Kaustubh

Results 36 issues of Kaustubh

This PR builds on top of #8041 Currently has the following `Generator` methods: - `Generator().shuffle()` - `Generator().permutation()`

2 - In Progress

## Reporting a bug - [x] I have tried using the latest released version of Numba (most recent is visible in the change log (https://github.com/numba/numba/blob/main/CHANGE_LOG). - [x] I have included...

bug - incorrect behavior

This PR adds support for passing multiple axes as a tuple to `np.expand_dims()`. ```python import numpy as np import numba orig_shape = (3,4,5) orig = np.random.normal(size=orig_shape) new_order_tuple = (1, 3)...

4 - Waiting on reviewer
Effort - medium

As an effort towards #801 , We want to implement a new class based `RandomState` in our Numba backend. It'll allow us to curcumvent Numba's limitation of a global random...

enhancement
important
Numba
NumPy compatibility

This PR adds the capability to effectively replace nodes affected due to variable replacement by cloning them whenever one of it's input is changed in the FunctionGraph. (Currently this is...

enhancement
important
graph rewriting

Resolves #506 This PR adds 3 new Ops, namely the `TransposeOp`, `SqueezeDimsOp` and `ExpandDimsOp` for the `np.transpose`, `np.squeeze` and `np.expand_dims` Numpy functionalities respectively. If implemented, this will go on as...

refactor
NumPy compatibility
Op implementation

Numba has recently added support for `Generator` and `BitGenerator` objects (https://github.com/numba/numba/pull/8031). This PR updates NumPy documentation regarding the same.

Builds on top of #8415 This PR refactors the array method `np.sum` to use the new `@overload` API. It also adds a way of indexing subarrays that supports axis values...

2 - In Progress

Original Comment: https://github.com/numba/numba/pull/8415#issuecomment-1250123916 MWR: ```python from numba import njit @njit def foo(a): return np.std(a) a = np.array([]) print(foo(a)) # raise => ZeroDivisionError: division by zero print(foo.py_func(a)) => emit a warning...

good second issue

This PR refactors methods in `numba.np.arraymath` to use the new `@overload` API. It refactors the following functions: - [x] `np.prod` - [x] `np.cumsum` - [x] `np.cumprod` - [x] `np.mean` -...

3 - Ready for Review