Skip to content

GitLab

Switch branch/tag
  1. 28 Apr, 2021 1 commit
    • Nicolas Hug's avatar
      [FBcode->GH] Parametrize test_perspective (#3748) (#3749) · 1b0bd0e3
      Nicolas Hug authored 
      Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/3748

This PR parametrizes the `perspective`-related tests, and avoids having deeply nested for-loops which will help debugging. "What" gets tested is left unchanged.

The newly-introduced `cpu_and_gpu()` generator along with the `dont_collect` mark is a logic that allows to not run CPU tests on GPU machines (and vice versa).

Reviewed By: fmassa

Differential Revision: D27908299

fbshipit-source-id: 24a10a89fe90ae0a9e62de4bc7e768a669ebf212
      1b0bd0e3
  3. 27 Apr, 2021 2 commits
  5. 26 Apr, 2021 5 commits
  7. 23 Apr, 2021 4 commits
  9. 22 Apr, 2021 1 commit
  11. 21 Apr, 2021 2 commits
  13. 20 Apr, 2021 1 commit
  15. 19 Apr, 2021 1 commit
  17. 16 Apr, 2021 1 commit
  19. 12 Apr, 2021 2 commits
  21. 09 Apr, 2021 1 commit
    • Prabhat Roy's avatar
      Added KITTI dataset (#3640) · 7da9afee
      Prabhat Roy authored 
      

* Added KITTI dataset

* Addressed review comments

* Changed type of target to List[Dict] and corrected the data types of the returned values.

* Updated unit test to rely on ImageDatasetTestCase

* Added kitti to dataset documentation

* Cleaned up test and some minor changes

* Made data_url a string instead of a list

* Removed unnecessary try and print
Co-authored-by: default avatarFrancisco Massa <fvsmassa@gmail.com>
      7da9afee
  23. 08 Apr, 2021 1 commit
    • Nicolas Hug's avatar
      Add Quantized version of RoIAlign (#3624) · ad9cc62a
      Nicolas Hug authored 
      * WIP

* clang

* docs

* extracted out common utils

* Use better quantization function and pass tensors as parameters

* proper dequantization

* Some tests

* Dequantization optimization, seems to gain a few ms

* clang-format

* again

* more correct test. Had to remove optimization although it almost works

* Also test aligned=True

* remove useless part

* more docs and comments

* Put back optimization with more robust test

* Added check for index upper bound

* avoid possible overflow

* Move common function into common.h

* oops

* scale=1,zero_point=0 makes more sense

* Force batch size of 1 to prevent any indexingbug

* format

* format again

* updated docstring

* put back description comment for pre_calc_bilinear_interpolate

* revert most changes to docstring as it's taken care of in another PR
      ad9cc62a
  25. 06 Apr, 2021 1 commit
  27. 30 Mar, 2021 5 commits
  29. 25 Mar, 2021 1 commit
  31. 24 Mar, 2021 2 commits
    • Philip Meier's avatar
      add tests for MNIST and variants (#3423) · aa365993
      Philip Meier authored 
      * add tests for MNIST and variants

* remove old tests and fakedata generation

* fix default config detection for if dataset has variable keywords

* use split="mnist" as default for EMNIST

* fix QMNIST tests

* lint

* fix special kwargs detection

* Revert "use split="mnist" as default for EMNIST"

This reverts commit 62c9b23597f4a391cad409cbd93edac1b3474acf.

* fix tests

* fix QMNIST test case name

* remove dead code from test

* Revert "remove old tests and fakedata generation"

This reverts commit a285b97c4827566a5bc06c288f5bba8d614a4f7a.

* remove old tests

* readd removed import
      aa365993
    • Philip Meier's avatar
      Improve error handling in make_dataset (#3496) · 0818c682
      Philip Meier authored 
      * factor out find_classes

* use find_classes in video datasets

* adapt old tests
      0818c682
  33. 22 Mar, 2021 2 commits
  35. 20 Mar, 2021 1 commit
  37. 19 Mar, 2021 1 commit
  39. 17 Mar, 2021 2 commits
  41. 16 Mar, 2021 2 commits
  43. 15 Mar, 2021 1 commit