Skip to content

Move _tensor_accumulation_impl extension and use it for dpnp#2791

Open
vlad-perevezentsev wants to merge 116 commits intoinclude-dpctl-tensorfrom
move_tensor_accumulation_impl
Open

Move _tensor_accumulation_impl extension and use it for dpnp#2791
vlad-perevezentsev wants to merge 116 commits intoinclude-dpctl-tensorfrom
move_tensor_accumulation_impl

Conversation

@vlad-perevezentsev
Copy link
Contributor

@vlad-perevezentsev vlad-perevezentsev commented Mar 2, 2026

This PR completely moves _tensor_accumulation_impl pybind11 extension into dpctl_ext.tensor and extends dpctl_ext.tensor Python API with the functions cumulative_logsumexp, cumulative_prod and cumulative_sum reusing them in dpnp

  • Have you provided a meaningful PR description?
  • Have you added a test, reproducer or referred to an issue with a reproducer?
  • Have you tested your changes locally for CPU and GPU devices?
  • Have you made sure that new changes do not introduce compiler warnings?
  • Have you checked performance impact of proposed changes?
  • Have you added documentation for your changes, if necessary?
  • Have you added your changes to the changelog?

@vlad-perevezentsev
Rename folder dpctl to dpctl_ext
d4fd805
@vlad-perevezentsev
Add simplify_iteration_space implementation to libtensor
c040713
@vlad-perevezentsev
Extend codespell ignore list for libtensor
14b466f
@vlad-perevezentsev
Add copy_and_cast kernels to libtensor
dcc421b
@vlad-perevezentsev
Add copy_usm_ndarray_into_usm_ndarray implementation
5a9c14c
@vlad-perevezentsev
Add pybind11 bindings for dpctl_ext.tensor._tensor_impl
4f63340
@vlad-perevezentsev
Add CMake build files for dpctl_ext
634579c
@vlad-perevezentsev
Add empty __init__ to dpctl_ext/
79d40f2
@vlad-perevezentsev
Enable _same_logical_tensors in _tensor_impl
7949c17
@vlad-perevezentsev
Add device_support_queries to enable default device types
29d6c02
@vlad-perevezentsev
Enable building and packaging of dpctl_ext
936e719
@vlad-perevezentsev
Use _tensor_impl from dpctl_ext.tensor in dpnp
cd85f1e
@vlad-perevezentsev
Move put() and take() to dpctl_ext/tensor
0c6780a
@vlad-perevezentsev
Use put/take from dpctl_ext.tensor in dpnp
87e5482
@vlad-perevezentsev
Move full() to dpctl_ext/tensor
b537f30
@vlad-perevezentsev
Use full and _full_usm_ndarray from dpctl_ext in dpnp
d50f263
@vlad-perevezentsev
Update .gitignore to ignore .so files in dpctl_ext
f189dc5
@vlad-perevezentsev
Move _zeros_usm_ndarray to dpctl_ext
f9a1817
@vlad-perevezentsev
Use _zeros_usm_ndarray from dpctl_ext in dpnp_fill.py
4b8505a
@vlad-perevezentsev
Move linear-sequence implementations to dpctl_ext/tensor
61106b2
@vlad-perevezentsev
Use _tensor_impl from dpctl_ext in dpnp_utils_fft.py
a030579
@vlad-perevezentsev
Move tril()/triu() to dpctl_ext/tensor
a1d6fa3
@vlad-perevezentsev
Use tril/triu/_tril from dpctl_ext.tensor in dpnp
f1d6e56
@vlad-perevezentsev
Disable pylint no-name-in-module for dpctl_ext
6680790
@vlad-perevezentsev
Add TODO comments
263b717
@vlad-perevezentsev
Use default_device_complex_type from dpctl_ext on test_array_api_info.py
4130c1b
@vlad-perevezentsev
Remove unused build_dpctl_ext function
17ca9ab
@vlad-perevezentsev
Apply remarks for CMake files
79cb2a4
@vlad-perevezentsev
Apply remarks for c++ files
4bf080e
@vlad-perevezentsev
Remove linear-sequence implementations
cfa6cd6
@vlad-perevezentsev
Move ti.moveaxis() to dpctl_ext.tensor and reuse it
9c88edb
@vlad-perevezentsev
Move ti.squeeze() to dpctl_ext.tensor and reuse it
ba51636
@vlad-perevezentsev
Move ti.stack() to dpctl_ext.tensor and reuse it
bb16c19
@vlad-perevezentsev
Move ti.swapaxes() to dpctl_ext.tensor and reuse it
ccad5f0
@vlad-perevezentsev
Move ti.tile() to dpctl_ext.tensor and reuse it
b5e3541
@vlad-perevezentsev
Move ti.unstack() to dpctl_ext.tensor and reuse it
325729b
@vlad-perevezentsev
Initialize _tensor_accumulation_impl extension and move _cumsum_over_…
4552e78 
…axis
@vlad-perevezentsev
Move ti.cumulative_sum() and reuse it in dpnp
6b81e7a
@vlad-perevezentsev
Move _cumprod_over_axis to dpctl_ext.tensor._tensor_accumulation_impl
5af94c8
@vlad-perevezentsev
Add missing include
91547cc
@vlad-perevezentsev
Move ti.cumulative_prod() and reuse it in dpnp
668f3fb
@vlad-perevezentsev
Move _cumlogsumexp_over_axis to dpctl_ext.tensor._tensor_accumulation…
72d2109 
…_impl
@vlad-perevezentsev
Move ti.cumulative_logsumexp() and reuse it in dpnp
d8c3680
@vlad-perevezentsev vlad-perevezentsev self-assigned this Mar 2, 2026
@vlad-perevezentsev vlad-perevezentsev changed the title Move _tensor_accumulation_impl extensions and use it for dpnp Move _tensor_accumulation_impl extension and use it for dpnp Mar 2, 2026
@github-actions
Copy link
Contributor

github-actions bot commented Mar 2, 2026

View rendered docs @ https://intelpython.github.io/dpnp/pull/2791/index.html

@github-actions
Copy link
Contributor

github-actions bot commented Mar 2, 2026 •
edited
Loading

Array API standard conformance tests for dpnp=0.20.0dev3=py313h509198e_130 ran successfully.
Passed: 1356
Failed: 1
Skipped: 16

@vlad-perevezentsev vlad-perevezentsev marked this pull request as ready for review March 2, 2026 20:31
@vlad-perevezentsev vlad-perevezentsev mentioned this pull request Mar 2, 2026
Open
antonwolfy
antonwolfy reviewed Mar 4, 2026
View reviewed changes
Base automatically changed from move_tensor_impl_linear_sequence to include-dpctl-tensor March 5, 2026 14:44
@vlad-perevezentsev vlad-perevezentsev force-pushed the move_tensor_accumulation_impl branch from 445ec2b to 85ef3e0 Compare March 5, 2026 18:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@ndgrigorian ndgrigorian ndgrigorian left review comments

@antonwolfy antonwolfy antonwolfy left review comments

At least 1 approving review is required to merge this pull request.

Assignees

@vlad-perevezentsev vlad-perevezentsev

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@vlad-perevezentsev @ndgrigorian @antonwolfy