OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
51,522 Commits 8,304 Branches 150 Tags 1.5 GiB
01c54ad6de
Commit Graph

811 Commits

Author SHA1 Message Date
Ivan Yashchuk
01c54ad6de Remove deprecated torch.eig (#70982) 
The time has come to remove deprecated linear algebra related functions. This PR removes `torch.eig`.

cc @jianyuh @nikitaved @pearu @mruberry @walterddr @IvanYashchuk @xwang233 @Lezcano
Pull Request resolved: https://github.com/pytorch/pytorch/pull/70982
Approved by: https://github.com/Lezcano, https://github.com/malfet
 2022-09-09 21:31:57 +00:00
Andrew M. James
9b115c7bd3 Sparse Compressed Transpose add support for Batch dims and BSR/BSC layouts (#82122) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/82122
Approved by: https://github.com/bhosmer
 2022-09-02 17:42:58 +00:00
joncrall
b136f3f310 More doctest refinements. (#83317) 
Follow up to #82797

Now that the doctests themselves are in a better state, we should be able to enable xdoctest on the CI so they stay that way.

@ezyang @vadimkantorov
Pull Request resolved: https://github.com/pytorch/pytorch/pull/83317
Approved by: https://github.com/ezyang
 2022-08-22 20:07:26 +00:00
PyTorch MergeBot
f534b2c627 Revert "Remove split functional wrapper (#74727)" 
This reverts commit a58876ace7.

Reverted https://github.com/pytorch/pytorch/pull/74727 on behalf of https://github.com/seemethere due to Fails internal use cases, might extend out to external use cases as well. Need to assess overall impact of this change more widely
 2022-08-10 19:45:23 +00:00
Peter Bell
a58876ace7 Remove split functional wrapper (#74727) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/74727
Approved by: https://github.com/albanD, https://github.com/khabinov
 2022-08-10 17:57:48 +00:00
Kurt Mohler
be5b3df6cc Update std_mean/var_mean/nanmean/nansum signatures with int[1]? dim (#82912) 
### Description
Change the type of the `dim` arg for `std_mean/var_mean/nanmean/nansum` to `int[1]?` in `native_functions.yaml`

### Issue
Part of #29137

### Testing

Pull Request resolved: https://github.com/pytorch/pytorch/pull/82912
Approved by: https://github.com/albanD
 2022-08-10 16:58:26 +00:00
Kurt Mohler
5ca9b2b6fa Enable dim=None for torch.var (#82765) 
### Description
Add support for `dim=None` in `torch.var`

### Issue
Part of #29137

### Testing
N/A
Pull Request resolved: https://github.com/pytorch/pytorch/pull/82765
Approved by: https://github.com/albanD
 2022-08-04 20:47:27 +00:00
Pearu Peterson
ff5399e528 Revise sparse docs regarding Sparse Compressed tensors (#82108) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/82108
Approved by: https://github.com/bhosmer
 2022-07-29 18:15:09 +00:00
ProGamerGov
8def154e00 Fix multiple docstring type mistakes (#82474) 
### Description

* Docstrings using `(tuple of ints)` shows up as `(tuple of python:ints)`, so I fixed them by making the `int` no longer plural. Example: https://pytorch.org/docs/stable/generated/torch.permute.html#torch.permute
* A docstring type in JIT had one of its types incorrectly highlighted as code. Example: https://pytorch.org/docs/stable/generated/torch.jit.script.html#torch.jit.script
* I found some docstring type usages of `string` that had not yet been converted to `str` after #82410
* Some docstrings incorrectly listed their defaults inside the docstring types.
* I also found a docstring that was missing its type

### Testing
No testing should be required.

---

In the developer guidelines, there should probably be standards listed for the docstring types.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/82474
Approved by: https://github.com/albanD
 2022-07-29 17:45:37 +00:00
Fabio Rocha
fd84c458f4 Add torch.unflatten and improve its docs (#81399) 
unflatten now has a free function version in torch.flatten in addition to
    the method in torch.Tensor.flatten.

    Updated docs to reflect this and polished them a little.
    For consistency, changed the signature of the int version of unflatten in
    native_functions.yaml.

    Some override tests were failing because unflatten has unusual
    characteristics in terms of the .int and .Dimname versions having
    different number of arguments so this required some changes
    to test/test_override.py

    Removed support for using mix of integer and string arguments
    when specifying dimensions in unflatten.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/81399
Approved by: https://github.com/Lezcano, https://github.com/ngimel
 2022-07-29 15:02:42 +00:00
Kurt Mohler
2bfae07a79 Enable dim=None for torch.mean (#81286) 
Part of #79525

This will require coordination with XLA before merging, just like #79881
Pull Request resolved: https://github.com/pytorch/pytorch/pull/81286
Approved by: https://github.com/albanD
 2022-07-28 22:34:56 +00:00
ProGamerGov
357b7d589c Fix docstring inconsistencies: string -> str, boolean -> bool (#82410) 
### Description

Throughout the PyTorch docs and codebase, the `string` type in docstrings is referred to by two separate names. This leads to inconsistent docs, like you can see here: https://pytorch.org/docs/stable/generated/torch.nn.Conv3d.html#torch.nn.Conv3d

This PR fixes this issue by ensuring that all mentions of the string type in docstrings, are using the same format that Sphinx generates hyperlinks for.

### Testing
No testing should be required for this change

Pull Request resolved: https://github.com/pytorch/pytorch/pull/82410
Approved by: https://github.com/jbschlosser
 2022-07-28 21:29:57 +00:00
Zhengyi Li
026ef78b22 [DOC] fix a mistake in torch.dist document (#82104) 
In document (https://pytorch.org/docs/stable/generated/torch.dist.html )the torch.dist(x, y, 0) returns inf.

However, it should return 4.0 and the pytorch (1.12)  does return 4.0

Pull Request resolved: https://github.com/pytorch/pytorch/pull/82104
Approved by: https://github.com/samdow
 2022-07-26 13:51:31 +00:00
Huy Do
12cb26509a Apply ufmt to torch internal (#81643) 
This is a big bang PR, merge conflicts are probably expected and will be addressed at merge.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/81643
Approved by: https://github.com/ezyang
 2022-07-22 02:19:50 +00:00
lezcano
e505796a2c [Array API] Add linalg.vecdot (#70542) 
This PR adds the function `linalg.vecdot` specified by the [Array
API](https://data-apis.org/array-api/latest/API_specification/linear_algebra_functions.html#function-vecdot)

For the complex case, it chooses to implement \sum x_i y_i. See the
discussion in https://github.com/data-apis/array-api/issues/356

Edit. When it comes to testing, this function is not quite a binopt, nor a reduction opt. As such, we're this close to be able to get the extra testing, but we don't quite make it. Now, it's such a simple op that I think we'll make it without this.

Resolves https://github.com/pytorch/pytorch/issues/18027.

cc @mruberry @rgommers @pmeier @asmeurer @leofang @AnirudhDagar @asi1024 @emcastillo @kmaehashi
Pull Request resolved: https://github.com/pytorch/pytorch/pull/70542
Approved by: https://github.com/IvanYashchuk, https://github.com/mruberry
 2022-07-12 14:28:54 +00:00
PyTorch MergeBot
7f3677d723 Revert "Remove split functional wrapper (#74727)" 
This reverts commit cc3126083e.

Reverted https://github.com/pytorch/pytorch/pull/74727 on behalf of https://github.com/mehtanirav due to Breaking multiple internals builds and tests
 2022-07-11 18:29:45 +00:00
Kurt Mohler
23bdb570cf Reland: Enable dim=None for torch.sum (#79881) 
Part of #29137

Reland of #75845
Pull Request resolved: https://github.com/pytorch/pytorch/pull/79881
Approved by: https://github.com/albanD, https://github.com/kulinseth
 2022-07-09 00:54:42 +00:00
PyTorch MergeBot
39f659c3ba Revert "[Array API] Add linalg.vecdot (#70542)" 
This reverts commit 74208a9c68.

Reverted https://github.com/pytorch/pytorch/pull/70542 on behalf of https://github.com/malfet due to Broke CUDA-10.2 for vecdot_bfloat16, see 74208a9c68
 2022-07-08 22:56:51 +00:00
Peter Bell
cc3126083e Remove split functional wrapper (#74727) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/74727
Approved by: https://github.com/albanD
 2022-07-08 19:21:22 +00:00
lezcano
74208a9c68 [Array API] Add linalg.vecdot (#70542) 
This PR adds the function `linalg.vecdot` specified by the [Array
API](https://data-apis.org/array-api/latest/API_specification/linear_algebra_functions.html#function-vecdot)

For the complex case, it chooses to implement \sum x_i y_i. See the
discussion in https://github.com/data-apis/array-api/issues/356

Edit. When it comes to testing, this function is not quite a binopt, nor a reduction opt. As such, we're this close to be able to get the extra testing, but we don't quite make it. Now, it's such a simple op that I think we'll make it without this.

Resolves https://github.com/pytorch/pytorch/issues/18027.

cc @mruberry @rgommers @pmeier @asmeurer @leofang @AnirudhDagar @asi1024 @emcastillo @kmaehashi
Pull Request resolved: https://github.com/pytorch/pytorch/pull/70542
Approved by: https://github.com/IvanYashchuk, https://github.com/mruberry
 2022-07-08 15:37:58 +00:00
lezcano
37a5819665 Make slogdet, linalg.sloget and logdet support metatensors (#79742) 
This PR also adds complex support for logdet, and makes all these
functions support out= and be composite depending on one function. We
also extend the support of `logdet` to complex numbers and improve the
docs of all these functions.

We also use `linalg_lu_factor_ex` in these functions, so we remove the
synchronisation present before.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/79742
Approved by: https://github.com/IvanYashchuk, https://github.com/albanD
 2022-07-01 16:09:21 +00:00
Rohit Goswami
04407431ff MAINT: Harmonize argsort params with array_api (#75162) 
Closes [#70922](https://github.com/pytorch/pytorch/issues/70922).
- Does what it says on the tin.
-  No non-standard implementation details.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/75162
Approved by: https://github.com/rgommers, https://github.com/nikitaved, https://github.com/mruberry
 2022-06-09 12:32:01 +00:00
lezcano
c6215b343c Deprecate torch.lu_solve 
**BC-breaking note**:

This PR deprecates `torch.lu_solve` in favor of `torch.linalg.lu_solve_factor`.
A upgrade guide is added to the documentation for `torch.lu_solve`.

Note this PR DOES NOT remove `torch.lu_solve`.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/77637

Approved by: https://github.com/malfet
 2022-06-07 22:50:14 +00:00
lezcano
f7b9a46880 Deprecate torch.lu 
**BC-breaking note**:

This PR deprecates `torch.lu` in favor of `torch.linalg.lu_factor`.
A upgrade guide is added to the documentation for `torch.lu`.

Note this PR DOES NOT remove `torch.lu`.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/77636

Approved by: https://github.com/malfet
 2022-06-07 22:50:14 +00:00
lezcano
f091b3fb4b Update torch.lu_unpack docs 
As per title

Pull Request resolved: https://github.com/pytorch/pytorch/pull/77635

Approved by: https://github.com/malfet
 2022-06-07 22:50:13 +00:00
Andrij David
bd08d085b0 Update argmin docs to reflect the code behavior (#78888) 
Fixes #78791

Pull Request resolved: https://github.com/pytorch/pytorch/pull/78888
Approved by: https://github.com/ezyang
 2022-06-07 00:44:39 +00:00
Kshiteej K
c461d8a977 [primTorch] refs: hsplit, vsplit (#78418) 
As per title

TODO:
* [x] Add error inputs (already exist)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78418
Approved by: https://github.com/mruberry
 2022-06-06 19:54:05 +00:00
kshitij12345
57c117d556 update signbit docs and add -0. to reference testing for unary and binary functions. (#78349) 
Fixes https://github.com/pytorch/pytorch/issues/53963
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78349
Approved by: https://github.com/mruberry
 2022-06-06 13:48:08 +00:00
John Clow
416f581eb1 Updating torch.log example 
Fixes issue  #78301

Pull Request resolved: https://github.com/pytorch/pytorch/pull/78776

Approved by: https://github.com/ngimel
 2022-06-03 00:57:35 +00:00
Gao, Xiang
388d44314d Fix docs for torch.real (#78644) 
Non-complex types are supported

```python
>>> import torch
>>> z = torch.zeros(5)
>>> torch.real(z.float())
tensor([0., 0., 0., 0., 0.])
>>> torch.real(z.int())
tensor([0, 0, 0, 0, 0], dtype=torch.int32)
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78644
Approved by: https://github.com/mruberry, https://github.com/anjali411
 2022-06-02 04:17:03 +00:00
Thomas J. Fan
3524428fad DOC Corrects default value for storage_offset in as_strided (#78202) 
Fixes #77730
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78202
Approved by: https://github.com/mruberry
 2022-05-31 19:28:36 +00:00
Yukio Siraichi
3f334f0dfd Fix asarray documentation formatting (#78485) 
Fixes #78290

Here's a screenshot of the modified doc:
![asarray](https://user-images.githubusercontent.com/3337141/170971723-aafe20a9-8e51-420f-ae98-67dc2df579a2.png)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/78485
Approved by: https://github.com/ngimel
 2022-05-30 19:28:10 +00:00
Mike Ruberry
089203f8bc Updates floor_divide to perform floor division (#78411) 
Fixes https://github.com/pytorch/pytorch/issues/43874

This PR changes floor_divide to perform floor division instead of truncation division.

This is a BC-breaking change, but it's a "bug fix," and we've already warned users for several releases this behavior would change.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78411
Approved by: https://github.com/ngimel
 2022-05-29 21:28:45 +00:00
Ryan Spring
2df1da09e1 Add Elementwise unary ops 4 references (#78216) 
Add reference implementations for `nan_to_num, positive, sigmoid, signbit, tanhshink`
Add prims for `minimum_value(dtype)` and `maximum_value(dtype)`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78216
Approved by: https://github.com/mruberry
 2022-05-27 21:55:34 +00:00
Brian Hirsh
07e4533403 reland of as_strided support for functionalization; introduce as_strided_scatter 
This reverts commit a95f1edd85.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/78199

Approved by: https://github.com/ezyang
 2022-05-24 22:40:44 +00:00
PyTorch MergeBot
a95f1edd85 Revert "as_strided support for functionalization; introduce as_strided_scatter" 
This reverts commit 3a921f2d26.

Reverted https://github.com/pytorch/pytorch/pull/77128 on behalf of https://github.com/suo due to This broke rocm tests on master 3a921f2d26. rocm tests are no longer run on PRs, you should add a `ciflow/trunk` label if you want to run them
 2022-05-24 20:19:12 +00:00
Brian Hirsh
3a921f2d26 as_strided support for functionalization; introduce as_strided_scatter 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/77128

Approved by: https://github.com/ezyang
 2022-05-24 18:20:31 +00:00
Jeff Daily
de86146c61 rocblas alt impl during backward pass only (#71881) 
In preparation of adopting future rocblas library options, it is necessary to track when the backward pass of training is executing.  The scope-based helper class `BackwardPassGuard` is provided to toggle state.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/71881
Approved by: https://github.com/albanD
 2022-05-18 19:42:58 +00:00
Xiang Gao
0975174652 Fix doc about type promotion of lshift and rshift (#77613) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/77613
Approved by: https://github.com/ngimel
 2022-05-17 00:28:48 +00:00
Mikayla Gawarecki
841c65f499 Unprivate _index_reduce and add documentation 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/76997

Approved by: https://github.com/cpuhrsch
 2022-05-13 19:48:38 +00:00
Xiang Gao
cc9d0f309e lshift and rshift stop support floating types (#77146) 
Fixes #74358

Pull Request resolved: https://github.com/pytorch/pytorch/pull/77146
Approved by: https://github.com/ngimel
 2022-05-11 22:29:30 +00:00
Ivan Yashchuk
890bdf13e1 Remove deprecated torch.solve (#70986) 
The time has come to remove deprecated linear algebra related functions. This PR removes `torch.solve`.

cc @jianyuh @nikitaved @pearu @mruberry @walterddr @IvanYashchuk @xwang233 @Lezcano
Pull Request resolved: https://github.com/pytorch/pytorch/pull/70986
Approved by: https://github.com/Lezcano, https://github.com/albanD
 2022-05-10 13:44:07 +00:00
PyTorch MergeBot
4ceac49425 Revert "Update torch.lu_unpack docs" 
This reverts commit 9dc8f2562f.

Reverted https://github.com/pytorch/pytorch/pull/73803 on behalf of https://github.com/malfet
 2022-05-09 19:09:43 +00:00
PyTorch MergeBot
1467e0dd5d Revert "Deprecate torch.lu" 
This reverts commit a5bbfd94fb.

Reverted https://github.com/pytorch/pytorch/pull/73804 on behalf of https://github.com/malfet
 2022-05-09 19:06:44 +00:00
PyTorch MergeBot
b042cc7f4d Revert "Deprecate torch.lu_solve" 
This reverts commit f84d4d9cf5.

Reverted https://github.com/pytorch/pytorch/pull/73806 on behalf of https://github.com/malfet
 2022-05-09 19:03:26 +00:00
lezcano
f84d4d9cf5 Deprecate torch.lu_solve 
**BC-breaking note**:

This PR deprecates `torch.lu_solve` in favor of `torch.linalg.lu_solve_factor`.
A upgrade guide is added to the documentation for `torch.lu_solve`.

Note this PR DOES NOT remove `torch.lu_solve`.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/73806

Approved by: https://github.com/IvanYashchuk, https://github.com/nikitaved, https://github.com/mruberry
 2022-05-05 19:19:19 +00:00
lezcano
a5bbfd94fb Deprecate torch.lu 
**BC-breaking note**:

This PR deprecates `torch.lu` in favor of `torch.linalg.lu_factor`.
A upgrade guide is added to the documentation for `torch.lu`.

Note this PR DOES NOT remove `torch.lu`.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/73804

Approved by: https://github.com/IvanYashchuk, https://github.com/mruberry
 2022-05-05 19:17:11 +00:00
lezcano
9dc8f2562f Update torch.lu_unpack docs 
As per title

Pull Request resolved: https://github.com/pytorch/pytorch/pull/73803

Approved by: https://github.com/IvanYashchuk, https://github.com/nikitaved, https://github.com/mruberry
 2022-05-05 19:12:23 +00:00
lezcano
7cb7cd5802 Add linalg.lu 
This PR modifies `lu_unpack` by:
- Using less memory when unpacking `L` and `U`
- Fuse the subtraction by `-1` with `unpack_pivots_stub`
- Define tensors of the correct types to avoid copies
- Port `lu_unpack` to be a strucutred kernel so that its `_out` version
does not incur on extra copies

Then we implement `linalg.lu` as a structured kernel, as we want to
compute its derivative manually. We do so because composing the
derivatives of `torch.lu_factor` and `torch.lu_unpack` would be less efficient.

This new function and `lu_unpack` comes with all the things it can come:
forward and backward ad, decent docs, correctness tests, OpInfo, complex support,
support for metatensors and support for vmap and vmap over the gradients.

I really hope we don't continue adding more features.

This PR also avoids saving some of the tensors that were previously
saved unnecessarily for the backward in `lu_factor_ex_backward` and
`lu_backward` and does some other general improvements here and there
to the forward and backward AD formulae of other related functions.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/67833

Approved by: https://github.com/IvanYashchuk, https://github.com/nikitaved, https://github.com/mruberry
 2022-05-05 09:17:05 +00:00
Pearu Peterson
5adf97d492 Add docstrings to sparse compressed tensor factory functions 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/76651

Approved by: https://github.com/cpuhrsch
 2022-05-04 03:36:14 +00:00