Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/30892
Fixes all outstanding lints and actually installs a properly configured
flake8
Test Plan: Imported from OSS
Differential Revision: D18862825
Pulled By: suo
fbshipit-source-id: 08e9083338a7309272e17bb803feaa42e348aa85
Summary:
With the CI failure caused in 8bbafa0b32 fixed (incorrect return type of the lambdas in CUDA kernels)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/30521
Differential Revision: D18770151
Pulled By: ailzhang
fbshipit-source-id: 02f0fe1d5718c34d24da6dbb5884ee8b247ce39a
Summary:
There is no `out` argument to `argsort` according to the source code.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/24335
Differential Revision: D16829134
Pulled By: vincentqb
fbshipit-source-id: 8f91154984cd4a753ba1d6105fb8a9bfa0da22b3
Summary:
In-tree changes to pytorch to support complex numbers are being submitted here.
Out-of-tree support for complex numbers is here: [pytorch-cpu-strided-complex extension](https://gitlab.com/pytorch-complex/pytorch-cpu-strided-complex)
Changes so far:
- [x] Renamed references to variable "I" that may be confused for "I" defined in complex.h. I did this to avoid crazy CI failures messages as complex.h is included by more source files.
- aten/src/ATen/native/cpu/Loops.h (Renamed I to INDEX)
- aten/src/ATen/native/cuda/Loops.cuh (Renamed I to INDEX)
- aten/src/ATen/core/ivalue_inl.h (Renamed I to INDEX)
- c10/util/Array.h (Renamed I to INDEX)
- c10/util/C++17.h (Renamed I to INDEX)
- c10/util/Metaprogramming.h (Renamed I to INDEX)
- c10/util/SmallVector.h (custom renaming)
- [x] Added complex support of Linear Algebra Ops.
- SVD needed to be modified to support mixed data types
- Example U(std::complex<double)), S(double), V(std::complex<double>)
- See before and after benchmark below (No observable change in performance).
- [x] Added complex support of Reduce Ops.
- var/std computations could have been faster if it was possible to interpret std::complex<double> Tensor as a double Tensor.
- [x] Added complex derivative support for autograd functionality.
- derivatives are the same as defined by numpy autograd library for real(), imag(), conj(), angle(). These functions only affect complex numbers.
- derivative of abs() has not been modified to not interfere with existing code.
- Autograd defines abs() for complex numbers and fabs() for real numbers. I will look into this further down the road.
----------------------------------------
PyTorch/Caffe2 Operator Micro-benchmarks Before Changes
----------------------------------------
Tag : short
Benchmarking PyTorch: svd
Mode: Eager
Name: svd_M512_N512
Input: M: 512, N: 512
Forward Execution Time (us) : 162339.425
Forward Execution Time (us) : 162517.479
Forward Execution Time (us) : 162847.775
----------------------------------------
PyTorch/Caffe2 Operator Micro-benchmarks After Changes
----------------------------------------
Tag : short
Benchmarking PyTorch: svd
Mode: Eager
Name: svd_M512_N512
Input: M: 512, N: 512
Forward Execution Time (us) : 162032.117
Forward Execution Time (us) : 161943.484
Forward Execution Time (us) : 162513.786
Pull Request resolved: https://github.com/pytorch/pytorch/pull/27653
Differential Revision: D17907886
Pulled By: ezyang
fbshipit-source-id: a88b6d0427591ec1fba09e97c880f535c5d0e513
Summary:
Following from https://github.com/pytorch/pytorch/issues/28479 let's remove the type information from the docstrings of these functions as well, making them valid python signatures matching the other signatures in the docstrings for the torch API.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/28556
Differential Revision: D18115641
Pulled By: ezyang
fbshipit-source-id: e4c3d56981b16f5acabe8be7bfbe6ae506972d7f
Summary:
Changelog:
- Changes the behavior of returning a zero tensor when eigenvectors=False, matching behavior of torch.eig
Pull Request resolved: https://github.com/pytorch/pytorch/pull/28338
Test Plan: - test_symeig has been modified appropriately for this change
Differential Revision: D18085280
Pulled By: ezyang
fbshipit-source-id: 43129a96dd01743997157974100e5a7270742b46
Summary:
The types don't appear in the docstrings for other functions in the `torch` namespace so I think this was included here because of a copy/paste error.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/28479
Differential Revision: D18086150
Pulled By: ezyang
fbshipit-source-id: 2481bccba6df36b12779a330f8c43d4aea68495f
Summary:
f362a5a04b reverted
5ca612b55e due to build time conerns (also
see https://github.com/pytorch/pytorch/issues/25254). Now we come back to this by reusing the underlying code in
comparison operators: Logical operators on non-bool variables are
essentially comparison operators that semantically output bool
values. Compared with the previous implementation, we compromise by
always applying XOR on the same input type, while output can be either
the input type or the bool type.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/27248
Differential Revision: D17929356
Pulled By: ezyang
fbshipit-source-id: dbac08c7614b36f05d24c69104fee9df9ca523d5
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/27782
Warnings show up when running `make html` to build documentation. All of
the warnings are very reasonable and point to bugs in our docs. This PR
attempts to fix most of those warnings.
In the future we will add something to the CI that asserts that there
are no warnings in our docs.
Test Plan: - build and view changes locally
Differential Revision: D17887067
Pulled By: zou3519
fbshipit-source-id: 6bf4d08764759133b20983d6cd7f5d27e5ee3166
Summary:
Currently when an integral tensor is divided by zero, it emits a
"floating point exception" (which can be different from system to
system). Clarify in the document that nothing would be guaranteed under
this circumstance.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/25968
Differential Revision: D17888097
Pulled By: ezyang
fbshipit-source-id: 7c3ce3ac4080479d637cc2710b6aa3ae7e42431d
Summary:
Added Complex support with AVX to unary ops and binary ops.
I need to add nan propagation to minimum() and maximum() in the future.
In-tree changes to pytorch to support complex numbers are being submitted here.
Out-of-tree support for complex numbers is here: pytorch-cpu-strided-complex extension
Preliminary Benchmarks are here.
I tried rrii and riri and found that riri is better in most situations.
Divide is very slow because you can't reduce 1/(x+y)
Sqrt is also very slow.
Reciprocal could be sped up after I add conj()
Everything else is typically within 20% of the real number performance.
Questions:
Why does macOS not support mil? #if AT_MKL_ENABLED() && !defined(__APPLE__) in vml.h. MKL does support some complex operations like Abs, so I was curious about trying it.
Is MKL just calling AVX?
Pull Request resolved: https://github.com/pytorch/pytorch/pull/26500
Differential Revision: D17835431
Pulled By: ezyang
fbshipit-source-id: 6746209168fbeb567af340c22bf34af28286bd54
Summary:
We do support inputs with dim > 2 in _out variants
Pull Request resolved: https://github.com/pytorch/pytorch/pull/26989
Differential Revision: D17785632
Pulled By: soumith
fbshipit-source-id: d42ba7ca9c225ad1a26ff3b410d0c5c08eaed001
Summary:
Change the doc of torch.where. The parameters are x and y instead of input and other
Pull Request resolved: https://github.com/pytorch/pytorch/pull/25554
Differential Revision: D17227193
Pulled By: soumith
fbshipit-source-id: 96d8a6f60ae8e788648247320ae715d0058de2b4
Summary:
Changelog:
- Enable broadcasting of RHS and LHS tensors for lu_solve. This means that you can now have RHS with size `3 x 2` and LHS with size `4 x 3 x 3` for instance
- Remove deprecated behavior of having 2D tensors for RHS. Now all tensors have to have a last dimension which equals the number of right hand sides
- Modified docs
Pull Request resolved: https://github.com/pytorch/pytorch/pull/24333
Test Plan: - Add tests for new behavior in test_torch.py with a port to test_cuda.py
Differential Revision: D17165463
Pulled By: zou3519
fbshipit-source-id: cda5d5496ddb29ed0182bab250b5d90f8f454aa6
Summary:
As in https://github.com/pytorch/pytorch/issues/23439, some descriptions of arguments in `_torch_docs.py` have been replaced by `common_args`, it would be helpful to check if any descriptions can be replaced for new docs in the future.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/24161
Differential Revision: D16889293
Pulled By: ezyang
fbshipit-source-id: bf6f581494482d6eb32e634f73e84a4586766230
Summary:
Many descriptions of arguments could be replaced by items in the template such as `factory_common_args`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23439
Differential Revision: D16688527
Pulled By: ezyang
fbshipit-source-id: 406ce45d72e297f46b5fa9ea5472b3284c8d4324
