Skip to content

GitLab

Switch branch/tag
  1. 09 Jun, 2022 4 commits
  3. 08 Jun, 2022 2 commits
    • Nicolas Hug's avatar
      [FBcode->GH] [quant][core][better-engineering] Rename files in quantized directory… (#6133) · a7e4fbdc
      Nicolas Hug authored 
      * [quant][core][better-engineering] Rename files in quantized directory to conform with non-quantized countertpart filenames (#77037)

Summary:
X-link: https://github.com/pytorch/pytorch/pull/77037



Names of analogous files in quantized directory (previously snake case) were inconsistent with
their non-quantized filename counterparts (pascal case). This is the first of a series of PRs that changes
all files in quantized (and sub-directories) dir to have pascal case.

`aten/src/ATen/native/quantized/qconv_unpack.cpp` has not been renamed yet
because (for reasons currently unknown) after making the name change, `import torch` produces the below error (`qlinear_unpack.cpp` renaming also seems to fail some phabricator CI tests for similar reasons). We suspect that these may be undefined errors and will revisit naming these files in a future PR.

```
terminate called after throwing an instance of 'c10::Error'
  what():  Type c10::intrusive_ptr<ConvPackedParamsBase<2> > could not be converted to any of the known types.
Exception raised from operator() at ../aten/src/ATen/core/jit_type.h:1735 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) + 0x55 (0x7f26745c0c65 in /data/users/dzdang/pytorch/torch/lib/libc10.so)
frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) + 0xb1 (0x7f26745bdcd1 in /data/users/dzdang/pytorch/torch/lib/libc10.so)
frame #2: <unknown function> + 0x1494e24 (0x7f2663b14e24 in /data/users/dzdang/pytorch/torch/lib/libtorch_cpu.so)
frame #3: <unknown function> + 0xfed0bc (0x7f266366d0bc in /data/users/dzdang/pytorch/torch/lib/libtorch_cpu.so)
frame #4: c10::detail::infer_schema::make_function_schema(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >&&, c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>, c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>) + 0x5a (0x7f266366d71a in /data/users/dzdang/pytorch/torch/lib/libtorch_cpu.so)
frame #5: c10::detail::infer_schema::make_function_schema(c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>, c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>) + 0x7b (0x7f266366e06b in /data/users/dzdang/pytorch/torch/lib/libtorch_cpu.so)
frame #6: <unknown function> + 0x1493f32 (0x7f2663b13f32 in /data/users/dzdang/pytorch/torch/lib/libtorch_cpu.so)
frame #7: <unknown function> + 0xe227dd (0x7f26634a27dd in /data/users/dzdang/pytorch/torch/lib/libtorch_cpu.so)
frame #8: <unknown function> + 0x14e0a (0x7f268c934e0a in /lib64/ld-linux-x86-64.so.2)
..........................truncated.............
```

Reviewed By: malfet

Differential Revision: D36862332

Pulled By: dzdang

fbshipit-source-id: 598c36656b4e71f906d940e7ff19ecf82d43031d

* empty commit

* empty commit

* empty commit
Co-authored-by: default avatardzdang <dzdang@umich.edu>
Co-authored-by: default avatarVasilis Vryniotis <datumbox@users.noreply.github.com>
      a7e4fbdc
    • Nikita Shulga's avatar
      GH: Add M1 conda build workflows (#6135) · ee6f6ec2
      Nikita Shulga authored 
      Clean up Conda build folder before every run
Enable artifact upload to GitHub for every workflow run, but upload to Conda/S3 only on nightly pushes

Test plan: `conda install -c pytorch-nightly torchvision; python -c "import torchvision;print(torchvision.io.read_image('hummingbird.jpg').shape)"`
      ee6f6ec2
  5. 06 Jun, 2022 3 commits
  7. 02 Jun, 2022 2 commits
    • Nikita Shulga's avatar
      [M1] Install "jpeg<=9b" rather than OpenJpeg (#6122) · c890a7e7
      Nikita Shulga authored 
      Explicitly set PATH to point to `conda` binary, otherwise libjpeg detection logic does not work
Pin libjpeg to the same version on x86 and m1
Add simple tests that jpeg can be decoded by a generated wheel
      c890a7e7
    • Nikita Shulga's avatar
      [BE] Unify version computation (#6117) · 59ef2ab0
      Nikita Shulga authored 
      * [BE] Unify version computation

Instead of hardcoding dev version in various script, use one from
`version.txt` if `setup_build_version` is called without arguments

Also, pass `--pre` option to M1 build/test pip install commands to build
TorchVision against nightly pytorch

* Pin torchvision dependency to a specific pytorch version
      59ef2ab0
  9. 31 May, 2022 2 commits
  11. 30 May, 2022 3 commits
  13. 26 May, 2022 4 commits
    • Nikita Shulga's avatar
      Add M1 wheels binary builds (#5948) · d5929257
      Nikita Shulga authored
      d5929257
    • Vasilis Vryniotis's avatar
      Change weights return type to Mapping (#6097) · 6aaa2b00
      Vasilis Vryniotis authored
      6aaa2b00
    • YosuaMichael's avatar
      Refactor swin transfomer so later we can reuse component for 3d version (#6088) · 952f4806
      YosuaMichael authored 
      * Use List[int] instead of int for window_size and shift_size

* Make PatchMerging and SwinTransformerBlock able to handle 2d and 3d cases

* Separate patch embedding from SwinTransformer and enable to get model without head by specifying num_heads=None

* Dont use if before padding so it is fx friendly

* Put the handling on window_size edge cases on separate function and wrap with torch.fx.wrap so it is excluded from tracing

* Update the weight url to the converted weight with new structure

* Update the accuracy of swin_transformer

* Change assert to Exception and nit

* Make num_classes optional

* Add typing output for _fix_window_and_shift_size function

* init head to None to make it jit scriptable

* Revert the change to make num_classes optional

* Revert unneccesarry changes that might be risky

* Remove self.head declaration
      952f4806
    • Philip Meier's avatar
      add tests for F.pad_bounding_box (#6038) · 1d50dfa0
      Philip Meier authored 
      

* add tests for F.pad_bounding_box

* Added correctness tests for pad and reimplemented bbox op to keep dtype

* Update _geometry.py
Co-authored-by: default avatarvfdev <vfdev.5@gmail.com>
      1d50dfa0
  15. 25 May, 2022 2 commits
  17. 24 May, 2022 2 commits
  19. 23 May, 2022 5 commits
  21. 20 May, 2022 5 commits
  23. 19 May, 2022 4 commits
  25. 18 May, 2022 2 commits