Add Weighted Quantile and Percentile Support to jax.numpy #32737

Aniketsy · 2025-10-18T15:25:15Z

This PR adds support for weighted quantile and percentile computations to jax.numpy.quantile, jax.numpy.nanquantile, jax.numpy.percentile, and jax.numpy.nanpercentile .

Please let me know if my approach or fix needs any improvements . I’m open to feedback and happy to make changes based on suggestions.
Thankyou !

gemini-code-assist · 2025-10-18T15:25:28Z

Summary of Changes

Hello @Aniketsy, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly enhances the jax.numpy module by introducing support for weighted quantile and percentile calculations. By integrating an optional weights argument into the existing quantile, nanquantile, percentile, and nanpercentile functions, users can now perform more sophisticated statistical analyses where individual data points contribute differently to the overall distribution. The change involves modifying function signatures and implementing the core weighted logic within the internal _quantile function, ensuring accurate results across various interpolation methods and proper handling of NaN values.

Highlights

Weighted Quantile/Percentile Support: The jax.numpy.quantile, jax.numpy.nanquantile, jax.numpy.percentile, and jax.numpy.nanpercentile functions now accept an optional weights parameter, enabling weighted computations.
Core Weighted Logic Implementation: The internal _quantile function has been extended to handle weighted inputs, including sorting data and weights, calculating cumulative normalized weights, and determining quantile values based on various interpolation methods.
NaN Handling with Weights: The weighted quantile logic correctly handles NaN values by masking them and adjusting weights accordingly when squash_nans is enabled.
New Test Case: A new unit test, test_weighted_quantile_linear, has been added to tests/lax_numpy_reducers_test.py to validate the correctness of the weighted quantile calculation.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This pull request introduces support for weighted quantiles and percentiles, which is a valuable addition. The overall approach is sound, but there are a couple of significant issues in the implementation within _quantile. Firstly, there's a structural issue where the new logic for weighted quantiles is added after the axis variable has been processed, leading to dead code and incorrect behavior for NaN handling when axis is None. Secondly, the implementation for the weighted case appears to be incorrect when q (the quantiles) is a vector. I've provided detailed comments and suggestions for fixes. I also recommend expanding the test suite to cover these edge cases to ensure the feature is robust.

Aniketsy · 2025-10-19T18:40:13Z

@jakevdp please review these changes, when you get a chance. I'm not sure if i correctly placed -handle weights also all tests are not covered yet, should i add. Thanks!

jakevdp

Hey - thanks for working on this – I got a bit of the way through and then stopped reviewing becuse this implementation has fundamental flaws. I'd suggest starting over, and writing test cases first (including JIT-compilation) and make sure they pass before you request review.

jax/_src/numpy/reductions.py

jakevdp · 2025-10-20T16:58:23Z

jax/_src/numpy/reductions.py

+  # Handle weights
+  if weights is not None:
+    a, weights = promote_dtypes_inexact(a, weights)
+    if axis is None:


Axis will never be None here, because of the if axis is None check above.

If we need to change the shape of weights based on the value of axis, that needs to be done before this block.

jax/_src/numpy/reductions.py

Aniketsy · 2025-10-24T08:55:14Z

Thanks for the earlier suggestions!
I’ve added additional tests and re-implemented the changes , all tests are now passing locally.
However, I’m unsure if I’ve fully addressed, as this part about -- a has already been promoted; we should promote only once. That requires making the previous promote_dtypes_inexact conditional on whether weights is supplied.
When I tried to apply that change, it led to some test failures. so I’ll look more into it.
Please feel free to guide me if I’m missing something or going in the wrong direction.

jakevdp

Hi - it looks like you resolved some merge conflicts in a way that led to your branch being invalid. Please take a look!

I'll review further once these issues are addressed.

jax/_src/numpy/reductions.py

jakevdp

We're getting there! A number of comments below.

jax/_src/numpy/reductions.py

tests/lax_numpy_reducers_test.py

Aniketsy · 2025-10-24T21:35:23Z

Thank you for the detailed feedback and for pointing out the corrections. I’ve updated the changes as per your suggestions. Please let me know if any further improvements are needed.

jakevdp

A few more comments!

Also, regarding the tests, you should be aware of this: numpy/numpy#30069

It looks like the NumPy release has some inconsistent behavior in some cases; if you see test failures in relevant cases, it could be due to this.

jax/_src/numpy/reductions.py

jakevdp · 2025-10-24T22:38:19Z

jax/_src/numpy/reductions.py

    Array([1.5, 3. , 4.5], dtype=float32)
  """
-  a, q = ensure_arraylike("nanquantile", a, q)
+  check_arraylike("nanquantile", a, q)


jax/_src/numpy/reductions.py

tests/lax_numpy_reducers_test.py

Aniketsy · 2025-10-25T05:39:23Z

Thanks! I’ve updated the changes. I also have a small query, should we add local import inside functions that use jnp ?

jakevdp

Please run tests locally to catch problems in your code: for a change like this relying on github CI will lead to the development cycle being very slow. There is information here: https://docs.jax.dev/en/latest/contributing.html#contributing-code-using-pull-requests

jax/_src/numpy/reductions.py

jakevdp · 2025-10-25T17:08:49Z

jax/_src/numpy/reductions.py

+      resh = [1] * a.ndim
+      resh[axis] = w_shape[0]
+      weights = lax.expand_dims(weights, axis)
+    weights = _broadcast_to(weights, a.shape)


Should this be in the if block? If weights.shape is already equal to a.shape, this is not needed.

jax/_src/numpy/reductions.py

tests/lax_numpy_reducers_test.py

Aniketsy · 2025-10-25T20:30:16Z

I've tested locally and getting test fails AttributeError: lax.count , interpolation deprecation error, please let me know, if I should update these.

cakedev0

I took a very quick look because I've been working on this subject those past days/weeks but I'm not familiar at all with the jax project.

I think this PR is not really aligned with numpy or scikit-learn implementations (sklearn.utils.stats._weighted_percentile), so I suggest to take a deep look at those before continuing this PR. The numpy's doc is pretty helpful too.
You can also take a look at my PR in array-api-extra: data-apis/array-api-extra#494 ^^ I believe it's the easiest to read, but I'm definitely biased here 😅

cakedev0 · 2025-10-25T20:33:56Z

jax/_src/numpy/reductions.py

+      if method == "linear":
+        denom = cw_next - cw_prev
+        denom = _where(denom == 0, 1, denom)
+        weight = (qi - cw_prev) / denom
+        out = val_prev * (1 - weight) + val * weight
+      elif method == "lower":
+        out = val_prev
+      elif method == "higher":
+        out = val
+      elif method == "nearest":
+        out = _where(lax.abs(qi - cw_prev) < lax.abs(qi - cw_next), val_prev, val)
+      elif method == "midpoint":
+        out = (val_prev + val) / 2
+      elif method == "inverted_cdf":
+        out = val


How did you come up with those formula?

While they seem relatively sounded, it worth noting that NumPy only support method "inverted_cdf" with weights. And scikit-learn's internal _weighted_percentile implements "inverted_cdf" and "average_inverted_cdf" (through parameter average=True/False).

Typically, in your implementation I see that both methods "higher" and "inverted_cdf" are equivalent, but for non-weighted case in numpy, there aren't equivalent, try this for instance:

import numpy as np a = np.random.rand(10) q = np.linspace(0, 1, num=11) np.allclose(np.quantile(a, q, method='higher'), np.quantile(a, q, method='inverted_cdf'))

So I don't think it's correct for both methods to be equivalent in the weighted case.

cakedev0 · 2025-10-25T20:34:59Z

jax/_src/numpy/reductions.py

  elif method == "midpoint":
    result = lax.mul(lax.add(low_value, high_value), lax._const(low_value, 0.5))
+  elif method == "inverted_cdf":
+    result = high_value


Same here: this is not consistent with numpy.

cakedev0 · 2025-10-25T20:38:17Z

jax/_src/numpy/reductions.py

+      idx = lax.clamp(0, idx, a_sorted.shape[axis] - 1)
+      val = jnp.take_along_axis(a_sorted, idx, axis)
+
+      idx_prev = lax.clamp(idx - 1, 0, a_sorted.shape[axis] - 1)


Given that some weights might be null, doing idx - 1 might not have the expected effect in some cases (typically adding adding 0-weight samples shouldn't affect the output, but with this logic it will)

jax/_src/numpy/reductions.py

jakevdp · 2025-10-26T15:55:58Z

Using from jax._src.lax import lax causes all tests to fail.

You'll have to figure out why it causes failures and then fix them

Also, _const was not applied by me.

Some change you made caused this to fail (did you overwrite the local variable lax?). In order to proceed, you'll have to debug this and fix the issue.

These AttributeErrors are the easiest issues you're going to encounter: we haven't even gotten to the hard stuff yet, which is making sure outputs produced by your code match those produced by NumPy.

Aniketsy · 2025-10-26T17:39:27Z

Sure, I’ll fix these soon.

Aniketsy · 2025-10-27T11:42:30Z

I've fixed Attribute errors, still getting some errors, I'm working on them please, correct me if these changes needs improvement. Thanks!

====================== short test summary info =======================
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testMedian0 - AssertionError: Tuples differ: (1, 1) != (1,)
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testMedian1 - AssertionError: Tuples differ: (1, 1, 1) != (1, 1)
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testMedian2 - AssertionError: Tuples differ: (1, 1, 1) != (1, 1)
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testMedian4 - AssertionError: Tuples differ: (1, 1, 1) != (1, 1)
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantile4 - AssertionError: Tuples differ: (4,) != (1, 4)
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantile5 - AssertionError: Tuples differ: (1, 7) != (1, 1, 7)
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantile7 - AssertionError: Tuples differ: (1, 4, 1, 1) != (4, 1, 1)       
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantileDeprecatedArgs0 - TypeError: nanquantile() argument interpolation was removed in JAX...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantileDeprecatedArgs2 - TypeError: quantile() argument interpolation was 
removed in JAX v0...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile0 - ValueError: Quantiles must be in the range [0, 1]      
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile1 - jax._src.dtypes.TypePromotionError: Input dtypes ('int8', 'float32...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile2 - jax._src.dtypes.TypePromotionError: Input dtypes ('int8', 'float32...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile3 - jax._src.dtypes.TypePromotionError:

jakevdp · 2025-10-27T12:03:02Z

It seems like most of the errors are related to incorrect shapes when the leading dimension is 1. I don't totally follow the reasoning behind all the nested if statements at the end of the main function when you reshape the result array, but I'm pretty sure that's where the issue lies.

Aniketsy · 2025-10-27T18:43:29Z

FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantileDeprecatedArgs0 - TypeError: nanquantile() argument interpolation was removed in JAX...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantileDeprecatedArgs2 - TypeError: quantile() argument interpolation was 
removed in JAX v0...

Just to confirm do i need to fix these as we already have this, in our code

if not isinstance(interpolation, DeprecatedArg):
    raise TypeError("nanquantile() argument interpolation was removed in JAX"
                    " v0.8.0. Use method instead.")

Also, could you please provide some insight on this TypePromotionError? Thanks!

FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile0 - ValueError: Quantiles must be in the range [0, 1]      
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile1 - jax._src.dtypes.TypePromotionError: Input dtypes ('int8', 'float32...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile2 - jax._src.dtypes.TypePromotionError: Input dtypes ('int8', 'float32...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::test_weighted_quantile_all_weights_zero - AssertionError: ValueError not raisedFAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::test_weighted_quantile_negative_weights - AssertionError: ValueError not raised

jakevdp · 2025-10-27T19:24:09Z

Just to confirm do i need to fix these as we already have this, in our code

Yes, if tests are failing you need to fix them. I'm not entirely sure how you're hitting that error case: perhaps you're passing an interpolation argument into the function somewhere?

Also, could you please provide some insight on this TypePromotionError? Thanks!

All JAX tests are run with jax_numpy_dtypes_promotion='strict', so implementations within JAX cannot rely on implicit dtype promotion. The reason for this is that if we used implicit promotion within JAX's core, then jax_numpy_dtype_promotion='strict' would be basically useless.

jakevdp · 2025-10-27T19:25:57Z

jax/_src/numpy/reductions.py

 import numpy as np

+import jax
+from jax import lax


I commented on this previously, but it still applies: you cannot import top-level APIs in source files. For things like jax.vmap, use api.vmap here. And instead of jax.lax, use jax._src.lax and its contents.

jakevdp · 2025-10-27T19:28:45Z

tests/lax_numpy_reducers_test.py

+    weights = np.abs(rng(weights_shape, a_dtype)) + 1e-3
+
+    def np_fun(a, q, weights):
+      return np.quantile(np.array(a), np.array(q), axis=axis, weights=np.array(weights), method=method, keepdims=keepdims)


These are already numpy arrays; no need to pass them through np.array

tests/lax_numpy_reducers_test.py

Aniketsy · 2025-10-29T08:51:16Z

I made a few additional modifications in this file apart from the parts where I was implementing weights, mainly to address some attribute errors. A few tests are still failing, but before making further updates, I wanted to check if these extra changes are acceptable or if I should revert them. Thank you!

Aniketsy · 2025-10-29T08:58:52Z

These are some of the test failures I’m currently working on. Could you please provide some insights into them?
I also apologize for bothering you repeatedly with these test failures, I really appreciate your guidance! Thanks

========================== short test summary info ========================== 
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantileDeprecatedArgs0 - TypeError: nanquantile() argument interpolation was removed in 
JAX v0.8.0...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testQuantileDeprecatedArgs2 - TypeError: quantile() argument interpolation was removed in JAX v0.8.0. U...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile0 - jax.errors.TracerBoolConversionError: Attempted boolean conversion of tra...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile1 - jax.errors.TracerBoolConversionError: Attempted boolean conversion of tra...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile2 - ValueError: Length of weights not compatible with specified axis.    
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile3 - UserWarning: Explicitly requested dtype float64 requested in astype is no...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile6 - ValueError: Length of weights not compatible with specified axis.    
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile7 - jax.errors.TracerBoolConversionError: Attempted boolean conversion of tra...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile8 - jax.errors.TracerBoolConversionError: Attempted boolean conversion of tra...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::testWeightedQuantile9 - UserWarning: Explicitly requested dtype float64 requested in astype is no...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::test_weighted_quantile_all_weights_zero - jax.errors.TracerBoolConversionError: Attempted boolean conversion of tra...
FAILED tests/lax_numpy_reducers_test.py::JaxNumpyReducerTests::test_weighted_quantile_negative_weights - jax.errors.TracerBoolConversionError: Attempted boolean conversion of tra...

jakevdp · 2025-10-29T16:29:03Z

These are some of the test failures I’m currently working on.

The interpolation failures are confusing to me. I don't see anywhere in the current changes where this would arise – do you have local changes that you haven't pushed?

Aniketsy · 2025-10-29T17:31:42Z

I’ve pushed the latest changes, these failures are from the latest commit. I’ll go through everything again to see what I might be missing, and please feel free to guide me if you notice something.

jakevdp · 2025-10-29T17:42:32Z

jax/_src/numpy/reductions.py

  return lax.transpose(a, perm)

-def _upcast_f16(dtype: DTypeLike) -> DType:
+def _upcast_f16(dtype: DTypeLike) -> DTypeLike:


This should stay -> DType, because it returns a dtype.

jax/_src/numpy/reductions.py

jakevdp

It looks like a number of unrelated changes snuck in here (possibly due to a bad resolution of a merge conflict)

Please revert all unrelated changes, and then we can continue debugging your code.

jax/_src/numpy/reductions.py

Aniketsy · 2025-10-30T14:23:42Z

I’ve reverted the changes.

jax/_src/numpy/reductions.py

jakevdp · 2025-10-31T14:13:13Z

I took a very quick look because I've been working on this subject those past days/weeks but I'm not familiar at all with the jax project.

I think this PR is not really aligned with numpy or scikit-learn implementations (sklearn.utils.stats._weighted_percentile), so I suggest to take a deep look at those before continuing this PR. The numpy's doc is pretty helpful too. You can also take a look at my PR in array-api-extra: data-apis/array-api-extra#494 ^^ I believe it's the easiest to read, but I'm definitely biased here 😅

@Aniketsy I wonder if you saw this comment – it's important. So far my reviews have not focused on the correctness of the implementation, as well-crafted tests will generally tell you whether you've gotten that right. But it's taken a lot of iteration to get to the point where the tests can even be run.

jakevdp · 2025-10-31T14:17:40Z

We've been at this for two weeks, and your responses to my reviews are generally two steps forward, one step back. Your commits keep referencing old APIs that no longer exist in the package (most recently, jax._src.numpy.util._complex_elem_type). This makes me think you're using an LLM and blindly pushing the commits without testing locally.

This wastes my time and yours – I think it would be best to not continue with this change.

Add weighted quantile and percentile support with tests

6a60ed4

gemini-code-assist bot reviewed Oct 18, 2025

View reviewed changes

jakevdp reviewed Oct 20, 2025

View reviewed changes

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

Add weighted quantile and percentile support with tests

f5d2177

Aniketsy force-pushed the support-32647 branch from dafc4c1 to f5d2177 Compare October 24, 2025 08:29

jakevdp requested changes Oct 24, 2025

View reviewed changes

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

Add weighted quantile and percentile support with tests

f7ab683

Aniketsy force-pushed the support-32647 branch from 85873ce to f7ab683 Compare October 24, 2025 17:41

Add weighted quantile and percentile support with tests

5f881bf

Aniketsy force-pushed the support-32647 branch from 4593d40 to 5f881bf Compare October 24, 2025 17:53

jakevdp requested changes Oct 24, 2025

View reviewed changes

Add weighted quantile and percentile support with tests

7b967cb

Aniketsy force-pushed the support-32647 branch from fb246ba to 7b967cb Compare October 24, 2025 21:24

jakevdp requested changes Oct 24, 2025

View reviewed changes

Add weighted quantile and percentile support with tests

4f522e6

Aniketsy force-pushed the support-32647 branch from 2ab7977 to 4f522e6 Compare October 25, 2025 05:30

jakevdp requested changes Oct 25, 2025

View reviewed changes

Add weighted quantile and percentile support with tests

cef1731

Aniketsy force-pushed the support-32647 branch from f3c407d to cef1731 Compare October 25, 2025 20:24

cakedev0 reviewed Oct 25, 2025

View reviewed changes

jakevdp reviewed Oct 25, 2025

View reviewed changes

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

jakevdp reviewed Oct 25, 2025

View reviewed changes

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

Add weighted quantile and percentile support with tests

a230e01

Add weighted quantile and percentile support with tests

7d50a32

Aniketsy force-pushed the support-32647 branch from dcbc485 to 7d50a32 Compare October 27, 2025 11:34

jakevdp reviewed Oct 27, 2025

View reviewed changes

Aniketsy force-pushed the support-32647 branch from 0a9de57 to 8e9c8d6 Compare October 29, 2025 08:41

jakevdp reviewed Oct 29, 2025

View reviewed changes

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

jakevdp reviewed Oct 29, 2025

View reviewed changes

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

jax/_src/numpy/reductions.py Outdated Show resolved Hide resolved

Add weighted quantile and percentile support with tests

ca1d95b

Aniketsy force-pushed the support-32647 branch from 4d0dfd4 to ca1d95b Compare October 30, 2025 14:14

Add weighted quantile and percentile support with tests

4ebbf21

Aniketsy force-pushed the support-32647 branch from 4b318a4 to 4ebbf21 Compare October 30, 2025 14:47

jakevdp reviewed Oct 30, 2025

View reviewed changes

Add weighted quantile and percentile support with tests

43cdb5c

Aniketsy force-pushed the support-32647 branch from 19ad7be to 43cdb5c Compare October 31, 2025 05:35

Merge branch 'main' into support-32647

d9c4614

jakevdp closed this Oct 31, 2025

Add Weighted Quantile and Percentile Support to jax.numpy #32737

Add Weighted Quantile and Percentile Support to jax.numpy #32737

Conversation

Aniketsy commented Oct 18, 2025

Uh oh!

gemini-code-assist bot commented Oct 18, 2025

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Aniketsy commented Oct 19, 2025

Uh oh!

jakevdp left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Aniketsy commented Oct 24, 2025

Uh oh!

jakevdp left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jakevdp left a comment

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!

Aniketsy commented Oct 24, 2025

Uh oh!

jakevdp left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

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!

Aniketsy commented Oct 25, 2025

Uh oh!

jakevdp left a comment

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!

Aniketsy commented Oct 25, 2025

Uh oh!

cakedev0 left a comment

Choose a reason for hiding this comment

Uh oh!

jakevdp commented Oct 26, 2025 •

edited

Loading