FEA Group aware Time-based cross validation #16236

getgaurav2 · 2020-01-26T09:01:08Z

Reference Issues/PRs

What does this implement/fix? Explain your changes.

Fixes #14257
1 . Assumes that groups are contiguous .
2. Split the groups into train and test indices using TimeSeriesSplit
3. Use these group indices to get the indices for original data.
4. Use the max_train_size parameter to trim the train_array for each iteration of split.

Any other comments?

Question:

Should we have separate parameter for max_train_size in group vs. actual data ?
Should we have separate parameter for gap in group vs. actual data ?

jnothman · 2020-01-26T22:26:02Z

@getgaurav2, tests may be failing. Let us know when you want review.

jnothman · 2020-01-26T22:30:53Z

Linting is failing, actually
.

getgaurav2 · 2020-01-27T03:26:53Z

@getgaurav2, tests may be failing. Let us know when you want review.

Sure @jnothman . Will do . Thank you .

getgaurav2 · 2020-03-03T07:45:33Z

@jnothman @ogrisel Could you please opine on the approach that I have taken for this feature ?
I can work on performance metrics , documentation etc. if you are ok with the general direction .
Thanks

getgaurav2 · 2020-03-13T17:05:46Z

@jnothman @ogrisel Could you please opine on the approach that I have taken for this feature ? ...

@getgaurav2 - gentle reminder @ogrisel

_{Sent with GitHawk}

getgaurav2 · 2020-04-18T00:06:04Z

...

@ogrisel @jnothman - can you pls review this PR ? Thank you

_{Sent with GitHawk}

pip-wheel-metadata/scikit_learn.dist-info/COPYING

sklearn/model_selection/tests/test_split.py

sklearn/model_selection/_split.py

albertvillanova · 2020-08-31T09:48:11Z

@getgaurav2 are you still working on this PR?

getgaurav2 · 2020-08-31T11:27:30Z

@getgaurav2 are you still working on this PR?

@albertvillanova - yes . I would like to finish this please. Have been busy at work lately . Will find some time this week to make some progress . Thanks

_{Sent with GitHawk}

albertvillanova · 2020-09-05T10:07:26Z

@getgaurav2 Great! Because I do really need this feature in the next release ;)

PimwipaV · 2020-09-05T21:34:51Z

hi @getgaurav2, I tried pushing here but I got an error remote: Permission to getgaurav2/scikit-learn.git denied to Pimpwhippa. My git remote -v are correct I think. Please could you check if you allow edits to this PR. Thank you.

albertvillanova · 2020-09-06T09:10:20Z

@getgaurav2, as you have seen, it is not a good idea to rebase onto master during a Pull Request. If you would like to sync branches, it is better you make a merge:

git merge upstream master

Anyway, it does not seem necessary to do it: see GitHub message: This branch has no conflicts with the base branch.

getgaurav2 · 2020-09-06T21:34:24Z

hi @getgaurav2, I tried pushing here but I got an error remote: Permission to getgaurav2/scikit-learn.git denied to Pimpwhippa. My git remote -v are correct I think. Please could you check if you allow edits to this PR. Thank you.

Let's see if this helps. Please add test cases into test_split.py

getgaurav2 · 2020-09-07T09:22:40Z

@getgaurav2 Great! Because I do really need this feature in the next release ;)

@albertvillanova - would you want to check the latest code base ?
I want to talk through the best way to handle cases when the function may yield empty training splits .

_{Sent with GitHawk}

PimwipaV · 2020-09-07T15:15:55Z

@getgaurav2 thank you. I just opened a PR.

albertvillanova · 2020-09-08T11:53:07Z

Fast check: why this empty file? sklearn/model_selection/conftest.py

sklearn/model_selection/tests/test_split.py

albertvillanova · 2020-09-08T13:33:38Z

I would like to point out an issue in your implementation. I have created a new test case:

groups = np.array(['a', 'a', 'a', 'a', 'a', 'a', 'b', 'b', 'b', 'b', 'b', 'c', 'c', 'c', 'c', 'd', 'd', 'd'])
gtss = GroupTimeSeriesSplit(n_splits=3)

and this is what I get:

In [49]: for train_idx, test_idx in gtss.split(groups, groups=groups):
    ...:     print(train_idx, test_idx)
    ...:     print(groups[train_idx], groups[test_idx])
    ...:     print()
    ...:
[0 1 2 3 4 5] [6 7 8 9]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'b' 'b' 'b']

[0 1 2 3 4 5] [10 11 12 13]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'c' 'c' 'c']

[ 0  1  2  3  4  5  6  7  8  9 10] [14 15 16 17]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'd' 'd' 'd']

As you can see, there are gaps between the train and the test indices, i.e. they are not contiguous.

I would expect the following result:

In [49]: for train_idx, test_idx in gtss.split(groups, groups=groups):
    ...:     print(train_idx, test_idx)
    ...:     print(groups[train_idx], groups[test_idx])
    ...:     print()
    ...:
[0 1 2 3 4 5] [6 7 8 9 10]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'b' 'b' 'b' 'b']

[0 1 2 3 4 5 6 7 8 9 10] [11 12 13 14]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c']

[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14] [15 16 17]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b' 'c' 'c' 'c' 'c'] ['d' 'd' 'd']

By the way, and concerning the gaps, if you look at the next v0.24 release version of TimeSeriesSplit, there are two new parameters: test_size and gap. Maybe these should be also implemented in GroupTimeSeriesSplit.

PimwipaV · 2020-09-08T22:01:18Z

'In each split, test indices must be higher than before, and thus shuffling in cross validator is inappropriate'.
You made the class description clearer already but does that mean i can still shuffle the groups, but not the time right?

PimwipaV · 2020-09-08T22:42:07Z

i mean this is allowed
groups = ['B', 'D', 'D', 'C', 'C', 'A', 'B', 'A']
and not only this
groups = ['A', 'A', 'B', 'B', 'C', 'C', 'D', 'D']

getgaurav2 · 2020-10-03T20:02:47Z

I would like to point out an issue in your implementation. I have created a new test case:

groups = np.array(['a', 'a', 'a', 'a', 'a', 'a', 'b', 'b', 'b', 'b', 'b', 'c', 'c', 'c', 'c', 'd', 'd', 'd'])
gtss = GroupTimeSeriesSplit(n_splits=3)

and this is what I get:

In [49]: for train_idx, test_idx in gtss.split(groups, groups=groups):
    ...:     print(train_idx, test_idx)
    ...:     print(groups[train_idx], groups[test_idx])
    ...:     print()
    ...:
[0 1 2 3 4 5] [6 7 8 9]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'b' 'b' 'b']

[0 1 2 3 4 5] [10 11 12 13]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'c' 'c' 'c']

[ 0  1  2  3  4  5  6  7  8  9 10] [14 15 16 17]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'd' 'd' 'd']

As you can see, there are gaps between the train and the test indices, i.e. they are not contiguous.

I would expect the following result:

In [49]: for train_idx, test_idx in gtss.split(groups, groups=groups):
    ...:     print(train_idx, test_idx)
    ...:     print(groups[train_idx], groups[test_idx])
    ...:     print()
    ...:
[0 1 2 3 4 5] [6 7 8 9 10]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'b' 'b' 'b' 'b']

[0 1 2 3 4 5 6 7 8 9 10] [11 12 13 14]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c']

[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14] [15 16 17]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b' 'c' 'c' 'c' 'c'] ['d' 'd' 'd']

By the way, and concerning the gaps, if you look at the next v0.24 release version of TimeSeriesSplit, there are two new parameters: test_size and gap. Maybe these should be also implemented in GroupTimeSeriesSplit.

@albertvillanova - Thanks for your review. I have incorporated your feedback in the latest Commit . Can you please check again .

albertvillanova · 2020-10-04T12:34:46Z

@getgaurav2, thanks for your contribution. Your implementation now gives the expected result in the case I pointed out above.

However, when using the new parameters (gap, test_size) I think I have found some unexpected behavior (that I am commenting below), besides that they are not documented in the docstring. Moreover, I think there are some peculiarities when dealing with groups that might deserve a discussion to decide the expected behavior, the definition of the corresponding API and its implementation. For example, I am thinking of test_size, which in TimeSeriesSplit corresponds to the number of samples in the test set; however, when dealing with groups, maybe it would be more useful to be able to specify the number of groups in the test set instead.

Maybe, I would suggest, forgetting about this parameters for the moment, so that this PR can already be finished. You could eventually open a new PR to implement those parameters. What do you think?

The unexpected behavior for the parameter gap. Current behavior:

In [10]: gtss = GroupTimeSeriesSplit(n_splits=2, test_size=2, gap=1)

In [11]: for train_idx, test_idx in gtss.split(groups, groups=groups):
    ...:     print(train_idx, test_idx)
    ...:     print(groups[train_idx], groups[test_idx])
    ...:     print()
    ...:
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [11, 12]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c']

[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14] [15, 16]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b' 'c' 'c' 'c' 'c'] ['d' 'd']

As you can see, there is no gap between training and test sets.

The unexpected behavior for the parameter test_size. Current behavior:

In [8]: gtss = GroupTimeSeriesSplit(n_splits=2, test_size=5)

In [9]: for train_idx, test_idx in gtss.split(groups, groups=groups):
   ...:     print(train_idx, test_idx)
   ...:     print(groups[train_idx], groups[test_idx])
   ...:     print()
   ...:
scikit-learn\sklearn\model_selection\_split.py:2375: UserWarning: The size=4 of group=c is smaller than test_size=5.
  warnings.warn(("The size=%d of group=%s is smaller"
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [11, 12, 13, 14]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c']

scikit-learn\sklearn\model_selection\_split.py:2375: UserWarning: The size=3 of group=d is smaller than test_size=5.
  warnings.warn(("The size=%d of group=%s is smaller"
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14] [15, 16, 17]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b' 'c' 'c' 'c' 'c'] ['d' 'd' 'd']

In this case, I think the result should be:

[0 1 2 3 4 5] [6 7 8 9 10]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'b' 'b' 'b' 'b']

[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [11, 12, 13, 14, 15]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c', 'd']

albertvillanova · 2020-10-04T12:52:34Z

Maybe I would also add the test case I pointed out above:

In [47]: groups = np.array(['a', 'a', 'a', 'a', 'a', 'a', 'b', 'b', 'b', 'b', 'b', 'c', 'c', 'c', 'c', 'd', 'd', 'd'])
In [48]: gtss = GroupTimeSeriesSplit(n_splits=3)

In [49]: for train_idx, test_idx in gtss.split(groups, groups=groups):
    ...:     print(train_idx, test_idx)
    ...:     print(groups[train_idx], groups[test_idx])
    ...:     print()
    ...:
[0 1 2 3 4 5] [6 7 8 9 10]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'b' 'b' 'b' 'b']

[0 1 2 3 4 5 6 7 8 9 10] [11 12 13 14]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c']

[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14] [15 16 17]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b' 'c' 'c' 'c' 'c'] ['d' 'd' 'd']

sklearn/model_selection/_split.py

sklearn/model_selection/tests/test_split.py

getgaurav2 · 2020-10-05T07:13:54Z

@getgaurav2, thanks for your contribution. Your implementation now gives the expected result in the case I pointed out above.

However, when using the new parameters (gap, test_size) I think I have found some unexpected behavior (that I am commenting below), besides that they are not documented in the docstring. Moreover, I think there are some peculiarities when dealing with groups that might deserve a discussion to decide the expected behavior, the definition of the corresponding API and its implementation. For example, I am thinking of test_size, which in TimeSeriesSplit corresponds to the number of samples in the test set; however, when dealing with groups, maybe it would be more useful to be able to specify the number of groups in the test set instead.

Maybe, I would suggest, forgetting about this parameters for the moment, so that this PR can already be finished. You could eventually open a new PR to implement those parameters. What do you think?

The unexpected behavior for the parameter gap. Current behavior:
In [10]: gtss = GroupTimeSeriesSplit(n_splits=2, test_size=2, gap=1)

In [11]: for train_idx, test_idx in gtss.split(groups, groups=groups):
    ...:     print(train_idx, test_idx)
    ...:     print(groups[train_idx], groups[test_idx])
    ...:     print()
    ...:
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [11, 12]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c']

[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14] [15, 16]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b' 'c' 'c' 'c' 'c'] ['d' 'd']
As you can see, there is no gap between training and test sets.

The unexpected behavior for the parameter test_size. Current behavior:
In [8]: gtss = GroupTimeSeriesSplit(n_splits=2, test_size=5)

In [9]: for train_idx, test_idx in gtss.split(groups, groups=groups):
   ...:     print(train_idx, test_idx)
   ...:     print(groups[train_idx], groups[test_idx])
   ...:     print()
   ...:
scikit-learn\sklearn\model_selection\_split.py:2375: UserWarning: The size=4 of group=c is smaller than test_size=5.
  warnings.warn(("The size=%d of group=%s is smaller"
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [11, 12, 13, 14]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c']

scikit-learn\sklearn\model_selection\_split.py:2375: UserWarning: The size=3 of group=d is smaller than test_size=5.
  warnings.warn(("The size=%d of group=%s is smaller"
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14] [15, 16, 17]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b' 'c' 'c' 'c' 'c'] ['d' 'd' 'd']
In this case, I think the result should be:
[0 1 2 3 4 5] [6 7 8 9 10]
['a' 'a' 'a' 'a' 'a' 'a'] ['b' 'b' 'b' 'b' 'b']

[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [11, 12, 13, 14, 15]
['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c', 'd']

@albertvillanova - Thanks for your feedback. Can you please review the latest commit .

For the test_size = 5 , the expected output that you have put is certainly more desirable . It will be interesting to determine how the 2nd split shall behave when there is conflict between gap and test_size like :

groups = np.array(['a', 'a', 'a', 'a', 'a', 'a', 'b', 'b', 'b', 'b', 'b', 'c', 'c', 'c', 'c', 'd', 'd', 'd', 'd', 'd'])
gtss = GroupTimeSeriesSplit(n_splits=2, test_size=5)

> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [11, 12, 13, 14, 15]
> ['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['c' 'c' 'c' 'c', 'd']
OR 
> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] [15, 16, 17, 18, 19]
> ['a' 'a' 'a' 'a' 'a' 'a' 'b' 'b' 'b' 'b' 'b'] ['d' 'd' 'd' 'd', 'd']

Would you recommend that I open another PR here or raise a new feature request to keep things separate ?

albertvillanova · 2020-10-06T13:31:36Z

@getgaurav2, as you have already done an amazing amount of work here (thank you), my suggestion would be to leave for another PR the implementation of the parameters test_size and gap. That way, you would already be able to finish this PR.

Just a minor addition: could you please add GroupTimeSeriesSplit to the test test_yields_constant_splits?

And please, add a whatsnew entry to the file v0.24.rst, in the model_selection section, as a |MajorFeature|, and reference this PR and your name and link. You can take the other entries as example.

getgaurav2 · 2020-10-07T04:18:16Z

@getgaurav2, as you have already done an amazing amount of work here (thank you), my suggestion would be to leave for another PR the implementation of the parameters test_size and gap. That way, you would already be able to finish this PR.

Just a minor addition: could you please add GroupTimeSeriesSplit to the test test_yields_constant_splits?

And please, add a whatsnew entry to the file v0.24.rst, in the model_selection section, as a |MajorFeature|, and reference this PR and your name and link. You can take the other entries as example.

@albertvillanova - Thank you . Can you please check now .

albertvillanova · 2020-10-07T04:33:30Z

@getgaurav2 thanks. I think your PR is ready for a review round. Could you please change its name from WIP to MRG?

cmarmo

Thanks @getgaurav2 for your patience.
Those two comments should fix the documentation error and rendering.

sklearn/model_selection/_split.py

getgaurav2 · 2022-07-16T22:07:30Z

Also the diagram for the 2 time-aware splitters have time-order classes. This is very likely to never happen in a practical setting and would prevent any machine learning model to behave correctly on such a task and is therefore confusing to included in the documentation.

Therefore I would advise to randomly scatter the samples of the 3 classes across the samples, at least for those 2 splitters.

@ogrisel - Do the attached plots match what you have in mind?

getgaurav2 · 2022-07-31T04:44:23Z

I think that we could an additional test with a bit more groups and imbalanced size such that we can make sure of the behaviour that I describe in one of the comments (equal test size/unequal group size vs. unequel test size/equal group size)

Added a test case with imbalanced groups . Imbalanced splits still pass the test as of now . I can change that if we agree on a function to that will give "closest group-aligned split" ( comment )

doc/modules/cross_validation.rst

sklearn/model_selection/_split.py

jnothman

The tests are currently quite hard to read. Tests need to be easy to read and understand that they are testing what we believe to be true.

sklearn/model_selection/_split.py

jnothman · 2022-12-28T12:49:05Z

sklearn/model_selection/_split.py

+        for idx in np.arange(n_samples):
+            if groups[idx] in group_dict:
+                if idx - group_dict[groups[idx]][-1] == 1:
+                    group_dict[groups[idx]].append(idx)


we should not need to store full lists... I think this is sufficient:

seen_groups = set() prev_group = None for group in groups: if group != prev_group and group in seen_groups: raise ValueError("The groups should be contiguous") prev_group = group

Alternatively, we could do:

reordered_unique_groups, indices, inverse = np.unique(groups, return_index=True, return_inverse=True) if (np.diff(inverse) < 1).any(): raise ValueError("The groups should be contiguous")

although this is much more obfuscated!

Used the first variant with enumerate to add the index in the error msg.

jnothman · 2022-12-28T12:53:48Z

sklearn/model_selection/tests/test_split.py

+    assert_array_equal(groups[test], ["d", "d", "d"])
+
+
+def test_group_time_series_non_overlap_group_2():


What is the benefit of having each of these test cases? If they test distinct capabilities, consider using pytest.parametrize and expressing your test in a more generalised way that tests the invariances of interest.

I have renamed the test cases to convey the purpose of each. Also , removed one of the tests that was kind of repetitive. Please let me know if you think it needs further clarification.

adrinjalali · 2024-03-06T09:27:22Z

@glemaitre this seems like a good candidate for a "time series related" "project board".

sluofoss · 2024-09-26T13:16:58Z

Hi, can I check what is the status of this PR?

Would I be correct that this PR would also close issue #6322 and related old PR #6351 's HeteroTSCV.?

Also I feel like a gap functionality that is similar to timegapsplit #13761 and the current gap parameter in https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.TimeSeriesSplit.html would be useful when it comes to chronological groups.

adrinjalali · 2024-09-26T13:47:28Z

@lu0x1a0 this is up for grabs. Now that we have metadata routing, this can nicely move forward if you're up for it.

getgaurav2 · 2024-09-26T16:41:43Z

@adrinjalali , @lu0x1a0 -- I would love to continue work on it to see this one to completion if that's okay .

adrinjalali · 2024-09-26T20:55:31Z

@getgaurav2 that'd be fantastic!

getgaurav2 · 2025-01-21T05:57:18Z

@getgaurav2 that'd be fantastic!

@adrinjalali - would you please be able to check the issue in doc build to and give some direction ....thank you !

adrinjalali · 2025-02-18T17:30:43Z

Something has gone wrong here probably in your merge with main. You have over 12k line change, and some large changes in the rst files. I'd start with fixing those issues if I were you.

getgaurav2 · 2025-07-30T05:59:43Z

Something has gone wrong here probably in your merge with main. You have over 12k line change, and some large changes in the rst files. I'd start with fixing those issues if I were you.

@adrinjalali - sorry for the long gap . Would you pls be able to review now ?

getgaurav2 · 2025-07-30T06:01:36Z

@glemaitre this seems like a good candidate for a "time series related" "project board".

@adrinjalali - Would you pls be able to share a link if this is public ..Thank you !

adrinjalali · 2025-08-04T14:50:53Z

sklearn/model_selection/tests/test_split.py

+        train,
+        np.array(
+            [
+                0,


in all these cases, we don't have to write out each individual case, you can use functions such as range, etc to make these much smaller diffs.

Sure ..fixed now.

adrinjalali · 2025-08-05T08:43:40Z

sklearn/model_selection/_split.py

+        # `np.unique` will reorder the group. We need to keep the original
+        # ordering.
+        reordered_unique_groups, indices = np.unique(groups, return_index=True)
+        unique_groups = reordered_unique_groups[np.argsort(indices)]


why do we care about the order? groups doesn't have any ordered semantics.

why do we care about the order? groups doesn't have any ordered semantics.

This is done to preserve the sequence in which the labels of the group appear in the dataset. The split logic will cut the list in the same order as the group label appear in the dataset.

PimwipaV mentioned this pull request Feb 17, 2020

[WIP] Fix #14257 (tests only) for GroupTimeSeriesSplit #16464

Closed

github-actions bot added the module:model_selection label Mar 2, 2020

jnothman reviewed May 20, 2020

View reviewed changes

albertvillanova reviewed Sep 8, 2020

View reviewed changes

sklearn/model_selection/tests/test_split.py Outdated Show resolved Hide resolved

albertvillanova reviewed Oct 4, 2020

View reviewed changes

thomasjpfan mentioned this pull request Jun 21, 2022

GroupShuffleSplit with optional shuffle #18147

Closed

cmarmo removed the Waiting for Reviewer label Jul 3, 2022

cmarmo reviewed Jul 10, 2022

View reviewed changes

sklearn/model_selection/_split.py Outdated Show resolved Hide resolved

sklearn/model_selection/_split.py Outdated Show resolved Hide resolved

cmarmo added the Waiting for Reviewer label Jul 11, 2022

he7d3r reviewed Oct 16, 2022

View reviewed changes

doc/modules/cross_validation.rst Outdated Show resolved Hide resolved

he7d3r reviewed Oct 16, 2022

View reviewed changes

sklearn/model_selection/_split.py Outdated Show resolved Hide resolved

glemaitre self-requested a review October 17, 2022 12:39

jnothman requested changes Dec 28, 2022

View reviewed changes

getgaurav2 requested a review from jnothman July 20, 2023 15:21

getgaurav2 mentioned this pull request Nov 16, 2024

Cython error while installing development version on MacOS M2 chip #30286

Closed

Clean Commit

7686284

getgaurav2 force-pushed the issue_14257 branch from e918480 to 7686284 Compare July 29, 2025 21:25

getgaurav2 added 4 commits July 29, 2025 18:02

fix linting error with unused import

0d3208f

Add change log

dbe4502

plot grouptimeseries split

a83f569

linting error

39b6466

adrinjalali reviewed Aug 5, 2025

View reviewed changes

getgaurav2 added 2 commits August 7, 2025 19:16

using reange and repeat to condense code

40b6ae8

ruff formatting

272853c

		assert_array_equal(groups[test], ["d", "d", "d"])


		def test_group_time_series_non_overlap_group_2():

Uh oh!

FEA Group aware Time-based cross validation #16236

Are you sure you want to change the base?

FEA Group aware Time-based cross validation #16236

Uh oh!

Conversation

getgaurav2 commented Jan 26, 2020 • 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!

jnothman commented Jan 26, 2020

Uh oh!

jnothman commented Jan 26, 2020

Uh oh!

getgaurav2 commented Jan 27, 2020

Uh oh!

getgaurav2 commented Mar 3, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

getgaurav2 commented Mar 13, 2020

Uh oh!

getgaurav2 commented Apr 18, 2020

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

albertvillanova commented Aug 31, 2020

Uh oh!

getgaurav2 commented Aug 31, 2020

Uh oh!

albertvillanova commented Sep 5, 2020

Uh oh!

PimwipaV commented Sep 5, 2020

Uh oh!

albertvillanova commented Sep 6, 2020

Uh oh!

getgaurav2 commented Sep 6, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

getgaurav2 commented Sep 7, 2020

Uh oh!

PimwipaV commented Sep 7, 2020

Uh oh!

albertvillanova commented Sep 8, 2020

Uh oh!

Uh oh!

albertvillanova commented Sep 8, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

PimwipaV commented Sep 8, 2020

Uh oh!

PimwipaV commented Sep 8, 2020

Uh oh!

getgaurav2 commented Oct 3, 2020

Uh oh!

albertvillanova commented Oct 4, 2020

Uh oh!

albertvillanova commented Oct 4, 2020

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

getgaurav2 commented Oct 5, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

albertvillanova commented Oct 6, 2020

Uh oh!

getgaurav2 commented Oct 7, 2020

Uh oh!

albertvillanova commented Oct 7, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cmarmo left a comment

Choose a reason for hiding this comment

Uh oh!

getgaurav2 commented Jan 26, 2020 •

edited

Loading

getgaurav2 commented Mar 3, 2020 •

edited

Loading

getgaurav2 commented Sep 6, 2020 •

edited

Loading

albertvillanova commented Sep 8, 2020 •

edited

Loading

getgaurav2 commented Oct 5, 2020 •

edited

Loading

albertvillanova commented Oct 7, 2020 •

edited

Loading

getgaurav2 commented Jul 16, 2022 •

edited

Loading

getgaurav2 commented Jul 31, 2022 •

edited

Loading

getgaurav2 Jan 1, 2023 •

edited

Loading

sluofoss commented Sep 26, 2024 •

edited

Loading

getgaurav2 commented Sep 26, 2024 •

edited

Loading

getgaurav2 commented Jan 21, 2025 •

edited

Loading