to compute the OOB score. We did not anticipate that `type_of_target` will return the invalid type. You PR as the right fix since that we can leverage already the `_estimator_type`. | [
{
"body": "### Describe the bug\n\nWhen having:\r\n- RandomForestRegressor\r\n- Multiple targets\r\n- integer values only (e.g., 1.0, 2.0, 3.0, ...) in the targets\r\n- oob_score=True\r\n\r\nThe check in `BaseForest` will raise the error:\r\n\r\n```\r\nValueError: The type of target cannot be used to compute OO... | c069511e067619eec658e8070edbd51fcd4545d9 | {
"head_commit": "bbb702b82fae0ff15348b75220e5afe3608b05e5",
"head_commit_message": "add tests for all combinations of conditions",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex f3f40fd26e27a..c4f525d3d3040 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rs... | [
{
"diff_hunk": "@@ -598,29 +598,36 @@ def test_forest_oob_warning(ForestEstimator):\n \n \n @pytest.mark.parametrize(\"ForestEstimator\", FOREST_CLASSIFIERS_REGRESSORS.values())\n-@pytest.mark.parametrize(\n- \"X, y, params, err_msg\",\n- [\n- (\n- iris.data,\n- iris.target,\n... | 83a512eda8cada094cd0a25c29dedb5f74713a78 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index f3f40fd26e27a..0469e38ad1798 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -331,6 +331,7 @@ Changelog
- |API| In :class:`ensemble.AdaBoostClassifier`, the `algorithm` argument `SAMME.R` was
deprecated and will be removed in 1.6... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-28322@65392e4 | scikit-learn/scikit-learn | Python | 28,322 | Fixed TimeSeriesSplit training set size formula | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
Fixes #28321.
#### What does this implement/fix? Explain your changes.
This fixes the incorre... | 2024-01-31T14:00:21Z | TimeSeriesSplit Train Size formula correction
### Describe the issue linked to the documentation
The documentation states:
> The training set has size `i * n_samples // (n_splits + 1) + n_samples % (n_splits + 1)` in the `i`th split, with a test set of size `n_samples // (n_splits + 1)` by default, where `n_samples... | [
{
"body": "### Describe the issue linked to the documentation\r\n\r\nThe documentation states:\r\n> The training set has size `i * n_samples // (n_splits + 1) + n_samples % (n_splits + 1)` in the `i`th split, with a test set of size `n_samples // (n_splits + 1)` by default, where `n_samples` is the number of sa... | dc1cad2b3fddb8b9069d7cfd89cb1039260baf8e | {
"head_commit": "65392e40b2f1453011aa5f138deb6cafdbac738f",
"head_commit_message": "Removed redundant parentheses",
"patch_to_review": "diff --git a/sklearn/model_selection/_split.py b/sklearn/model_selection/_split.py\nindex 93898a5114191..8147f79ded361 100644\n--- a/sklearn/model_selection/_split.py\n+++ b/skl... | [
{
"diff_hunk": "@@ -1167,10 +1167,10 @@ class TimeSeriesSplit(_BaseKFold):\n \n Notes\n -----\n- The training set has size ``i * n_samples // (n_splits + 1)\n- + n_samples % (n_splits + 1)`` in the ``i`` th split,\n- with a test set of size ``n_samples//(n_splits + 1)`` by default,\n- where ... | 2c9d3bbde78f4d1577024b28094a0c56dfade505 | diff --git a/sklearn/model_selection/_split.py b/sklearn/model_selection/_split.py
index bae48e5b93d77..b2e0f64af1c9d 100644
--- a/sklearn/model_selection/_split.py
+++ b/sklearn/model_selection/_split.py
@@ -1183,7 +1183,9 @@ class TimeSeriesSplit(_BaseKFold):
The training set has size ``i * n_samples // (n_split... | {
"difficulty": "low",
"estimated_review_effort": 3,
"problem_domain": "Documentation Updates"
} | |
scikit-learn__scikit-learn-27841@f946f3e | scikit-learn/scikit-learn | Python | 27,841 | ENH Warn in SelectKBest when k > n_features | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-11-24T17:14:18Z | SelectKBest shouldn't raise if k > n_samples
### Describe the workflow you want to enable
Let's say I want to build a logistic regression model with at most 50 features. I could do that with something like this:
``make_pipeline(ColumnTransformer(...OneHotEncoder(), remainder="passthrough"), SelectKBest(k=50), Logi... | I would not be opposed to raise a warning instead of the error mentioning that all available features are returned.
That would work for me, as the warning would be expected I could filter it out here.
To make sure I understand the case properly: you have a dataset where the number of features passed to `SelectKBest` is... | [
{
"body": "### Describe the workflow you want to enable\r\n\r\nLet's say I want to build a logistic regression model with at most 50 features. I could do that with something like this: \r\n``make_pipeline(ColumnTransformer(...OneHotEncoder(), remainder=\"passthrough\"), SelectKBest(k=50), LogisticRegression())`... | 5ac2eb117a0d242244ac4cff0d9f22abc6873ae2 | {
"head_commit": "f946f3ee3c70cf15fad3a86cc3240e940cffa91d",
"head_commit_message": "DOC Adds whats new",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex f3f40fd26e27a..9f3f4cdf022b2 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -345,6 +345,11 @@ Ch... | [
{
"diff_hunk": "@@ -345,6 +345,11 @@ Changelog\n not check for nans during input validation.\n :pr:`21807` by `Thomas Fan`_.\n \n+- |API| :class:`feature_selection.SelectKBest` and\n+ :class:`feature_selection.GenericUnivariateSelect` with `mode='k_best'",
"line": null,
"original_line": 349,
"o... | efb6f2a76ac2067c7a2ba762075b42c68a41edae | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index 42543f7591266..44d5cfa01c8bb 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -374,6 +374,11 @@ Changelog
not check for nans during input validation.
:pr:`21807` by `Thomas Fan`_.
+- |Enhancement| :class:`feature_selection.Sel... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27937@e3154e9 | scikit-learn/scikit-learn | Python | 27,937 | add feature_names_in_ and n_features_in_ attributes to dummy estimators | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-12-11T08:48:16Z | Dummy estimators don't have the `feature_names_in_` nor `n_features_in_` attributes
### Describe the bug
`DummyClassifier` and `DummyRegressor` estimators don't have the `feature_names_in_` nor `n_features_in_` attributes. The reason is that they don't call `self._validate_data` during `fit` like other estimators do.
... | probably it's more a feature request than a bug. I am willing to make a PR for this.
these estimators are not really checking the input `X` and hence they don't store feature names. I'm not sure if it's a good idea to add that validation step.
It makes sense to not do the validation, but they could still store feature ... | [
{
"body": "### Describe the bug\n\n`DummyClassifier` and `DummyRegressor` estimators don't have the `feature_names_in_` nor `n_features_in_` attributes. The reason is that they don't call `self._validate_data` during `fit` like other estimators do.\n\n### Steps/Code to Reproduce\n\n```python\r\nfrom sklearn.dat... | db971d1e63acdac0d00ac1c32636ceec904f48df | {
"head_commit": "e3154e96fed588fdff6fbb676323e377fe2a1c95",
"head_commit_message": "add changelog and fix test",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex a2aca5841cd9e..8e832db2da603 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -35,6 +35,10... | [
{
"diff_hunk": "@@ -110,6 +110,13 @@ class prior probabilities.\n Frequency of each class observed in `y`. For multioutput classification\n problems, this is computed independently for each output.\n \n+ n_features_in_ : int\n+ Number of features seen during fit.\n+\n+ feature_names... | 7166583078113b5d7bf33dc3208905c097320e61 | diff --git a/doc/whats_new/v1.5.rst b/doc/whats_new/v1.5.rst
index 96cbd21021f08..4a44bd6666615 100644
--- a/doc/whats_new/v1.5.rst
+++ b/doc/whats_new/v1.5.rst
@@ -25,12 +25,20 @@ Changelog
:pr:`123456` by :user:`Joe Bloggs <joeongithub>`.
where 123455 is the *pull request* number, not the issue number.
+
... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27650@9ddc618 | scikit-learn/scikit-learn | Python | 27,650 | ENH Add warning when only one label found in `confusion_matrix` | #### Reference Issues/PRs
fixes #19756
#### What does this implement/fix? Explain your changes.
Add warning when only one label found in `confusion_matrix` and advise user to pass all labels via `labels` parameter.
| 2023-10-24T05:59:03Z | Confusion Matrix is of wrong shape
#### Describe the bug
Having 100 percent true positive rate gives a 1x1 matrix. This can cause bugs if you are using the confusion matrix in a function to calculate evaluation parameters such as sensitivity, auc, specifity, accuracy.
```
from sklearn.metrics import confusio... | If `y_true` and `y_pred` does not contain all the known labels, then you can use `labels` to pass them in:
```py
from sklearn.metrics import confusion_matrix, roc_auc_score
ground_truth, predictions = np.ones((2,1))
cf_matrix = confusion_matrix(ground_truth, predictions, labels=[0, 1])
cf_matrix
# array([[0, ... | [
{
"body": "\r\n\r\n#### Describe the bug\r\n\r\nHaving 100 percent true positive rate gives a 1x1 matrix. This can cause bugs if you are using the confusion matrix in a function to calculate evaluation parameters such as sensitivity, auc, specifity, accuracy.\r\n\r\n```\r\nfrom sklearn.metrics import confusion_... | 1f1329f7ecb001eda2ff8e6d6a68bc2054c4962f | {
"head_commit": "9ddc6188e619bdd3cb220786423f99a24029f52b",
"head_commit_message": "black",
"patch_to_review": "diff --git a/sklearn/metrics/_classification.py b/sklearn/metrics/_classification.py\nindex d1dc3232c53ab..73260a2be1d52 100644\n--- a/sklearn/metrics/_classification.py\n+++ b/sklearn/metrics/_classif... | [
{
"diff_hunk": "@@ -382,6 +382,15 @@ def confusion_matrix(\n cm = cm / cm.sum()\n cm = np.nan_to_num(cm)\n \n+ if cm.shape == (1, 1):\n+ warnings.warn(\n+ (\n+ \"Only one label was found in 'y_true' and 'y_pred', use the 'labels' \"",
"line": null,
... | 33fe2a09b0158828809bef4255e3a8a2b2fcdb3f | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index a947dfe225b04..0063fa77bf87a 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -376,6 +376,9 @@ Changelog
`predict_proba`). Such scorer are specific to classification.
:pr:`26840` by :user:`Guillaume Lemaitre <glemaitre>`.
+- |... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27721@369719e | scikit-learn/scikit-learn | Python | 27,721 | ENH support unsupervised feature selection for percentile and k-best | closes #27609
Allow `SelectKBest`, `SelectPercentile`, and `GenericUnivariateSelection` (with k-best and percentile modes) to work with unsupervised scoring function.
| 2023-11-04T14:21:01Z | Add a version of `GenericUnivariateSelect`/`SelectPercentile`/`SelectKBest` that allows input of X with missing values and `y=None`.
### Describe the workflow you want to enable
- Select features by the percentage of missing values of X
- Select features only by statistical properties of X before y is available
### ... | Thanks for creating the issue. Could you add a bit more context on use-cases where it would be useful to have this and maybe examples in the wild where this is being used. This will help to decide if we want to include this in scikit-learn.
> Select features by the percentage of missing values of X
For example, if o... | [
{
"body": "### Describe the workflow you want to enable\n\n- Select features by the percentage of missing values of X\r\n- Select features only by statistical properties of X before y is available\n\n### Describe your proposed solution\n\nCreate a version with weaker X, y checks.\n\n### Describe alternatives yo... | 0027893288dd58806c731de7efc52b6c8e6c131a | {
"head_commit": "369719eab21427419c70d8f97606037e73a47abe",
"head_commit_message": "Update doc/whats_new/v1.4.rst",
"patch_to_review": "diff --git a/doc/modules/feature_selection.rst b/doc/modules/feature_selection.rst\nindex 1190d17019183..7fcec524e7168 100644\n--- a/doc/modules/feature_selection.rst\n+++ b/doc... | [
{
"diff_hunk": "@@ -680,6 +689,9 @@ class SelectKBest(_BaseFilter):\n Ties between features with equal scores will be broken in an unspecified\n way.\n \n+ This filter supports unsupervised feature selection that only request `X` for",
"line": null,
"original_line": 692,
"original_start_l... | 43220b1b0366ae19a624160e005f6ea7a88f57fb | diff --git a/doc/modules/feature_selection.rst b/doc/modules/feature_selection.rst
index 1190d17019183..7fcec524e7168 100644
--- a/doc/modules/feature_selection.rst
+++ b/doc/modules/feature_selection.rst
@@ -108,6 +108,12 @@ applied to non-negative features, such as frequencies.
Beware not to use a regression sco... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-27412@69dd9e3 | scikit-learn/scikit-learn | Python | 27,412 | FIX make `roc_auc_score` consistent with `roc_curve` with a single class | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-09-18T21:08:32Z | Inconsistency in AUC ROC and AUPR API
### Describe the bug
When only one class is present on the groundtruth. The function `roc_auc_score` throws an `ValueError` and exits while the `average_precision_score` returns `-0.0`. I feel that both functions should return similar output in this case.
### Steps/Code to Reprod... | `average_precision_score` is not an area under precision-recall curve, it's a different metric
@GLevV isn't it an discrete estimation of the area under the PR curve?
https://glassboxmedicine.com/2019/03/02/measuring-performance-auprc/
EDIT- added the correct link
Actually, they are different.
Check [sklearn.metri... | [
{
"body": "### Describe the bug\n\nWhen only one class is present on the groundtruth. The function `roc_auc_score` throws an `ValueError` and exits while the `average_precision_score` returns `-0.0`. I feel that both functions should return similar output in this case.\n\n### Steps/Code to Reproduce\n\n```pytho... | 364cafe979592abdf90b56722e18c27f3b02b9b2 | {
"head_commit": "69dd9e38d68a0ac581bc04f742458d0d3e91c336",
"head_commit_message": "bugfix",
"patch_to_review": "diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst\nindex e93eb01d3d32d..629944b7e52be 100644\n--- a/doc/whats_new/v1.6.rst\n+++ b/doc/whats_new/v1.6.rst\n@@ -326,6 +326,10 @@ Changelog\n ... | [
{
"diff_hunk": "@@ -355,6 +354,24 @@ def test_roc_curve_toydata():\n assert_array_almost_equal(fpr, [0, 1])\n assert_almost_equal(roc_auc, 0.5)\n \n+ y_true = [1, 1]",
"line": null,
"original_line": 357,
"original_start_line": null,
"path": "sklearn/metrics/tests/test_ranking.py",
... | 8cbac3e1e20ed0ec9488d56c8221b4b4102633f2 | diff --git a/doc/whats_new/v1.6.rst b/doc/whats_new/v1.6.rst
index e93eb01d3d32d..629944b7e52be 100644
--- a/doc/whats_new/v1.6.rst
+++ b/doc/whats_new/v1.6.rst
@@ -326,6 +326,10 @@ Changelog
classification labels.
:pr:`29738` by `Adrin Jalali`_.
+- |Fix| :func:`metrics.roc_auc_score` will now correctly return ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27387@b00ad8c | scikit-learn/scikit-learn | Python | 27,387 | DOC: add note about restarting conda environment after creation | #### Reference Issues/PRs
Fixes #27356
#### What does this implement/fix? Explain your changes.
Adds an explanatory notes section that the conda environment after creation needs to be deactivated and reactivated before environment variables become active. It is the same notes section as for the macOS conda bui... | 2023-09-15T20:16:55Z | Building from source fails using conda instructions
### Describe the bug
I am trying to build the development version of scikit-learn in Windows Subsystem for Linux running Ubuntu 20.04.5 and conda 23.7.2.
After creating the environment, building scikit-learn fails as cython is not found for some reason.
### S... | Could you close and reopen a terminal between the step 1 and step 2. Since `compilers` will modify some environment variables, it can happen that those don't take effect.
I tried on my linux machine and the above command are working as expected.
Thank you @glemaitre , that worked. Is this worth being noted in the do... | [
{
"body": "### Describe the bug\r\n\r\nI am trying to build the development version of scikit-learn in Windows Subsystem for Linux running Ubuntu 20.04.5 and conda 23.7.2.\r\n\r\nAfter creating the environment, building scikit-learn fails as cython is not found for some reason.\r\n\r\n### Steps/Code to Reproduc... | 8faa92011e76bace1df02230bce83ee3a5e083b0 | {
"head_commit": "b00ad8ce20f34d71bd1c731484f70677bd532e35",
"head_commit_message": "add accidentally deleted freebsd header",
"patch_to_review": "diff --git a/doc/developers/advanced_installation.rst b/doc/developers/advanced_installation.rst\nindex a9461e965af2a..bfcbd1649d56f 100644\n--- a/doc/developers/advan... | [
{
"diff_hunk": "@@ -286,7 +292,13 @@ scikit-learn from source:\n .. prompt:: bash $\n \n conda create -n sklearn-dev -c conda-forge python numpy scipy cython \\\n- joblib threadpoolctl pytest compilers llvm-openmp\n+ joblib threadpoolctl pytest compilers",
"line": null,
"original_line"... | 3532a44cc8a7481d892805fb9762e7076588d900 | diff --git a/doc/developers/advanced_installation.rst b/doc/developers/advanced_installation.rst
index a9461e965af2a..2eab1bb06d979 100644
--- a/doc/developers/advanced_installation.rst
+++ b/doc/developers/advanced_installation.rst
@@ -69,6 +69,12 @@ feature, code or documentation improvement).
.. prompt:: bash $
... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27700@4141451 | scikit-learn/scikit-learn | Python | 27,700 | ENH Add `chain_method` to `ClassifierChain` | #### Reference Issues/PRs
Fixes #9247
Closes #9316 (supersedes)
#### What does this implement/fix? Explain your changes.
Add `chain_method` to `ClassifierChain`. Supports `{'predict', 'predict_proba', 'predict_log_proba', 'decision_function'}` (as suggested in https://github.com/scikit-learn/scikit-learn/pull/931... | 2023-11-01T02:27:08Z | Chain on decision_function or predict_proba in ClassifierChain
ClassifierChain uses the prediction from the classifier for each label as a feature for the next label's classifier. Instead of chaining on the categorical prediction it should be able to select a `method` to chain on, one of { predict, predict_proba, decis... | Hi, I would like to work on it
Go ahead
On Thu, Jun 29, 2017 at 11:43 AM, RAKOTOARISON Herilalaina <
notifications@github.com> wrote:
> Hi, I would like to work on it
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub
> <https://github.com/sciki... | [
{
"body": "ClassifierChain uses the prediction from the classifier for each label as a feature for the next label's classifier. Instead of chaining on the categorical prediction it should be able to select a `method` to chain on, one of { predict, predict_proba, decision_function }.",
"number": 9247,
"t... | e94eae47b57ce44038982de5321c4ad44b5a7bcd | {
"head_commit": "41414511514fadb95e6816546371cf4ff2edc00d",
"head_commit_message": "lint",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex 204cbb9e9c890..00b0be9917961 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -529,6 +529,12 @@ Changelog\n ob... | [
{
"diff_hunk": "@@ -492,10 +492,14 @@ def generate_multilabel_dataset_with_correlations():\n return X, Y_multi\n \n \n-def test_classifier_chain_fit_and_predict_with_linear_svc():\n+@pytest.mark.parametrize(\"chain_method\", [\"predict\", \"decision_function\"])\n+def test_classifier_chain_fit_and_predict_w... | 969d8892a6e78ae91c028374f3479d0c94a9cb86 | diff --git a/doc/whats_new/v1.5.rst b/doc/whats_new/v1.5.rst
index 53f0fbd8a74e8..d70c9cc2f1f23 100644
--- a/doc/whats_new/v1.5.rst
+++ b/doc/whats_new/v1.5.rst
@@ -158,6 +158,12 @@ Changelog
- |Enhancement| :term:`CV splitters <CV splitter>` that ignores the group parameter now
raises a warning when groups are pas... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-28509@e6a74da | scikit-learn/scikit-learn | Python | 28,509 | FEA add zero_division to matthews_corrcoef | #### Reference Issues/PRs
Partially address: https://github.com/scikit-learn/scikit-learn/issues/29048
Fixes #25258
See also #19977
#### What does this implement/fix? Explain your changes.
The Matthews correlation coefficient is ill-defined (due to zero division) when only one class is present in either the tr... | 2024-02-22T21:10:44Z | Matthews correlation coefficient gives wrong results for single-class result
### Describe the bug
for a single class input, (ie perfect correlation) MCC returns 0, when it should return 1.
since MCC is a special case of pearson correlation, it should fall back to pearson when appropriate
### Steps/Code to Reproduc... | First, I thought that we could quickly fix the problem you pointed out with something like:
```python
if all(y_true == y_pred):
return 1
```
But this problem is more general: MCC and Pearson Correlation (as far as I understand) are not defined if we have a constant array/variable because the standard deviation... | [
{
"body": "### Describe the bug\n\nfor a single class input, (ie perfect correlation) MCC returns 0, when it should return 1.\r\n\r\nsince MCC is a special case of pearson correlation, it should fall back to pearson when appropriate\n\n### Steps/Code to Reproduce\n\nfrom sklearn.metrics import matthews_corrcoef... | 087d8b7a4aeef796ce175c8c8012b4211c2a175b | {
"head_commit": "e6a74da862233370d7e77263c78d6c5e9473dcaf",
"head_commit_message": "Fixed a typo",
"patch_to_review": "diff --git a/sklearn/metrics/_classification.py b/sklearn/metrics/_classification.py\nindex b6d59d25696ce..a5db319cea4cb 100644\n--- a/sklearn/metrics/_classification.py\n+++ b/sklearn/metrics/_... | [
{
"diff_hunk": "@@ -886,12 +886,13 @@ def test_matthews_corrcoef():\n y_true_inv2 = np.where(y_true_inv2, \"a\", \"b\")\n assert_almost_equal(matthews_corrcoef(y_true, y_true_inv2), -1)\n \n- # For the zero vector case, the corrcoef cannot be calculated and should\n- # output 0\n- assert_almost... | a8a4ca0bda212ba58c253352c94c39e8b24f00f4 | diff --git a/doc/whats_new/upcoming_changes/sklearn.metrics/28509.feature.rst b/doc/whats_new/upcoming_changes/sklearn.metrics/28509.feature.rst
new file mode 100644
index 0000000000000..755d586dbce2b
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.metrics/28509.feature.rst
@@ -0,0 +1,3 @@
+- Adds `zero_divi... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27645@cc576c2 | scikit-learn/scikit-learn | Python | 27,645 | FIX IsolationForest accepts sparse matrix with float value contamination | closes #27626
This PR makes sure to convert to CSR the validated CSC matrix when contamination is a floating number. We still don't want to call `score_samples` because it will cost an additional validation, more expensive than the expected `X.asformat("csr")`. | 2023-10-23T17:48:40Z | Isolation Forest Bug with Sparse Matrix and Contamination as Float
### Describe the bug
### Environment:
```
Python 3.11 and 3.8
Scikit-learn library 1.3.1
Isolation Forest algorithm
Sparse matrix input (tested csr and csc)
Contamination parameter set as a float
```
### Bug Summary:
When using the Iso... | I can reproduce. The thing is that we call `_score_samples` to avoid a second input validation. However, the first validation make sure that we have a CSC matrix while `_score_samples` expect a CSR matrix. We should therefore make the conversion before calling the private function since we still want to avoid all extra... | [
{
"body": "### Describe the bug\r\n\r\n### Environment:\r\n\r\n```\r\nPython 3.11 and 3.8\r\nScikit-learn library 1.3.1\r\nIsolation Forest algorithm\r\nSparse matrix input (tested csr and csc)\r\nContamination parameter set as a float\r\n```\r\n\r\n### Bug Summary:\r\n\r\nWhen using the Isolation Forest algori... | 1471fa7c5816e6340d25e67c6f9555f142bc3aea | {
"head_commit": "cc576c269a7035ed2522111bf2536319bcc3c11a",
"head_commit_message": "add changelog",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex a947dfe225b04..896b018c89620 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -276,6 +276,10 @@ Changel... | [
{
"diff_hunk": "@@ -348,7 +348,9 @@ def fit(self, X, y=None, sample_weight=None):\n \n # Else, define offset_ wrt contamination parameter\n # To avoid performing input validation a second time we call\n- # _score_samples rather than score_samples\n+ # X to a CSR matrix while it was... | e5cb2295be942ea1eaed2321bcb155fd0eadf782 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index 2f73e0a03d68f..ccfec4c9c2bb5 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -368,6 +368,10 @@ Changelog
is a predefined metric listed in :func:`metrics.get_scorer_names` and
early stopping is enabled. :pr:`26163` by `Thomas Fa... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27309@87c19e2 | scikit-learn/scikit-learn | Python | 27,309 | FIX raise an error if nan is not the last element in user provided categories | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-09-07T01:24:38Z | Wrong infrequent categories and error in OrdinalEncoder
### Describe the bug
When I manually set the numpy object to `categories` in OrdinalEncoder, I got wrong `infrequent_categories_`.
If I run `fit_transform`, then I got an error. See the code below.
### Steps/Code to Reproduce
```python
import numpy as np
fr... | For numerical categories, nan should be the last element if there are nans. But for other types of categories, e.g., object, do we need to add this constraint?
Would you like to take a look at this? @thomasjpfan
Yea, the current infrequent API assumes that nans are the last element even for object dtypes. For now, I t... | [
{
"body": "### Describe the bug\n\nWhen I manually set the numpy object to `categories` in OrdinalEncoder, I got wrong `infrequent_categories_`. \r\nIf I run `fit_transform`, then I got an error. See the code below.\n\n### Steps/Code to Reproduce\n\n```python\r\nimport numpy as np\r\nfrom sklearn.preprocessing ... | a5fed0d77842a229a98655485530264c82ca59d6 | {
"head_commit": "87c19e251a6cf4677a99e10f9fa45fe486fd4b11",
"head_commit_message": "add an entry in changelog v1.4",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex 2a631d43ac5c2..1390ee594ceff 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -450,6 +... | [
{
"diff_hunk": "@@ -123,13 +123,13 @@ def _fit(\n )\n raise ValueError(msg)\n \n- last_idx = len(cats) - 1\n- for idx, category in enumerate(cats):\n- if is_scalar_nan(category) and idx != last_idx:\n+ # `nan... | 8c18bcc05d7563f3d65872417c98165aa04758a5 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index 2a631d43ac5c2..ce0a2c175ea48 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -450,6 +450,11 @@ Changelog
- |Enhancement| :class:`preprocessing.TargetEncoder` now supports `target_type`
'multiclass'. :pr:`26674` by :user:`Lucy Liu... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27167@75661f6 | scikit-learn/scikit-learn | Python | 27,167 | Fix low level crash in BisectingKMeans.predict on rescaled data | Fixes #27081.
I will first open this PR with the non-regression test that should hopefully make the macOS Intel CI workers segfault (for some reason, this bug seems to stay silent on other platforms)...
Then I will push a fix for the underlying root cause. | 2023-08-25T16:03:08Z | BisectingKMeans floating point exception fatal error on OSX
### Describe the bug
Hi, thanks for this fantastic library 🚀 🎉
I believe I found a bug. In the BisectingKMeans clustering, the `predict` function cannot be used when the to-be-predicted-data is on a different numerical scale than the fitted data. Othe... | This is unexpected and is likely to be something specific to your machine being a MacOS or your environment.
One thing I noticed in your versions is that you have a different OpenBLAS version from numpy and scipy. Could it be the issue, hard to tell.
Could you follow the instructions from https://github.com/sciki... | [
{
"body": "### Describe the bug\r\n\r\nHi, thanks for this fantastic library 🚀 🎉 \r\n\r\nI believe I found a bug. In the BisectingKMeans clustering, the `predict` function cannot be used when the to-be-predicted-data is on a different numerical scale than the fitted data. Other clustering methods like e.g., K... | d90dd114ab85b85e49facdb7b37242b7362696c8 | {
"head_commit": "75661f6d2145b602e6369427abe1c6c9f3909e14",
"head_commit_message": "DOC update changelog",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex d97ccea9469eb..05e9584028647 100644\n--- a/doc/whats_new/v1.3.rst\n+++ b/doc/whats_new/v1.3.rst\n@@ -18,6 +18,13 @@ Ch... | [
{
"diff_hunk": "@@ -133,3 +133,17 @@ def test_float32_float64_equivalence(csr_container):\n \n assert_allclose(km32.cluster_centers_, km64.cluster_centers_)\n assert_array_equal(km32.labels_, km64.labels_)\n+\n+\n+def test_non_regression_segfault_on_empty_bisections():",
"line": null,
"original_... | a4a5abf23aed937d5f4917140a40ed1ff9ff2600 | diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst
index d97ccea9469eb..05e9584028647 100644
--- a/doc/whats_new/v1.3.rst
+++ b/doc/whats_new/v1.3.rst
@@ -18,6 +18,13 @@ Changes impacting all modules
Changelog
---------
+:mod:`sklearn.cluster`
+......................
+
+- |Fix| :class:`cluster.BisectingKM... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27454@bf88e93 | scikit-learn/scikit-learn | Python | 27,454 | CI raise error for all kind of deprecation warnings in doc build | closes #27430
We are currently catching any deprecation warnings in the test and source code but not in our examples. This PRs intend to raise an error if there is any.
Since we are using lock file, we should not have warning until and if any when updating the lock files, then we should fix them.
It is expecte... | 2023-09-22T21:09:37Z | Catching deprecation warnings from examples
Up-to-now, we never check deprecation warnings that are raised when executing our examples.
I assume that we could scrap the RST generated file to find such warning:
```python
from pathlib import Path
from pprint import pprint
from sklearn.utils.fixes import VisibleD... | For instance this is the current output:
```shell
> python warning_scraping_from_generated_examples.py
/home/glemaitre/Documents/packages/scikit-learn/doc/conf.py:81: UserWarning:
jupyterlite_sphinx is not installed, you need to install it if you want JupyterLite links to appear in each example
['auto_exampl... | [
{
"body": "Up-to-now, we never check deprecation warnings that are raised when executing our examples.\r\nI assume that we could scrap the RST generated file to find such warning:\r\n\r\n```python\r\nfrom pathlib import Path\r\nfrom pprint import pprint\r\n\r\nfrom sklearn.utils.fixes import VisibleDeprecationW... | bd162e5548e04264589221af2dcc6162fee40863 | {
"head_commit": "bf88e93fe046e32eef36880c57d246d46ca1b2e4",
"head_commit_message": "[doc build][azure parallel] trigger documentation",
"patch_to_review": "diff --git a/doc/conf.py b/doc/conf.py\nindex 1d4b061f9afb5..6dbbb1be670fd 100644\n--- a/doc/conf.py\n+++ b/doc/conf.py\n@@ -692,6 +692,8 @@ def setup(app):\... | [
{
"diff_hunk": "@@ -700,6 +702,11 @@ def setup(app):\n \" non-GUI backend, so cannot show the figure.\"\n ),\n )\n+# Raise warning as error in example to catch warnings when building the documentation\n+# Since we are using lock files to build the documentation, we should not have any\n+# warnings o... | 2e766d7e3dabcd6c8e861204feb6f8e891e8ec69 | diff --git a/doc/conf.py b/doc/conf.py
index 0851d243d17c8..4ef9cb61519ec 100644
--- a/doc/conf.py
+++ b/doc/conf.py
@@ -696,6 +696,8 @@ def setup(app):
),
)
+from sklearn.utils.fixes import VisibleDeprecationWarning
+
warnings.filterwarnings(
"ignore",
category=UserWarning,
@@ -704,6 +706,18 @@ def ... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Test Suite / CI Enhancements"
} |
scikit-learn__scikit-learn-27456@8d7a266 | scikit-learn/scikit-learn | Python | 27,456 | FEA Allow string input for pairwise distances | #### Reference Issues/PRs
closes #15932
closes #17991
closes #24674
supersedes #17991
supersedes #24674
#### What does this implement/fix? Explain your changes.
This allows the user to compute pairwise distance using a custom metric for non-metric data types (e.g. string or boolean).
#### Any other comm... | 2023-09-23T22:44:51Z | compute pairwise_distance with custom metric function for non-numeric data
This is a feature request: I'm working with strings and have a custom metric function to compute similarities between the strings. I would love to use the pairwise_distance function together with this custom metric to compute the whole distance/... | Thanks for the feature request. Sounds useful indeed (though I have not studied the question in detail). See earlier related discussion in https://github.com/scikit-learn/scikit-learn/issues/3737 and https://github.com/scikit-learn/scikit-learn/issues/4208
Can I work on this request?
You're welcome to work on this @ven... | [
{
"body": "This is a feature request: I'm working with strings and have a custom metric function to compute similarities between the strings. I would love to use the pairwise_distance function together with this custom metric to compute the whole distance/similarity matrix for my data. However, I'm getting a `V... | f4c058ac0913895318b5b0f5001bedf54c94bd36 | {
"head_commit": "8d7a266a62cf0b008e312f59ccdbf3335fad030c",
"head_commit_message": "Update sklearn/metrics/tests/test_pairwise.py",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex 07f4d335058b4..43b654ef511b4 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.r... | [
{
"diff_hunk": "@@ -130,6 +135,14 @@ def check_pairwise_arrays(\n .. versionchanged:: 0.23\n Accepts `pd.NA` and converts it into `np.nan`.\n \n+ only_check_num_samples : bool, default=False\n+ Whether to only check the length of the array.",
"line": null,
"original_line": 1... | 3df1d5ce9afff5ff2cf927768f6ce12bc48ddb99 | diff --git a/doc/whats_new/v1.5.rst b/doc/whats_new/v1.5.rst
index aad9fe591def7..65c9558b88bf2 100644
--- a/doc/whats_new/v1.5.rst
+++ b/doc/whats_new/v1.5.rst
@@ -218,6 +218,11 @@ Changelog
:mod:`sklearn.metrics`
......................
+- |Feature| :func:`metrics.pairwise_distances` accepts calculating pairwise d... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-26931@8eb9b28 | scikit-learn/scikit-learn | Python | 26,931 | FIX Adds more informative error message for OHE | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-07-29T00:48:43Z | Improve error message with pandas output and sparse data
Currently, we raise an error:
```
ValueError: Pandas output does not support sparse data.
```
when a transformer does output sparse data and we try to wrap it inside a dataframe due to `set_output(transform="pandas")`. I assume that at this point, we coul... | I am +1 with adding a more informative error message for transformers with `sparse_output`.
Hi @thomasjpfan ,
I am new to open source and would love to work on this issue with your help and guidance.
@sqali You should start by going through the contribution guideline: https://scikit-learn.org/dev/developers/contribu... | [
{
"body": "Currently, we raise an error:\r\n\r\n```\r\nValueError: Pandas output does not support sparse data.\r\n```\r\n\r\nwhen a transformer does output sparse data and we try to wrap it inside a dataframe due to `set_output(transform=\"pandas\")`. I assume that at this point, we could be extra nice and chec... | 2b0eef892d4ee973789089ac82f24669a89a3859 | {
"head_commit": "8eb9b28fe20802883d09fb21a8af8f5bb8fc4ffd",
"head_commit_message": "TST Adjust comment test",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex 1f66619ae2219..1bdce6df4d6c7 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -150,6 +150,13 ... | [
{
"diff_hunk": "@@ -1008,6 +1009,14 @@ def transform(self, X):\n returned.\n \"\"\"\n check_is_fitted(self)\n+ transform_output = _get_output_config(\"transform\", estimator=self)[\"dense\"]\n+ if transform_output == \"pandas\" and self.sparse_output:\n+ rais... | 713207f2fc7ffb4142a729be6fa31ddd93cfabe8 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index 1f66619ae2219..1bdce6df4d6c7 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -150,6 +150,13 @@ Changelog
metric parameter `p` is in the range `0 < p < 1`, regardless of the `dtype` of `X`.
:pr:`26760` by :user:`Shreesha Kumar B... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-28043@2c261c9 | scikit-learn/scikit-learn | Python | 28,043 | ENH Add support for np.nan values in SplineTransformer | #### Reference Issues/PRs
Closes #26793
#### What does this implement/fix? Explain your changes.
Adds support for np.nan values in SplineTransformer.
- adds param `handle_missing : {'error', 'zeros'}` to __init__, where `error` preserves the previous behaviour and `zeros` handles nan values by setting their spli... | 2024-01-02T12:05:59Z | Handle np.nan / missing values in SplineTransformer
I think it would be quite natural to add an option to `SplineTransformer` to accept inputs with missing values as follows:
- `handle_missing="error"` the default (keep current behavior)
- `handle_missing="zero"`/"constant": encode missing values by setting all out... | Note: we could also add `handle_missing="passthrough"` for consistency with other preprocessors that allow this. I am not sure it would be that useful though. Maybe `handle_missing="passthrough"` would be a YAGNI.
I will take this issue.
@lorentzenchr do you have any opinion on which options are best for this estimator... | [
{
"body": "I think it would be quite natural to add an option to `SplineTransformer` to accept inputs with missing values as follows:\r\n\r\n- `handle_missing=\"error\"` the default (keep current behavior)\r\n- `handle_missing=\"zero\"`/\"constant\": encode missing values by setting all output features for that... | c92330fee2f602802446a0c3c6023f1908004d13 | {
"head_commit": "2c261c939c6385da0bb08748f88c36fc2fa6ac72",
"head_commit_message": "delete unused code",
"patch_to_review": "diff --git a/doc/whats_new/v1.5.rst b/doc/whats_new/v1.5.rst\nindex f313d0c675ceb..26010252f7abc 100644\n--- a/doc/whats_new/v1.5.rst\n+++ b/doc/whats_new/v1.5.rst\n@@ -131,7 +131,7 @@ Cha... | [
{
"diff_hunk": "@@ -642,6 +643,14 @@ class SplineTransformer(TransformerMixin, BaseEstimator):\n Order of output array in the dense case. `'F'` order is faster to compute, but\n may slow down subsequent estimators.\n \n+ handle_missing : {'error', 'zeros'}, default='error'\n+ Specifies... | 08a72304bf37a11fc68875a437af176cbf1e37c0 | diff --git a/doc/whats_new/upcoming_changes/sklearn.preprocessing/28043.enhancement.rst b/doc/whats_new/upcoming_changes/sklearn.preprocessing/28043.enhancement.rst
new file mode 100644
index 0000000000000..8195352292539
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.preprocessing/28043.enhancement.rst
@@ -... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-26681@b0447ad | scikit-learn/scikit-learn | Python | 26,681 | MNT linter bot: tool versions and finished linting check | This fixes a few issues:
- linter should fail if downloading `linting.sh` fails (hence moving `set +e`)
- linter should fail if `linting.sh` failes to finish (@glemaitre noticed this in some PRs)
- comment includes the version of the tool used (@lucyleeow, Fixes https://github.com/scikit-learn/scikit-learn/issues/... | 2023-06-23T11:19:17Z | CI: Add black version in linter bot
Great job on the linter bot @adrinjalali !
Would it be a good idea to add in the comment what the desired black (and ruff ??) versions are? I think I recall for black at least that different versions can affect formatting... | [
{
"body": "Great job on the linter bot @adrinjalali !\r\n\r\nWould it be a good idea to add in the comment what the desired black (and ruff ??) versions are? I think I recall for black at least that different versions can affect formatting...",
"number": 26676,
"title": "CI: Add black version in linter ... | 02dce942d7232e82ecfd266a613d12f4e2428ad6 | {
"head_commit": "b0447ad20ab2fc0cb81a0ecbe03812bc742a694b",
"head_commit_message": "MNT linter bot: tool versions and finished linting check",
"patch_to_review": "diff --git a/.github/workflows/lint.yml b/.github/workflows/lint.yml\nindex e4b606f36b7f8..9648f66d840d4 100644\n--- a/.github/workflows/lint.yml\n+++... | [
{
"diff_hunk": "@@ -7,6 +7,26 @@\n import requests\n \n \n+def get_versions(versions_file):\n+ \"\"\"Get the versions of the packages used in the linter job.\n+\n+ Parameters\n+ ----------\n+ versions_file : str\n+ The path to the file that contains the versions of the packages.\n+\n+ Retu... | ccdb54c602dd7be58f250c56c510ad5aa2da4b42 | diff --git a/.github/workflows/lint.yml b/.github/workflows/lint.yml
index e4b606f36b7f8..9648f66d840d4 100644
--- a/.github/workflows/lint.yml
+++ b/.github/workflows/lint.yml
@@ -34,14 +34,19 @@ jobs:
source build_tools/shared.sh
# Include pytest compatibility with mypy
pip install py... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Test Suite / CI Enhancements"
} | |
scikit-learn__scikit-learn-26772@fa77f00 | scikit-learn/scikit-learn | Python | 26,772 | FIX Dataframe support in KNN.predict | Fixes https://github.com/scikit-learn/scikit-learn/issues/26768
knn.predict doesn't validate X right away because it call `kneighbors` that does validate X.
The issue is that ``ArgKMin.is_usable_for`` assumes that it's a numpy array. This PR simply adds a check there. Note that it is not usable for a pandas datafra... | 2023-07-05T09:35:07Z | AttributeError "Flags object has no attribute 'c_contiguous'" when using KNeighborsClassifier predict
### Describe the bug
I tried to build a K-nearest neighbor model. I used the default ``weights`` parameter and called the ``predict`` method with a pandas DataFrame object and got the following error:
AttributeEr... | Thanks for the report @wejaeger. A fix is proposed in #26772. | [
{
"body": "### Describe the bug\r\n\r\nI tried to build a K-nearest neighbor model. I used the default ``weights`` parameter and called the ``predict`` method with a pandas DataFrame object and got the following error:\r\n\r\nAttributeError: 'Flags' object has no attribute 'c_contiguous'\r\n\r\n### Steps/Code t... | 34446739dfbe5cb61d2ddccb1f4fcfaa1efd7a9b | {
"head_commit": "fa77f008c1d244f4903a54c8cb07e603bb85c6b4",
"head_commit_message": "ref",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex e3c444b3ae9ff..a559646fcb48e 100644\n--- a/doc/whats_new/v1.3.rst\n+++ b/doc/whats_new/v1.3.rst\n@@ -2,6 +2,22 @@\n \n .. currentmodule... | [
{
"diff_hunk": "@@ -96,7 +96,9 @@ def is_usable_for(cls, X, Y, metric) -> bool:\n \"\"\"\n \n def is_numpy_c_ordered(X):\n- return hasattr(X, \"flags\") and X.flags.c_contiguous\n+ if not hasattr(X, \"flags\"):",
"line": null,
"original_line": 99,
"original_star... | 61467a0885bffaafc43c3ee08f553aa3bf82d5be | diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst
index 8dcb07b271985..237072fd938da 100644
--- a/doc/whats_new/v1.3.rst
+++ b/doc/whats_new/v1.3.rst
@@ -7,7 +7,7 @@
Version 1.3.1
=============
-**TODO: set date**
+**In development**
Changelog
---------
@@ -15,6 +15,10 @@ Changelog
:mod:`sklearn.nei... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26597@ca3185d | scikit-learn/scikit-learn | Python | 26,597 | DOC FIx the formula for support_fraction in the covariance estimators | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-06-16T19:15:42Z | ValueError: The covariance matrix of the support data is equal to 0 - Elliptic Envelope
### Describe the bug
I have been using Elliptic Envelope on a simple time series dataset with default parameters and only setting `contamination` value to `0.5`. However it throws me error like below:
`ValueError: The covarian... | Thank you for reporting this. Could you please extend your "Steps/Code to Reproduce" with a (minimal) `X` that reproduces the problem?
Here are the steps to reproduce it:
```
X = ([[1.04725631],
[1.84803131],
[1.29789207],
[1.59315107],
[0.79657553],
[0. ],
[0. ... | [
{
"body": "### Describe the bug\r\n\r\nI have been using Elliptic Envelope on a simple time series dataset with default parameters and only setting `contamination` value to `0.5`. However it throws me error like below:\r\n\r\n`ValueError: The covariance matrix of the support data is equal to 0, try to increase ... | 714c50092b1bfafad2b568a07d635d5db1301d58 | {
"head_commit": "ca3185d7429cb1fae901468b273474cdd247dc70",
"head_commit_message": "Merge branch 'main' into Doc_support_fraction",
"patch_to_review": "diff --git a/sklearn/covariance/_elliptic_envelope.py b/sklearn/covariance/_elliptic_envelope.py\nindex fe109dddd5303..2bf81cd78a22a 100644\n--- a/sklearn/covari... | [
{
"diff_hunk": "@@ -373,7 +373,7 @@ def fast_mcd(\n The proportion of points to be included in the support of the raw\n MCD estimate. Default is `None`, which implies that the minimum\n value of `support_fraction` will be used within the algorithm:\n- `(n_sample + n_features + 1) ... | 3ab6417589290171623ddd4da75c6aa9bad8d42c | diff --git a/sklearn/covariance/_elliptic_envelope.py b/sklearn/covariance/_elliptic_envelope.py
index fe109dddd5303..ed99a38c0ee56 100644
--- a/sklearn/covariance/_elliptic_envelope.py
+++ b/sklearn/covariance/_elliptic_envelope.py
@@ -35,7 +35,7 @@ class EllipticEnvelope(OutlierMixin, MinCovDet):
support_fractio... | {
"difficulty": "low",
"estimated_review_effort": 1,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26893@98f9412 | scikit-learn/scikit-learn | Python | 26,893 | FIX `param_distribution` param of `HalvingRandomSearchCV` accepts list of dicts | #### What does this implement/fix? Explain your changes.
Closes #26885
Fixed that the `param_distribution` param of `HalvingRandomSearchCV` accepts lists of dicts and updated documentation.
I also tried to implement a test, using` test_random_search_cv_results` as a template, as you suggested @glemaitre , but ... | 2023-07-24T23:29:37Z | HalvingRandomSearchCV does not support param_distribution as a list
### Describe the bug
Before scikit-learn version 1.3.0 (e.g. 1.2.0) HalvingRandomSearchCV could be used with a `list[dict]` as the input for param_distribution (similar to RandomizedSearchCV).
The type hint in the documentation states that only `d... | cc @jeremiedbb @glemaitre
I can take care of it, thanks @adrinjalali.
So we need to enable it and have some unit tests (since we did not break anything it seems to not be tested :)). @StefanieSenger Go ahead, happy to review the PR.
The param description in the docstring should also be updated. It should be like the o... | [
{
"body": "### Describe the bug\r\n\r\nBefore scikit-learn version 1.3.0 (e.g. 1.2.0) HalvingRandomSearchCV could be used with a `list[dict]` as the input for param_distribution (similar to RandomizedSearchCV). \r\nThe type hint in the documentation states that only `dict` is possible but the description talks ... | 38a06e4be504f3971d109d6741b8b4c7192d7323 | {
"head_commit": "98f941287e07d3be5a472ed48b1e15aff76325ef",
"head_commit_message": "bug fix and cleanup in test_grid_search_cv_results and test_random_search_cv_results",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex 8d39ca2fed143..6ba517f4e24a9 100644\n--- a/doc/whats_n... | [
{
"diff_hunk": "@@ -967,31 +963,35 @@ def test_grid_search_cv_results():\n check_cv_results_keys(cv_results, param_keys, score_keys, n_candidates)\n # Check masking\n cv_results = search.cv_results_\n- n_candidates = len(search.cv_results_[\"params\"])\n- assert all(\n+\n+ poly_results = [\... | 5afd7d014d60a44ac3343c8fa5912f77f452b4bf | diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst
index d486b177b0f2c..e9961ab7567a3 100644
--- a/doc/whats_new/v1.3.rst
+++ b/doc/whats_new/v1.3.rst
@@ -23,6 +23,10 @@ Changelog
:attr:`sklearn.neighbors.KDTree.valid_metrics` as public class attributes.
:pr:`26754` by :user:`Julien Jerphanion <jjerphan... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-27002@24875f8 | scikit-learn/scikit-learn | Python | 27,002 | FIX accept multilabel-indicator in _get_response_values | closes #26817
Accept "multilabel-indicator" in `_get_response_values`.
| 2023-08-03T13:34:42Z | Scorer not working on ClassifierChain
### Describe the bug
ClassifeirChain fails when trying to make scoring predictions. The reason is: ClassifeirChain's attribute `classes_` is a list and not a np.array as the `_get_response_values` function expects.
### Steps/Code to Reproduce
```python
from sklearn.multi... | Thanks for reporting this. I can reproduce the problem.
A slight complication is that `ClassifierChain.classes_` is defined as a list of the classes for each estimator in the chain. This is a bit different from what `classes_` should be IMHO. I think it makes sense to fix this and stick to the uniform interface for ... | [
{
"body": "### Describe the bug\r\n\r\nClassifeirChain fails when trying to make scoring predictions. The reason is: ClassifeirChain's attribute `classes_` is a list and not a np.array as the `_get_response_values` function expects. \r\n\r\n### Steps/Code to Reproduce\r\n\r\n```python\r\nfrom sklearn.multioutpu... | 0d701e8fef043b3604008868bcd94a96d1217f4b | {
"head_commit": "24875f8c779f726ad645ce33a06ddd1b04fc7bfa",
"head_commit_message": "Apply suggestions from code review\n\nCo-authored-by: Thomas J. Fan <thomasjpfan@gmail.com>",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex d486b177b0f2c..0a1778a3277b9 100644\n--- a/doc/... | [
{
"diff_hunk": "@@ -94,10 +95,14 @@ def _get_response_values(\n \"classifier with two classes.\"\n )\n raise ValueError(err_msg)\n+ elif target_type == \"multilabel-indicator\" and isinstance(y_pred, list):\n+ y_pred = np.... | 8aeb8b472e573c3d3223877beca2652127faa801 | diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst
index 985cdc5e6411c..893502a1568a6 100644
--- a/doc/whats_new/v1.3.rst
+++ b/doc/whats_new/v1.3.rst
@@ -43,6 +43,13 @@ Changelog
``transform`` when ``add_indicator`` is set to ``True`` and missing values are observed
during ``fit``. :pr:`26600` by :user... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26602@b391e12 | scikit-learn/scikit-learn | Python | 26,602 | FIX ravel prediction of `PLSRegression` when fitted on 1d `y` | #### Reference Issues/PRs
Fixes: #26549.
#### What does this implement/fix? Explain your changes.
This implements https://github.com/scikit-learn/scikit-learn/issues/26549#issuecomment-1594874351 for `PLSRegression`. Please let me know if the alternative (https://github.com/scikit-learn/scikit-learn/issues/265... | 2023-06-18T12:12:27Z | PLSRegression not working with VotingRegressor
Hi,
Unfortunately, PLSRegression does not work in VotingRegressor. Likely related to PLSRegression returning predictions in shape of (n_samples, 1) instead of (n_samples,) like other single-target regressors.
thanks!
```python
import numpy as np
from sklearn.l... | `PLSRegresson `supports multi-target regression, hence the predicition is 2d-array.
As a proposed solution, I suggest creating a custom subclass of `PLSRegression`, which we can call `PLS1D`. This subclass overrides the `predict `method of `PLSRegression `to ensure it returns a 1-dimensional array, making it compati... | [
{
"body": "Hi, \r\n\r\nUnfortunately, PLSRegression does not work in VotingRegressor. Likely related to PLSRegression returning predictions in shape of (n_samples, 1) instead of (n_samples,) like other single-target regressors. \r\n\r\nthanks!\r\n\r\n```python\r\nimport numpy as np\r\nfrom sklearn.linear_model ... | 04860335c82d557e663b4cfa218663d1c7bf65fd | {
"head_commit": "b391e126889f52b024a8379caabb7d6a8bfd36a0",
"head_commit_message": "changelof entry updated",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex 7bdbe7841f0d4..cdcde6561ff30 100644\n--- a/doc/whats_new/v1.3.rst\n+++ b/doc/whats_new/v1.3.rst\n@@ -276,6 +276,13 ... | [
{
"diff_hunk": "@@ -276,6 +276,13 @@ Changelog\n :func:`covariance.graphical_lasso_path`, and :class:`covariance.GraphicalLassoCV`.\n :pr:`26033` by :user:`Genesis Valencia <genvalen>`.\n \n+:mod:`sklearn.cross_decomposition`\n+..................................\n+\n+- |Enhancement| :class:`cross_decomposit... | 53d4907836402206b61e8c9ae254c940599d4e02 | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index e058419c81ca8..0ce08c6eea751 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -66,6 +66,13 @@ Changelog
- |Enhancement| :func:`base.clone` now supports `dict` as input and creates a
copy. :pr:`26786` by `Adrin Jalali`_.
+:mod:`s... | {
"difficulty": "low",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26281@3b4c8a2 | scikit-learn/scikit-learn | Python | 26,281 | Include informative error message when checking classification target is of a non-regression type | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-04-25T10:04:11Z | Unhelpful error message when running a classifier on a regression outcome
### Describe the bug
When running a classifier on a regression outcome, we get a really unhelpful error message:
### Steps/Code to Reproduce
```python
In [2]: from sklearn import linear_model, datasets
In [3]: X, y = datasets.make_regres... | I will take this up and open a PR ASAP. | [
{
"body": "### Describe the bug\n\nWhen running a classifier on a regression outcome, we get a really unhelpful error message:\r\n\n\n### Steps/Code to Reproduce\n\n```python\r\nIn [2]: from sklearn import linear_model, datasets\r\n\r\nIn [3]: X, y = datasets.make_regression()\r\n\r\nIn [4]: model = linear_mode... | ae78c25b00e0c5b76468c28724c7a1f3e710483e | {
"head_commit": "3b4c8a29eb63304dff53a849ce0b7d74615305b4",
"head_commit_message": "Include informative error message",
"patch_to_review": "diff --git a/sklearn/utils/multiclass.py b/sklearn/utils/multiclass.py\nindex 02725f8986948..6b05c750d8331 100644\n--- a/sklearn/utils/multiclass.py\n+++ b/sklearn/utils/mul... | [
{
"diff_hunk": "@@ -215,7 +215,11 @@ def check_classification_targets(y):\n \"multilabel-indicator\",\n \"multilabel-sequences\",\n ]:\n- raise ValueError(\"Unknown label type: %r\" % y_type)\n+ raise ValueError(\n+ \"Unknown label type: %r\"\n+ \"Maybe yo... | 8bf4c481fc32ea1c132ed586d84e837870697d61 | diff --git a/sklearn/utils/multiclass.py b/sklearn/utils/multiclass.py
index 02725f8986948..24e528d10a0a2 100644
--- a/sklearn/utils/multiclass.py
+++ b/sklearn/utils/multiclass.py
@@ -215,7 +215,11 @@ def check_classification_targets(y):
"multilabel-indicator",
"multilabel-sequences",
]:
- ... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26840@8f36235 | scikit-learn/scikit-learn | Python | 26,840 | ENH/DEPR add new response_method and deprecate needs_* in make_scorer | closes #26817
While working in https://github.com/scikit-learn/scikit-learn/pull/26120, I find myself doing dirty tricks to overwrite the response method used in `_ThresholdScorer` while we should be able to pass a list of priorities and delegate to `_get_response_values`.
I introduce `response_method` in `make_s... | 2023-07-15T11:39:59Z | Scorer not working on ClassifierChain
### Describe the bug
ClassifeirChain fails when trying to make scoring predictions. The reason is: ClassifeirChain's attribute `classes_` is a list and not a np.array as the `_get_response_values` function expects.
### Steps/Code to Reproduce
```python
from sklearn.multi... | Thanks for reporting this. I can reproduce the problem.
A slight complication is that `ClassifierChain.classes_` is defined as a list of the classes for each estimator in the chain. This is a bit different from what `classes_` should be IMHO. I think it makes sense to fix this and stick to the uniform interface for ... | [
{
"body": "### Describe the bug\r\n\r\nClassifeirChain fails when trying to make scoring predictions. The reason is: ClassifeirChain's attribute `classes_` is a list and not a np.array as the `_get_response_values` function expects. \r\n\r\n### Steps/Code to Reproduce\r\n\r\n```python\r\nfrom sklearn.multioutpu... | d91e1a5c4622ad6ed64c37b4236a2538ff8f09bd | {
"head_commit": "8f36235db896e210e94b1b209e9820f99277741a",
"head_commit_message": "TST add mulitlabel test",
"patch_to_review": "diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst\nindex c2b7d19404af9..478fc8a1f1c2a 100644\n--- a/doc/whats_new/v1.4.rst\n+++ b/doc/whats_new/v1.4.rst\n@@ -111,6 +111,17 ... | [
{
"diff_hunk": "@@ -452,20 +469,15 @@ def _score(self, method_caller, clf, X, y, **kwargs):\n raise ValueError(\"{0} format is not supported\".format(y_type))\n \n if is_regressor(clf):\n- y_pred = method_caller(clf, \"predict\", X)\n+ response_method = self._response_m... | b80e944082b7562b508d0fdf7164e83db934d994 | diff --git a/doc/modules/model_evaluation.rst b/doc/modules/model_evaluation.rst
index 2ebb7dff7810b..a88a92604767e 100644
--- a/doc/modules/model_evaluation.rst
+++ b/doc/modules/model_evaluation.rst
@@ -181,9 +181,15 @@ take several parameters:
of the python function is negated by the scorer object, conforming to
... | {
"difficulty": "medium",
"estimated_review_effort": 4,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26337@9d8ef58 | scikit-learn/scikit-learn | Python | 26,337 | FIX KernelPCA inverse transform when gamma is not given | #### Reference Issues/PRs
Fixes #26280.
#### What does this implement/fix? Explain your changes.
Quote @jeremiedbb:
> Originally when `gamma` is `None`, its set to `1 / n_features` each time the kernel is called, not once at the beginning of fit. It means that when the kernel is called at `inverse_transform`,... | 2023-05-05T20:12:00Z | KernelPCA inverse transform behaves unexpectly.
### Describe the bug
Hi!
we wanted to use inverse transform in kernel PCA. There is a parameter gamma, which is defined as 1/num_features if gamma=None. However if gamma is provided as gamma=1/num_features, the resultant inverse transform results is different. Why is th... | From what I recall and what the documentation says, the `inverse_transform` is a learnt approximation: https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.KernelPCA.html#sklearn.decomposition.KernelPCA.inverse_transform
It is not surprising to not recover the original data.
> It is not surprising... | [
{
"body": "### Describe the bug\n\nHi!\r\nwe wanted to use inverse transform in kernel PCA. There is a parameter gamma, which is defined as 1/num_features if gamma=None. However if gamma is provided as gamma=1/num_features, the resultant inverse transform results is different. Why is that? How the parameter gam... | ea046f024694b9a558c882b8c2610c52dad95e29 | {
"head_commit": "9d8ef580ff3de1f695de6451bbceac7adca67e78",
"head_commit_message": "changelog added",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex 5965cd90c1b89..903a70c9367d9 100644\n--- a/doc/whats_new/v1.3.rst\n+++ b/doc/whats_new/v1.3.rst\n@@ -230,6 +230,11 @@ Chang... | [
{
"diff_hunk": "@@ -523,3 +523,28 @@ def test_kernel_pca_feature_names_out():\n \n names = kpca.get_feature_names_out()\n assert_array_equal([f\"kernelpca{i}\" for i in range(2)], names)\n+\n+\n+def test_kernel_pca_inverse_correct_gamma():\n+ \"\"\"Check that gamma is set correctly when not provided.... | 676725e6e0de8c757a6467d774d25b6f79f60716 | diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst
index 9cab0db995c5d..b319d5096ca8c 100644
--- a/doc/whats_new/v1.3.rst
+++ b/doc/whats_new/v1.3.rst
@@ -43,6 +43,14 @@ random sampling procedures.
:user:`Jérémie du Boisberranger <jeremiedbb>`,
:user:`Guillaume Lemaitre <glemaitre>`.
+- |Fix| :class:`... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26289@cb58a4e | scikit-learn/scikit-learn | Python | 26,289 | FIX `export_text` and `export_graphviz` accepts feature and class names as array-like | #### Reference Issues/PRs
Fixes #26265.
#### What does this implement/fix? Explain your changes.
Make `tree.export_text` accept both feature names and class names as numpy arrays. Before modification, both `tree.export_graphviz` and `tree.export_text` accept can deal with class names which are numpy arrays, bu... | 2023-04-27T13:39:27Z | sklearn.tree.export_text failing when feature_names supplied
folks, I'm not sure why this works for
```py
import sklearn.tree
print(my_feature_names)
['0' '0 trump' '0 trump versus' ... 'zur' 'zur ckhalten' 'zur ckhalten muss']
tree.export_graphviz(clf, out_file=None, max_depth=4, feature_names=my_feature_names)... | Could you please post a minimal reproducible? (something we can copy paste in its entirety to produce the issue).
@NickKanellos From the error message, it seems that the feature names you passed in is an array, but as [documented](https://scikit-learn.org/stable/modules/generated/sklearn.tree.export_graphviz.html), `fe... | [
{
"body": "folks, I'm not sure why this works for\r\n```py\r\nimport sklearn.tree\r\nprint(my_feature_names)\r\n['0' '0 trump' '0 trump versus' ... 'zur' 'zur ckhalten' 'zur ckhalten muss']\r\n\r\ntree.export_graphviz(clf, out_file=None, max_depth=4, feature_names=my_feature_names)\r\n```\r\nbut not for \r\n\r\... | 8521819eecbacb93deb87fce28842680ab1a5301 | {
"head_commit": "cb58a4e46387500eaadc12400c6625f9f93c5ae4",
"head_commit_message": "added changelog",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex bb245aa466152..6a50d0cb00d1a 100644\n--- a/doc/whats_new/v1.3.rst\n+++ b/doc/whats_new/v1.3.rst\n@@ -491,6 +491,10 @@ Chang... | [
{
"diff_hunk": "@@ -953,16 +953,16 @@ def export_text(\n It can be an instance of\n DecisionTreeClassifier or DecisionTreeRegressor.\n \n- feature_names : list of str, default=None\n- A list of length n_features containing the feature names.\n+ feature_names : array-like of shape (n... | d1cb4a4b19c292d0d4569c56e8265230537d3099 | diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst
index 5965cd90c1b89..5a3e1a6d6d82d 100644
--- a/doc/whats_new/v1.3.rst
+++ b/doc/whats_new/v1.3.rst
@@ -514,6 +514,10 @@ Changelog
for each target class in ascending numerical order.
:pr:`25387` by :user:`William M <Akbeeh>` and :user:`crispinlogan <cri... | {
"difficulty": "medium",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26410@4e81d2a | scikit-learn/scikit-learn | Python | 26,410 | FIX `KNeighborsClassifier` raise when all neighbors of some sample have zero weights | #### Reference Issues/PRs
Fixes #25854.
#### What does this implement/fix? Explain your changes.
Raises a `ValueError` when the weights for all neighbors of some sample are zero. Previous behavior: `predict` predicts the first class, `predict_proba` returns all zero probability. | 2023-05-21T07:30:24Z | `KNeighborsClassifier` with Default Label
### Describe the workflow you want to enable
In `KNeighborsClassifier`, for a user-provided callable `weights` parameter, it would be nice to allow for a situation with no winner. Say I want to give a weight of `0.0` to samples that are "too far away". Then, there can be a sit... | I agree that the current behavior is problematic. But adding a new public parameter for such an edge case sounds too much.
So my suggestion would be to:
- do a PR to raise an informative error when this case happens (am not sure about the type of exception though) so that they can change there `weights` callable to... | [
{
"body": "### Describe the workflow you want to enable\n\nIn `KNeighborsClassifier`, for a user-provided callable `weights` parameter, it would be nice to allow for a situation with no winner. Say I want to give a weight of `0.0` to samples that are \"too far away\". Then, there can be a situation in which *al... | 93e199d517aca98c0eeb222ebfaa2ac99b368e0a | {
"head_commit": "4e81d2a41c7bc0d70fa877657a1c535a82edebb4",
"head_commit_message": "add any_zero_row to api (is this necessary?)",
"patch_to_review": "diff --git a/doc/developers/utilities.rst b/doc/developers/utilities.rst\nindex 2525b2b1365ed..f5f2bf048328a 100644\n--- a/doc/developers/utilities.rst\n+++ b/doc... | [
{
"diff_hunk": "@@ -24,6 +24,25 @@ def min_pos(const floating[:] X):\n return min_val\n \n \n+def any_zero_row(floating[:, :] X):",
"line": null,
"original_line": 27,
"original_start_line": null,
"path": "sklearn/utils/arrayfuncs.pyx",
"start_line": null,
"text": "@user1:\nI would ma... | 2fee151b16641062089a041f1985798ea14f740c | diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
index c0261a51384c6..ba0facf7cdd13 100644
--- a/doc/whats_new/v1.4.rst
+++ b/doc/whats_new/v1.4.rst
@@ -19,6 +19,15 @@ Changelog
is read-only, e.g., a `numpy.memmap` instance.
:pr:`28111` by :user:`Yao Xiao <Charlie-XIAO>`.
+:mod:`sklearn.neighbors`
+... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-26286@43e493b | scikit-learn/scikit-learn | Python | 26,286 | FIX Improve ColumnTransformer error when indexes are different | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-04-26T15:16:44Z | confusing fit failure when using set_config(transform_output="pandas")
### Describe the bug
Using `set_config(transform_output="pandas")` in combination with `ColumnTransformer` may result in a failed fit, since indices do not align as expected. The error will not be especially helpful: "ValueError: Found input variab... | > doesn't work with ColumnTransformer if other transformers maintain aligned indices
Could you please provide a reproducer snippet that triggers the problem with `ColumnTransformer` and include the full traceback?
I assume the problem is related to the use of non-positional indices in the dataframe but it would b... | [
{
"body": "### Describe the bug\n\nUsing `set_config(transform_output=\"pandas\")` in combination with `ColumnTransformer` may result in a failed fit, since indices do not align as expected. The error will not be especially helpful: \"ValueError: Found input variables with inconsistent numbers of samples: [705,... | e5c65906ee7fdab81c2950fab0fe4d04ed7ad522 | {
"head_commit": "43e493b60276278038d5fee9e2f39d3cba4f32ac",
"head_commit_message": "Apply suggestions from code review\n\nCo-authored-by: Christian Lorentzen <lorentzen.ch@gmail.com>",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex bb245aa466152..739ced82f9ec9 100644\n---... | [
{
"diff_hunk": "@@ -2157,3 +2157,31 @@ def test_empty_selection_pandas_output(empty_selection):\n ct.set_params(verbose_feature_names_out=False)\n X_out = ct.fit_transform(X)\n assert_array_equal(X_out.columns, [\"a\", \"b\"])\n+\n+\n+def test_raise_error_if_index_not_aligned():\n+ \"\"\"Check co... | ce041612d3993d727b7cb383053a5f4dd5ee828a | diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst
index b55e5e7578155..a9a2e8d2d3b56 100644
--- a/doc/whats_new/v1.3.rst
+++ b/doc/whats_new/v1.3.rst
@@ -228,6 +228,13 @@ Changelog
:pr:`26385` by :user:`Meekail Zain <micky774>`
+:mod:`sklearn.compose`
+......................
+
+- |Fix| `compose.Column... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "Bug Fixes"
} |
scikit-learn__scikit-learn-26093@608e226 | scikit-learn/scikit-learn | Python | 26,093 | API Accept 'auto' option for 'dual' parameter in LinearSVC and LinearSVR | Fixes #24651
closes #24731
<!--
copilot:summary
-->
### <samp>🤖 Generated by Copilot at cd87daa</samp>
Set `dual` to "auto" for `LinearSVC` and `LinearSVR` in some tests to avoid future warnings. This is a temporary fix for compatibility with the upcoming version of scikit-learn. | 2023-04-05T10:18:14Z | Default value for solving an SVM in primal and dual should be determined automatically
### Describe the workflow you want to enable
Currently the default value for LinearSVC and LinearSVR with regards to solving the problem in dual is True. This means that for users that are unfamiliar with support-vector machines a... | @agramfort I don't know enough about the potential side-effect of solving the SVM problem in the primal. Would having an `"auto"` option indeed be better to switch automatically from dual to primal depending on the shape of `X`?
indeed. Dual should be faster when n_samples < n_features. This would deserve a tiny benchm... | [
{
"body": "### Describe the workflow you want to enable\r\n\r\nCurrently the default value for LinearSVC and LinearSVR with regards to solving the problem in dual is True. This means that for users that are unfamiliar with support-vector machines and do not know the importance of this parameter they could be us... | 86541f2b3bc8a96264e265cb810cf79858544340 | {
"head_commit": "608e226353effbb93440dc4cca2c8334e10ecb0b",
"head_commit_message": "lint",
"patch_to_review": "diff --git a/doc/whats_new/v1.3.rst b/doc/whats_new/v1.3.rst\nindex e47c74a54edd6..480ec9e740031 100644\n--- a/doc/whats_new/v1.3.rst\n+++ b/doc/whats_new/v1.3.rst\n@@ -443,6 +443,12 @@ Changelog\n Th... | [
{
"diff_hunk": "@@ -1,14 +1,35 @@\n from numbers import Integral, Real\n+import warnings\n \n import numpy as np\n \n-from ._base import _fit_liblinear, BaseSVC, BaseLibSVM\n+from ._base import _fit_liblinear, _get_liblinear_solver_type, BaseSVC, BaseLibSVM\n from ..base import BaseEstimator, RegressorMixin, Ou... | 290b254fb331b6ddb168effb54b3470681247a66 | diff --git a/benchmarks/bench_mnist.py b/benchmarks/bench_mnist.py
index 821fe303e8bf7..4bc28ea1a165d 100644
--- a/benchmarks/bench_mnist.py
+++ b/benchmarks/bench_mnist.py
@@ -86,10 +86,10 @@ def load_data(dtype=np.float32, order="F"):
"ExtraTrees": ExtraTreesClassifier(),
"RandomForest": RandomForestClassif... | {
"difficulty": "medium",
"estimated_review_effort": 3,
"problem_domain": "New Feature Additions"
} |
scikit-learn__scikit-learn-25940@518d878 | scikit-learn/scikit-learn | Python | 25,940 | TST fix unstable test_newrand_set_seed | Fixes #25937.
This test would not break on the CI because the ordering is fixed. On my laptop where pytest run tests in non-deterministic order for some reason, this would fail randomly. | 2023-03-22T10:59:22Z | Unstable sklearn/svm/tests/test_bounds.py::test_newrand_set_seed
On my local laptop I often observe a random failure for `test_newrand_set_seed`.
I am running macos m1 with Python 3.11 from conda-forge. scikit-learn has been built with the clang compilers from conda-forge.
Here is a typical run with 10 repetition... | I think I understand the problem: the first time the rng is used the result is 81, but then its state is mutated so the outcome is different.
On my laptop, the ordering of the pytest test in my Python 3.11 environment is not deterministic, so I trigger this often but not on the CI.
Let me do a PR to remove this t... | [
{
"body": "On my local laptop I often observe a random failure for `test_newrand_set_seed`.\r\n\r\nI am running macos m1 with Python 3.11 from conda-forge. scikit-learn has been built with the clang compilers from conda-forge.\r\n\r\nHere is a typical run with 10 repetitions using `pytest-repeat`:\r\n\r\n<detai... | c64e88d69ec13ec8a9b4e91b5eb97a5ea18b9306 | {
"head_commit": "518d878e8fd5402404e7a35ae5cea97e011c255e",
"head_commit_message": "TST fix unstable test_newrand_set_seed",
"patch_to_review": "diff --git a/sklearn/svm/tests/test_bounds.py b/sklearn/svm/tests/test_bounds.py\nindex 23d6be2f44e98..090959a5d1a83 100644\n--- a/sklearn/svm/tests/test_bounds.py\n+++... | [
{
"diff_hunk": "@@ -72,13 +72,30 @@ def test_ill_posed_min_c():\n _MAX_UNSIGNED_INT = 4294967295\n \n \n-@pytest.mark.parametrize(\"seed, val\", [(None, 81), (0, 54), (_MAX_UNSIGNED_INT, 9)])\n-def test_newrand_set_seed(seed, val):\n+def test_newrand_default():\n+ \"\"\"Test that bounded_rand_int_wrap withou... | b7fdb4b594e596c152dbe1e7bba612ee1f76de09 | diff --git a/sklearn/svm/tests/test_bounds.py b/sklearn/svm/tests/test_bounds.py
index 23d6be2f44e98..d51865717e2fa 100644
--- a/sklearn/svm/tests/test_bounds.py
+++ b/sklearn/svm/tests/test_bounds.py
@@ -72,13 +72,24 @@ def test_ill_posed_min_c():
_MAX_UNSIGNED_INT = 4294967295
-@pytest.mark.parametrize("seed, va... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Test Suite / CI Enhancements"
} |
scikit-learn__scikit-learn-25900@427fcce | scikit-learn/scikit-learn | Python | 25,900 | DOC note on calibration impact on ranking | <!--
Thanks for contributing a pull request! Please ensure you have taken a look at
the contribution guidelines: https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to th... | 2023-03-18T12:59:23Z | Isotonic calibration changes rank-based test metrics values
[As discussed with @ogrisel]
#### Describe the bug
Using isotonic calibration changes metrics values. This is because it is a non-strictly monotonic calibration. Sigmoid calibration being strictly monotonic doesn't suffer from this.
#### Steps/Code to R... | Thank you very much @dsleo for the detailed report. I started from your code and pushed the analysis further.
If one introspect the calibrator, on can observe the following:
```python
y_pred_calib = isotonic.predict_proba(X_test)
print(roc_auc_score(y_test, y_pred_calib[:, 1]))
calibrator = isotonic.calibr... | [
{
"body": "[As discussed with @ogrisel]\r\n\r\n#### Describe the bug\r\nUsing isotonic calibration changes metrics values. This is because it is a non-strictly monotonic calibration. Sigmoid calibration being strictly monotonic doesn't suffer from this.\r\n\r\n#### Steps/Code to Reproduce\r\nHere is quick examp... | e3e880f9a749df95727a9c3102feb755387d33bb | {
"head_commit": "427fcce7b55d1d1fffeaa0595d08c15d071316b7",
"head_commit_message": "DOC fix doi",
"patch_to_review": "diff --git a/doc/developers/contributing.rst b/doc/developers/contributing.rst\nindex 9a3efd3d2c8db..a0680ed87887e 100644\n--- a/doc/developers/contributing.rst\n+++ b/doc/developers/contributing... | [
{
"diff_hunk": "@@ -736,7 +736,7 @@ To also generate the example gallery you can use:\n \n .. prompt:: bash $\n \n- make html\n+ make doc",
"line": null,
"original_line": 739,
"original_start_line": null,
"path": "doc/developers/contributing.rst",
"start_line": null,
"text": "@user... | f87a5db519b502a383cae3bc24fe23062cbde44f | diff --git a/doc/developers/contributing.rst b/doc/developers/contributing.rst
index 9a3efd3d2c8db..8ff924fa04a9f 100644
--- a/doc/developers/contributing.rst
+++ b/doc/developers/contributing.rst
@@ -714,7 +714,7 @@ Building the documentation requires installing some additional packages:
pip install sphinx sphi... | {
"difficulty": "low",
"estimated_review_effort": 2,
"problem_domain": "Bug Fixes"
} |
Subsets and Splits
Review Test Instances
Retrieves basic metadata about code review instances but doesn't provide any analytical insights or patterns beyond raw data access.