Added sample weight handling to BinMapper under HGBT #29641

snath-xoc · 2024-08-08T21:46:57Z

Fixes #29640 . See also #27117.

Calls sample weight within BinMapper and passes it on to _find_binning_thresholds where bin midpoints are calculated using weighted percentile. Sample weights also passed to rng.choice() subsampling in BinMapper for samples larger than 2e5

NOTE: when the n_bins<discrete_values the best workaround was to set the bins as the midpoints. In future, it may be worth getting rid of this altogether, however at the risk of getting inhomogeneous array from weighted_percentile. We will need to agree on the best methods of trimming.

Major changes proposed:

Subsampling with weights after which sample_weight is not propagated in _BinMapper (set to None)
Tests for sample_weight invariance for _BinMapper under deterministic case added
Some tests failed with the new changes and n_samples had to be increased e.g., test_interaction_cst_numerically
Some tests failed under special cases when sample weights are passed through but distinct values<max_bins. To be checked what to do
Statistical test added in comments below

NOTE: this also allows HGBT to pass weighted tests for more than 256 samples (but still less than 2e6)

TO DO:

Update tests under test_gradient_boosting so that sample weight is passed as positional argument
Fix test failures, something fishy changes such that e.g., test_interaction_cst_numerically is not working
Modified tests to have larger number of samples otherwise expected results are not attaines (i.e., r2 value threshold etc. is not reached).... was expecting this to be the other way so not sure if I set off another bug.
When distinct values are less than max_bins and sample weights are given we get a discrepancy between weighted and repeated under special cases (see test_zero_sample_weights_classification under test_gradient_boosting.py). Need to see what to do here.
Further test failure when max_bins are specified as > distinct values under weighted and repeated cases
Check that sample weight invariance holds under deterministic case (test added under test_binning)
For the case n_samples > subsample conduct a statistical analysis to check for that the repeated/reweighted equivalence holds for the binning procedure in expectation, similar to issue Incorrect sample weight handling in KBinsDiscretizer #29906. Conduct analysis for many different values random_state for match of the mean bin edges.

github-actions · 2024-08-08T21:48:23Z

✔️ Linting Passed

All linting checks passed. Your pull request is in excellent shape! ☀️

_{Generated for commit: 39ed7f3. Link to the linter CI: here}

ogrisel

I did not analyse the test failures yet but here is some early feedback.

For the case n_samples > subsample we would need to conduct a statistical analysis to check for that the repeated/reweighted equivalence holds for the binning procedure in expectation.

Once the tests pass for the deterministic case, we should conduct such an analysis (e.g. using a side notebook, not included in the repo, where we rerun the binning for many different values random_state and then check for match of the mean bin edges).

Please add a TODO item to the description of this PR not to forget about this.

sklearn/ensemble/_hist_gradient_boosting/binning.py

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py

sklearn/ensemble/_hist_gradient_boosting/tests/test_gradient_boosting.py

…oosting.py Co-authored-by: Olivier Grisel <[email protected]>

snath-xoc · 2025-03-21T13:29:43Z

Statistical tests were conducted for n_samples>subsample (subsample=int(2e5)) using the following code:

import numpy as np
from sklearn.ensemble._hist_gradient_boosting.binning import _BinMapper
from scipy.stats import kstest
import matplotlib.pyplot as plt

BONFERRONI_CORRECTION = 1 #To adjust later according  to agreed test dim.

rng = np.random.RandomState(0)

n_samples = int(3e5)
X = rng.randint(0, 30, size=(n_samples,3))
# Use random integers (including zero) as weights.
sw = rng.randint(0, 5, size=n_samples)

X_repeated = np.repeat(X, sw,axis=0)
assert len(X_repeated)>int(2e5)

bin_thresholds_weighted=[]
bin_thresholds_repeated = []
for seed in np.arange(100):
    est_weighted = _BinMapper(n_bins=6, random_state=seed).fit(
                X, sample_weight=sw
            )

    est_repeated = _BinMapper(n_bins=6, random_state=seed+500).fit(
                X_repeated, sample_weight=None)

    bin_thresholds_weighted.append(est_weighted.bin_thresholds_)
    bin_thresholds_repeated.append(est_repeated.bin_thresholds_)

bin_thresholds_weighted = np.asarray(bin_thresholds_weighted)
bin_thresholds_repeated = np.asarray(bin_thresholds_repeated)

fig,axs = plt.subplots(3,4,figsize=(14,12))

j=0
for i,ax in enumerate(axs.flatten()):
    if i>0 and i%4==0:
        j+=1
    ax.hist(bin_thresholds_weighted[:,j,i%4].flatten())
    ax.hist(bin_thresholds_repeated[:,j,i%4].flatten(),alpha=0.5)
    pval = kstest(bin_thresholds_weighted[:,j,i%4].flatten(),
                                    bin_thresholds_repeated[:,j,i%4].flatten()).pvalue
    if pval<(0.05*BONFERRONI_CORRECTION):
        ax.set_title(f'p-value: {pval:.4f},failed')
    else:
        ax.set_title(f'p-value: {pval:.4f},passed')

The output is as follows:

snath-xoc · 2025-03-21T13:40:08Z

The ARM lint test is still failing due to the test_find_binning_thresholds_small_regular_data assertion error but I can't reproduce it locally.

ogrisel

Modified tests to have larger number of samples otherwise expected results are not attaines (i.e., r2 value threshold etc. is not reached).... was expecting this to be the other way so not sure if I set off another bug.

This is annoying, but it's probably caused by the fact that we switched from np.percentile(..., method="linear") to np.percentile(..., method="averaged_inverted_cdf") in the sample_weight=None case. Assuming this change is really the cause and since it's necessary to fix the weight/repetition equivalence, we might consider this behavior change as a bug fix. It should be clearly documented as such in the changelog, possibly with a dedicated note in the "Changed models" section of the release notes to better warn the users about this change.

Another concern is that _averaged_weighted_percentile implementation is currently very naive and will cause a performance regression whenever the users passes sample_weight != None. We might want to postpone the final merge of this PR to wait for the review and merge of #30945 first.

ogrisel · 2025-03-21T14:07:41Z

sklearn/ensemble/_hist_gradient_boosting/binning.py

+        func = _averaged_weighted_percentile
+        if len(distinct_values) <= max_bins:
+            max_bins = len(distinct_values)
+            func = _weighted_percentile


Why would _averaged_weighted_percentile not work when len(distinct_values) <= max_bins?

>>> import numpy as np >>> from sklearn.utils.stats import _weighted_percentile, _averaged_weighted_percentile >>> a = np.random.randint(0, 10, size=10000).astype(np.float64) >>> max_bins = 30 >>> max_bins > np.unique(a).shape[0] True >>> np.array( ... [_averaged_weighted_percentile(a, np.ones_like(a), percentile) for percentile in percentiles] ... ) array([0., 0., 0., 1., 1., 1., 1., 2., 2., 2., 3., 3., 3., 4., 4., 4., 5., 5., 5., 6., 6., 6., 7., 7., 7., 8., 8., 8., 9., 9., 9.]) >>> np.array( ... [_weighted_percentile(a, np.ones_like(a), percentile) for percentile in percentiles] ... ) array([0., 0., 0., 1., 1., 1., 1., 2., 2., 2., 3., 3., 3., 4., 4., 4., 5., 5., 5., 6., 6., 6., 7., 7., 7., 8., 8., 8., 9., 9., 9.])

When distinct values are less than max_bins and sample weights are given we get a discrepancy between weighted and repeated under special cases (see test_zero_sample_weights_classification under test_gradient_boosting.py). Need to see what to do here.

The manual midpoints between distinct values in the weighted case is probably responsible for this discrepancy.

We should probably explore if it's possible to post process the output of _averaged_weighted_percentile to trim duplicated thresholds. If the number of trimmed thresholds is lower than max_bins then it might be possible to further post-process to recover thresholds that match the unweighted case by shifting the thresholds to use 0.5 midpoints.

sklearn/ensemble/_hist_gradient_boosting/binning.py

sklearn/ensemble/_hist_gradient_boosting/tests/test_gradient_boosting.py

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py

sklearn/ensemble/_hist_gradient_boosting/binning.py

sklearn/ensemble/_hist_gradient_boosting/tests/test_gradient_boosting.py

Co-authored-by: Olivier Grisel <[email protected]>

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py

Co-authored-by: Olivier Grisel <[email protected]>

sklearn/ensemble/_hist_gradient_boosting/binning.py

snath-xoc · 2025-04-25T19:46:11Z

I can reproduce the test_sample_weight_effect failure locally, it only happens for specific seeds though e.g., glbal_random_seed=3. This may be due to some larger issue, but it is the remaining test failure @ogrisel and @antoinebaker if you have any opinions?

Never mind: it seems finicky to sample size, I have lowered it to 200 now.

snath-xoc · 2025-06-13T10:28:02Z

@ogrisel reviving this, i added a changelog, let me know what you think!

ogrisel

Here is a new pass of review:

sklearn/ensemble/_hist_gradient_boosting/binning.py

ogrisel · 2025-07-02T14:56:12Z

sklearn/ensemble/_hist_gradient_boosting/binning.py

+        # Remove duplicated midpoints if they exist and shift
+        unique_bin_values = np.unique(bin_thresholds)
+        if unique_bin_values.shape[0] != bin_thresholds.shape[0]:
+            bin_thresholds = unique_bin_values


Let's try to craft a test case to cover this line. I think the following might help trigger this case:

import numpy as np >>> col_data = np.asarray(([1] * 1000) + [2, 3, 4, 5, 6]) >>> col_data.shape (1005,) >>> np.unique(col_data).shape (6,) >>> max_bins = 4 >>> percentiles = np.linspace(0, 100, num=max_bins + 1)[1:-1] >>> np.percentile(col_data, percentiles) array([1., 1., 1.])

Then the test could also check that the result matches what we get from:

>>> col_data = np.asarray(([1, 2, 3, 4, 5, 6]) >>> sample_weight = np.asarray(([1000, 1, 1, 1, 1, 1])

sklearn/ensemble/_hist_gradient_boosting/binning.py

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py

sklearn/ensemble/_hist_gradient_boosting/binning.py

ogrisel · 2025-07-02T16:02:31Z

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py

@@ -198,6 +198,27 @@ def test_bin_mapper_repeated_values_invariance(n_distinct):
    assert_array_equal(binned_1, binned_2)


+@pytest.mark.parametrize("n_bins", [50, None])
+def test_binmapper_weighted_vs_repeated_equivalence(global_random_seed, n_bins):


I found an occurrence of a test failure when global_random_seed=63 when running:

SKLEARN_TESTS_GLOBAL_RANDOM_SEED="all" pytest -vl sklearn/ensemble/_hist_gradient_boosting/tests -k test_binmapper_weighted_vs_repeated_equivalence

Can you reproduce?

Here is the error message I get:

=========================================================================== test session starts ============================================================================ platform darwin -- Python 3.13.3, pytest-8.4.0, pluggy-1.6.0 -- /Users/ogrisel/miniforge3/envs/dev/bin/python3.13 cachedir: .pytest_cache rootdir: /Users/ogrisel/code/scikit-learn configfile: pyproject.toml plugins: xdist-3.7.0, run-parallel-0.4.3, anyio-4.9.0 collected 451 items / 449 deselected / 2 selected Collected 0 items to run in parallel sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py::test_binmapper_weighted_vs_repeated_equivalence[63-50] FAILED [ 50%] sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py::test_binmapper_weighted_vs_repeated_equivalence[63-None] PASSED [100%] ================================================================================= FAILURES ================================================================================= __________________________________________________________ test_binmapper_weighted_vs_repeated_equivalence[63-50] __________________________________________________________ global_random_seed = 63, n_bins = 50 @pytest.mark.parametrize("n_bins", [50, None]) def test_binmapper_weighted_vs_repeated_equivalence(global_random_seed, n_bins): rng = np.random.RandomState(global_random_seed) n_samples = 200 X = rng.randn(n_samples, 3) if n_bins is None: n_bins = np.unique(X[:, rng.randint(3)]).shape[0] + rng.randint(5) + 1 sw = rng.randint(0, 5, size=n_samples) X_repeated = np.repeat(X, sw, axis=0) est_weighted = _BinMapper(n_bins=n_bins).fit(X, sample_weight=sw) est_repeated = _BinMapper(n_bins=n_bins).fit(X_repeated, sample_weight=None) > assert_allclose(est_weighted.bin_thresholds_, est_repeated.bin_thresholds_) E AssertionError: E Not equal to tolerance rtol=1e-07, atol=0 E E Mismatched elements: 6 / 144 (4.17%) E Max absolute difference among violations: 0.022745 E Max relative difference among violations: 0.02664642 E ACTUAL: array([[-1.712595, -1.285816, -1.226962, -1.009446, -0.960101, -0.882069, E -0.83596 , -0.69571 , -0.675828, -0.603487, -0.399731, -0.361047, E -0.340069, -0.237937, -0.207451, -0.087749, -0.075609, 0.032785,... E DESIRED: array([[-1.712595, -1.285816, -1.226962, -1.009446, -0.960101, -0.882069, E -0.83596 , -0.69571 , -0.675828, -0.593016, -0.399731, -0.361047, E -0.340069, -0.237937, -0.207451, -0.087749, -0.075609, 0.032785,... X = array([[-2.13897865e+00, 1.11206124e+00, 3.58015526e-02], [-6.30157742e-01, 3.54051160e-05, -1.20895742e+00]... [-3.99730932e-01, 9.48906177e-01, 2.40632339e-01], [-3.83554471e-02, -1.19674458e+00, -1.27746484e+00]]) X_repeated = array([[-2.13897865, 1.11206124, 0.03580155], [ 0.29162807, -1.65065515, -1.50712909], [ 0.29162807, -...3234], [-0.39973093, 0.94890618, 0.24063234], [-0.39973093, 0.94890618, 0.24063234]], shape=(392, 3)) est_repeated = _BinMapper(n_bins=50) est_weighted = _BinMapper(n_bins=50) global_random_seed = 63 n_bins = 50 n_samples = 200 rng = RandomState(MT19937) at 0x13D5D3240 sw = array([1, 0, 4, 2, 4, 0, 4, 0, 4, 0, 4, 0, 2, 1, 3, 4, 4, 3, 0, 0, 3, 1, 0, 4, 2, 2, 0, 1, 2, 2, 0, 4, 3, 4, 0,...4, 4, 1, 3, 2, 4, 2, 2, 0, 4, 0, 2, 1, 0, 2, 2, 0, 0, 4, 3, 1, 4, 1, 0, 0, 0, 2, 4, 4, 3, 3, 2, 2, 4, 0]) sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py:215: AssertionError -------------------------------------------------------------------------- Captured stdout setup --------------------------------------------------------------------------- I: Seeding RNGs with 819376607 ========================================================================= short test summary info ========================================================================== FAILED sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py::test_binmapper_weighted_vs_repeated_equivalence[63-50] - AssertionError: =============================================================== 1 failed, 1 passed, 449 deselected in 0.46s ================================================================

So the problem happens when max_bins=50 < np.unique(X[sw != 0], axis=0).shape[0]=155, which means that it's not related to the branch of the code that uses midpoints as bin edges but rather uses the branch of the code that uses the _averaged_weighted_percentile. I think we try to extract a minimal reproducer to understand the root cause of this failure case.

Maybe the root cause is a bug in _averaged_weighted_percentile that does not implement the expected weighted/repeated equivalence? If that the case we should try to extract a minimal reproducer for that function in isolation and debug this and fix this in a dedicated PR.

I can confirm that I get the same failure for seed 63, will investigate a bit more, strange :/

sklearn/ensemble/_hist_gradient_boosting/gradient_boosting.py

antoinebaker

Here a first round of review !

antoinebaker · 2025-07-02T15:10:13Z

sklearn/ensemble/_hist_gradient_boosting/gradient_boosting.py

+            self._bin_mapper = self._bin_mapper.fit(
+                X, sample_weight=sample_weight
+            )  # F-aligned array
+            X_binned = self._bin_mapper.transform(X)


Suggested change

self._bin_mapper = self._bin_mapper.fit(

X, sample_weight=sample_weight

) # F-aligned array

X_binned = self._bin_mapper.transform(X)

self._bin_mapper.fit(X, sample_weight=sample_weight)

X_binned = self._bin_mapper.transform(X) # F-aligned array

sklearn/ensemble/_hist_gradient_boosting/binning.py

antoinebaker · 2025-07-02T15:45:24Z

sklearn/ensemble/_hist_gradient_boosting/binning.py

+    # if sample weight is not None and null values exist
+    # we need to remove those before calculating the
+    # distinct points
+    if sample_weight is not None:
+        col_data_non_null = col_data[sample_weight != 0]
+        distinct_values = np.unique(col_data_non_null).astype(X_DTYPE)
+    else:
+        distinct_values = np.unique(col_data).astype(X_DTYPE)


Suggested change

# if sample weight is not None and null values exist

# we need to remove those before calculating the

# distinct points

if sample_weight is not None:

col_data_non_null = col_data[sample_weight != 0]

distinct_values = np.unique(col_data_non_null).astype(X_DTYPE)

else:

distinct_values = np.unique(col_data).astype(X_DTYPE)

if sample_weight is not None:

# A zero sample_weight should be equivalent to removing the sample.

# We discard sample_weight=0 when computing the distinct values.

distinct_values = np.unique(col_data[sample_weight != 0]).astype(X_DTYPE)

else:

distinct_values = np.unique(col_data).astype(X_DTYPE)

antoinebaker · 2025-07-02T15:51:44Z

sklearn/ensemble/_hist_gradient_boosting/binning.py

        # We could compute approximate midpoint percentiles using the output of
        # np.unique(col_data, return_counts) instead but this is more
        # work and the performance benefit will be limited because we
        # work on a fixed-size subsample of the full data.


I think this comment belongs to the elif sample_weight is None: block above.

Besides, maybe the comment should be updated (midpoints are now replaced with the averaged_inverted_cdf method) ?

I updated this comment now but not sure if it is what you had in mind, feel free to check and provide suggestions!

antoinebaker · 2025-07-02T15:58:02Z

sklearn/ensemble/_hist_gradient_boosting/binning.py

        percentiles = np.linspace(0, 100, num=max_bins + 1)
        percentiles = percentiles[1:-1]
-        midpoints = np.percentile(col_data, percentiles, method="midpoint").astype(
-            X_DTYPE
+        sample_weight = sample_weight[~missing_mask]


I think we should do this in the if missing_mask.any(): above to make sure col_data and sample_weight are always aligned.

Gives a bit of a spaghetti if but am O.K. with it for now

Let me know if the fix makes sense for you now!

sklearn/ensemble/_hist_gradient_boosting/binning.py

ogrisel · 2025-07-02T16:25:34Z

BTW, I re-ran #29641 (comment) on the current state of the PR and the results are good for the subsampling branch of the code:

Co-authored-by: Olivier Grisel <[email protected]>

Co-authored-by: antoinebaker <[email protected]>

snath-xoc added 4 commits August 2, 2024 15:52

add weighted percentile to binning threshold calculation in HGBT

43da2a6

add weighted percentile to binning threshold calculation in HGBT

b7b1ce2

add sample_weight pass through to binning

d21a19c

fix _find_binning_thresholds and BinMapper sample weight handling

6b3f8ea

github-actions bot added the module:ensemble label Aug 8, 2024

snath-xoc added 4 commits August 8, 2024 17:49

minor fix

cf1149b

updated sampe weight handling in test_binning

fbd0365

updated sampe weight handling in test_binning

22c4c11

[all random seeds]

90665f1

ogrisel reviewed Aug 12, 2024

View reviewed changes

ogrisel mentioned this pull request Oct 25, 2024

List of estimators with known incorrect handling of sample_weight #16298

Open

54 tasks

snath-xoc and others added 11 commits March 13, 2025 21:43

Merge branch 'main' into HGBT_bin_weights

da8344e

Update sklearn/ensemble/_hist_gradient_boosting/tests/test_gradient_b…

00ead80

…oosting.py Co-authored-by: Olivier Grisel <[email protected]>

fix _find_binning_thresholds and BinMapper sample weight handling

2d74df6

[all random seeds]

398598e

comment

8d9544f

add None handling to _find_binning_threshold

3ffcbca

change sampling prob to None when no sample weights passed

0758886

fix distinct value handling and modify tests for convergence

4f58074

Merge branch 'main' into HGBT_bin_weights

b7ce7a9

update tests

664d9e4

Merge branch 'main' into HGBT_bin_weights

47e6b65

snath-xoc marked this pull request as ready for review March 21, 2025 13:32

ogrisel reviewed Mar 21, 2025

View reviewed changes

sklearn/ensemble/_hist_gradient_boosting/binning.py Show resolved Hide resolved

ogrisel reviewed Mar 21, 2025

View reviewed changes

sklearn/ensemble/_hist_gradient_boosting/binning.py Outdated Show resolved Hide resolved

ogrisel reviewed Mar 21, 2025

View reviewed changes

sklearn/ensemble/_hist_gradient_boosting/tests/test_gradient_boosting.py Outdated Show resolved Hide resolved

snath-xoc and others added 3 commits March 28, 2025 18:38

Apply suggestions from code review

b790d11

Co-authored-by: Olivier Grisel <[email protected]>

change midpoints name and handling of distinc_values<max_bins

e255e46

Merge branch 'main' into HGBT_bin_weights

2584e43

ogrisel reviewed Apr 10, 2025

View reviewed changes

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py Outdated Show resolved Hide resolved

ogrisel reviewed Apr 10, 2025

View reviewed changes

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py Outdated Show resolved Hide resolved

sklearn/ensemble/_hist_gradient_boosting/tests/test_binning.py Outdated Show resolved Hide resolved

snath-xoc and others added 5 commits April 10, 2025 16:09

Apply suggestions from code review

8573b61

Co-authored-by: Olivier Grisel <[email protected]>

Merge branch 'main' into HGBT_bin_weights

0b24d8a

remove redundant comment

d3d7ae8

add handling of null waited values for distinct value edge case

2935cce

Merge branch 'main' into HGBT_bin_weights

759cf12

snath-xoc commented Apr 10, 2025

View reviewed changes

sklearn/ensemble/_hist_gradient_boosting/binning.py Outdated Show resolved Hide resolved

snath-xoc commented Apr 10, 2025

View reviewed changes

sklearn/ensemble/_hist_gradient_boosting/binning.py Show resolved Hide resolved

snath-xoc and others added 3 commits April 10, 2025 20:09

Apply suggestions from code review

34b97f4

[all random seeds] test_sample_weight_effect

14180d2

Merge branch 'main' into HGBT_bin_weights

fc7556f

snath-xoc and others added 5 commits April 25, 2025 20:55

fix sample size in test_sample_weight_effect

546648f

fix documentation

2dc5e0a

Merge branch 'main' into HGBT_bin_weights

2966b2b

Merge branch 'main' into HGBT_bin_weights

00235d8

add changelog

955eb43

ogrisel reviewed Jul 2, 2025

View reviewed changes

antoinebaker reviewed Jul 2, 2025

View reviewed changes

snath-xoc and others added 5 commits July 4, 2025 18:16

Apply suggestions from code review

7064bde

Co-authored-by: Olivier Grisel <[email protected]>

Apply suggestions from code review

b43232b

Co-authored-by: antoinebaker <[email protected]>

Merge branch 'main' into HGBT_bin_weights

13c5bd1

implement reviews

f7b1785

add test_unique_bins_repeated_weighted

39ed7f3

Uh oh!

Added sample weight handling to BinMapper under HGBT #29641

Are you sure you want to change the base?

Added sample weight handling to BinMapper under HGBT #29641

Uh oh!

Conversation

snath-xoc commented Aug 8, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Aug 8, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✔️ Linting Passed

Uh oh!

ogrisel left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

snath-xoc commented Mar 21, 2025

Uh oh!

snath-xoc commented Mar 21, 2025

Uh oh!

ogrisel left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

snath-xoc commented Apr 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

snath-xoc commented Jun 13, 2025

Uh oh!

ogrisel left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

antoinebaker left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

snath-xoc commented Aug 8, 2024 •

edited

Loading

github-actions bot commented Aug 8, 2024 •

edited

Loading

ogrisel left a comment •

edited

Loading

snath-xoc commented Apr 25, 2025 •

edited

Loading

ogrisel left a comment •

edited

Loading