Codestin Search App

g4brielvs · 2021-10-23T16:24:58Z

Reference Issues/PRs

Fixes #19894.

What does this implement/fix? Explain your changes.

Add data_transposed argument (default=True) to datasets.make_sparse_coded_signal to avoid confusion.
Implement the logic when data_transposed=True.
Add data_transposed argument description to docstring.
Add FutureWarning to inform future change of the default from True to False.
Add test to check future warning message.
Add test to check shapes.

Any other comments?

Thanks #Data Umbrella sprint. Thanks so much for reviewing this PR!

g4brielvs · 2021-10-23T16:27:37Z

@amueller Thank you so much for your kind help! If any changes are necessary, I'd be happy to continue working on this issue. Please let me know we should also change the example on Orthogonal Matching Pursuit.

g4brielvs · 2021-10-23T16:30:29Z

Looking into the check failure on Check Changelog. The future warning should be included to the CHANGELOG.

amueller · 2021-10-23T16:36:13Z

Yes please add an entry to the changelog :)

Co-authored-by: Andreas Mueller <[email protected]>

amueller · 2021-10-23T18:07:26Z

You will also need to add an entry to the changelog, then you're all set!

jeremiedbb

Thanks @g4brielvs !

I think that to be coherent we should also transpose the code and the dictionary. Actually, instead of transposing them at the end, we should change the code to directly create them with the right shape. For instance, D = generator.randn(n_features, n_components) becomes D = generator.randn(n_components, n_features).

Then you can add a test to check that the shapes of all 3 returned arrays are the expected shapes

…ded_signal

reshamas · 2021-11-05T16:12:17Z

Hello @g4brielvs, How is this PR going?

g4brielvs · 2021-11-05T16:21:40Z

@reshamas I'm sorry for the delay. I'll submit the changes before the end of the week. I'd really appreciate the opportunity, if possible.

g4brielvs · 2022-02-15T22:30:35Z

@thomasjpfan Thank you so much and my apologies! I am aware of the contributing guide, but unfortunately I had black unpinned on the environment. I updated to black==22.1.0 and I'll wait to see if the tests are passing.

thomasjpfan

Thanks for the update. I left a comment regarding the API.

…-make_sparse_coded_signal

thomasjpfan · 2022-03-05T23:10:13Z

From the discussion in #21425 (comment), we can move forward with changing the order when data_transposed=False. Basically this behavior:

Y, D, X = make_sparse_coded_signal(..., data_transposed=True)
# Y = D @ X

Y, X, D = make_sparse_coded_signal(..., data_transposed=False)
# Y = X @ D

We would also need to update the documentation to describe the order change.

jeremiedbb · 2022-03-15T12:23:24Z

@g4brielvs I synced with main and directly pushed the changes we discussed.

@thomasjpfan I think it's ready for a final review.

thomasjpfan

This looks good to me. I think it would be good to have one more opinion for changing the order based on a kwarg.

jeremiedbb · 2022-03-18T14:51:44Z

I think it would be good to have one more opinion for changing the order based on a kwarg.

Note for future reviewer: this is temporary and will last until 1.3. After that there will be only 1 return statement as before.

ping @glemaitre, you assigned yourself to the issue, maybe you want to take a look here.

glemaitre

I find the swap code-dictionary quite surprising

glemaitre · 2022-03-21T13:57:21Z

+        Y = XD has shape (n_samples, n_features).
+
+        .. versionadded:: 1.1
+        .. deprecated:: 1.1


Is it deprecated or versionchanged here

Actually it's neither. It will be both in 1.3 when we do change the default and we deprecate this arg

glemaitre · 2022-03-21T14:06:05Z

    n_features=n_samples,
    n_nonzero_coefs=n_nonzero_coefs,
    random_state=0,
+    data_transposed=True,


Do we want to change the test to deal now with the change of shape?

I think we can leave that to when we deprecate the param

glemaitre · 2022-03-21T14:06:05Z

    n_features=n_samples,
    n_nonzero_coefs=n_nonzero_coefs,
    random_state=0,
+    data_transposed=True,


Do we want to change the test to deal now with the change of shape?

glemaitre

LGTM

reshamas · 2022-03-22T17:12:10Z

Congrats to @g4brielvs for completing this PR for the October sprint and everyone for reviewing it as well.
I will be updating our PR spreadsheet from the sprint. :)

jeremiedbb · 2022-03-22T17:58:20Z

Thanks @g4brielvs !

…nal (scikit-learn#21425) Co-authored-by: Andreas Mueller <[email protected]> Co-authored-by: jeremie du boisberranger <[email protected]> Co-authored-by: Jérémie du Boisberranger <[email protected]>

Add data_tranposed argument and warning to make_sparse_coded_signal

6f3753a

github-actions Bot added the module:datasets label Oct 23, 2021

amueller reviewed Oct 23, 2021

View reviewed changes

Comment thread sklearn/datasets/_samples_generator.py Outdated

Deprecate argument data_transposed and behave like True

ac261e4

reshamas added the Sprint label Oct 23, 2021

g4brielvs changed the title ~~Add data_tranposed argument and warning to make_sparse_coded_signal~~ Add data_transposed argument and warning to make_sparse_coded_signal Oct 23, 2021

amueller reviewed Oct 23, 2021

View reviewed changes

Comment thread sklearn/datasets/_samples_generator.py Outdated

Fix typo

91fbd43

Co-authored-by: Andreas Mueller <[email protected]>

g4brielvs marked this pull request as draft October 23, 2021 18:08

Add entry to the changelog

760a0fe

g4brielvs marked this pull request as ready for review October 23, 2021 18:12

amueller approved these changes Oct 23, 2021

View reviewed changes

Fix entry on the changelog

bdb8bd8

g4brielvs requested a review from amueller October 23, 2021 18:17

g4brielvs mentioned this pull request Oct 23, 2021

make_sparse_coded_signal returns data transposed #19894

Closed

jeremiedbb reviewed Oct 24, 2021

View reviewed changes

Comment thread sklearn/datasets/_samples_generator.py Outdated

Comment thread sklearn/datasets/_samples_generator.py Outdated

Transpose data, dict and code

b57c99f

g4brielvs marked this pull request as draft October 26, 2021 23:34

g4brielvs added 3 commits October 26, 2021 20:00

Include test for make_sparse_coded_signal

b8b4cd2

Merge remote-tracking branch 'upstream/main' into warn-make_sparse_co…

9edb9b6

…ded_signal

Formatting

928a5b3

g4brielvs marked this pull request as ready for review October 27, 2021 00:11

g4brielvs requested a review from jeremiedbb October 27, 2021 00:11

jeremiedbb reviewed Oct 27, 2021

View reviewed changes

Comment thread doc/whats_new/v1.1.rst Outdated

Comment thread sklearn/datasets/tests/test_samples_generator.py Outdated

Comment thread sklearn/datasets/tests/test_samples_generator.py Outdated

Comment thread sklearn/datasets/_samples_generator.py Outdated

g4brielvs added 2 commits February 15, 2022 17:24

Add minor edits

e9f8060

Merge branch 'main' into warn-make_sparse_coded_signal

fe21987

g4brielvs requested a review from jeremiedbb February 15, 2022 22:27

g4brielvs force-pushed the warn-make_sparse_coded_signal branch from 84b728d to 00b83c6 Compare February 15, 2022 22:40

Formatting with black==22.1.0

dd2841a

g4brielvs force-pushed the warn-make_sparse_coded_signal branch from 00b83c6 to dd2841a Compare February 15, 2022 22:43

thomasjpfan reviewed Feb 16, 2022

View reviewed changes

Comment thread doc/whats_new/v1.1.rst Outdated

Comment thread sklearn/datasets/_samples_generator.py

g4brielvs added 2 commits March 5, 2022 17:19

Merge branch 'main' of github.com:scikit-learn/scikit-learn into warn…

1f05642

…-make_sparse_coded_signal

Revert unrelated change

6ebbe3f

jeremiedbb added 2 commits March 15, 2022 12:51

Merge branch 'master' into pr/g4brielvs/21425

0a9dbbd

switch return order

ff7416a

cln

83b4f4f

jeremiedbb reviewed Mar 15, 2022

View reviewed changes

Comment thread sklearn/datasets/_samples_generator.py Outdated

Update sklearn/datasets/_samples_generator.py

23aa688

thomasjpfan reviewed Mar 18, 2022

View reviewed changes

Comment thread sklearn/datasets/_samples_generator.py Outdated

Comment thread sklearn/datasets/_samples_generator.py Outdated

Comment thread sklearn/datasets/_samples_generator.py Outdated

Comment thread sklearn/datasets/tests/test_samples_generator.py Outdated

thomasjpfan changed the title ~~Add data_transposed argument and warning to make_sparse_coded_signal~~ API Add data_transposed argument and warning to make_sparse_coded_signal Mar 18, 2022

address comments

3f31cf6

glemaitre reviewed Mar 21, 2022

View reviewed changes

jeremiedbb and others added 3 commits March 21, 2022 18:38

address comments, same return order

dee9195

Merge branch 'main' into warn-make_sparse_coded_signal

888da13

Update _samples_generator.py

5bd7044

glemaitre merged commit ffa0898 into scikit-learn:main Mar 22, 2022

glemaitre approved these changes Mar 22, 2022

View reviewed changes

Uh oh!

Uh oh!

Conversation

g4brielvs commented Oct 23, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Reference Issues/PRs

What does this implement/fix? Explain your changes.

Any other comments?

Uh oh!

g4brielvs commented Oct 23, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

g4brielvs commented Oct 23, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

amueller commented Oct 23, 2021

Uh oh!

Uh oh!

Uh oh!

amueller commented Oct 23, 2021

Uh oh!

jeremiedbb left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

reshamas commented Nov 5, 2021

Uh oh!

g4brielvs commented Nov 5, 2021

Uh oh!

g4brielvs commented Feb 15, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

thomasjpfan left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

thomasjpfan commented Mar 5, 2022

Uh oh!

jeremiedbb commented Mar 15, 2022

Uh oh!

Uh oh!

thomasjpfan left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jeremiedbb commented Mar 18, 2022

Uh oh!

glemaitre left a comment

Choose a reason for hiding this comment

Uh oh!

glemaitre Mar 21, 2022

Choose a reason for hiding this comment

Uh oh!

jeremiedbb Mar 21, 2022

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

glemaitre Mar 21, 2022

Choose a reason for hiding this comment

Uh oh!

jeremiedbb Mar 21, 2022

Choose a reason for hiding this comment

Uh oh!

glemaitre Mar 21, 2022

Choose a reason for hiding this comment

Uh oh!

glemaitre left a comment

g4brielvs commented Oct 23, 2021 •

edited

Loading

g4brielvs commented Oct 23, 2021 •

edited

Loading

g4brielvs commented Oct 23, 2021 •

edited

Loading

g4brielvs commented Feb 15, 2022 •

edited

Loading