add __repr__ and __getitem__ to vector #2163
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
johnnv1
pushed a commit
to johnnv1/kornia
that referenced
this pull request
Apr 5, 2023
* add __repr__ and __getitem__ to vector * silly mistake fix
edgarriba
added a commit
that referenced
this pull request
Sep 22, 2023
* Fixe DoG accuracy, add `upscale_double` (#2105)
* Difference of gaussians fixes/improvments
* right upscaling
* DoG test
* removing unused vars
* remove matplotlib
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* removed PIL
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fixing test
* PR comments reflected
* test_dog updated
dog test update
dog test update
dog test update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* removed unused files, better import
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update test/geometry/transform/test_pyramid.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* TestUpscaleDouble: more illustrative checks
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fixing test path
* deleted too heavy-weight test (see https://github.com/kornia/kornia-benchmark/pull/7)
* assert_close
* fixing test
* added tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/geometry/transform/pyramid.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update test/geometry/transform/test_pyramid.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* dtype and device params provided
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update documentation
* TestUpscaleDouble fixed
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update `nightly` labeled condition (#2140)
* update labeled condition
* rerun CI
to test if will run the nightly again
* improve `TestUpscaleDouble` (#2147)
- implement cardinality test
- use `assert_close`
- fix gradcheck test
- enable fast mode
* fix adalam tests (#2145)
* add `fail-fast:false` as default on tests workflow (#2146)
* Added Face detection Interactive demo (#2142)
* Bump pytest from 7.2.0 to 7.2.1 (#2148)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.2.0 to 7.2.1.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.2.0...7.2.1)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* add docs (#2150)
* Explicitly cast output to input type to avoid type mismatch errors (#1842)
* cast transformation matrix and output to input type to avoid type mismatch errors (fixes #1737)
* code style
* extend test and check seed behavior
* also add explicit type conversion in MixAugmentationBase class
* fix autocast test, be more explicit about the desired batch_prob
* select only needed sigma values
* skip test if autocast is not available
* revoke changes in gaussian blur, fix autocast test
* Fix params computation for `LongestMaxSize` and `SmallestMaxSize` (#2131)
* Fix the side resize for `LongestMaxSize`
* Fix the side resize for `SmallestMaxSize`
* Fix param random generator
* torch_version_geq -> torch_version_ge according to todo (#2157)
* torch_version_geq -> torch_version_ge according to todo
* fix doc build - `sphinx-autodoc-typehints==1.21.3` (#2159)
The 1.21.4 version is crashing the documentation build
* ScaleSpaceDetector -> Fast ScaleSpaceDetector (#2154)
* ScaleSpaceDetector ->MultiResolutionDetector
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Improve losses tests, add `TestSSIM3d`, and `BaseTester.gradcheck` (#2152)
* Improve `TestSSIM3DLoss`
- Add BaseTester
- Implement unit test
* add `gradcheck` on `BaseTester`
gradcheck as a method of BaseTester to avoid need to pass `fast_mode=True` on each gradcheck test
* add test for `metrics.SSIM3d`
`kornia.metrics.SSIM3D` was added in #2130, and here are the tests
* improve `TestWelschLoss`
* improve `TestCauchyLoss`
* improve `TestGemanMcclureLossLoss`
- remove unecessary tests from `TestCauchyLoss`
* improve `TestCharbonnierLoss`
* Update docs
* fix typing
* add non uniform shape to smoke test of `TestSSIM3DLoss`
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* modify comments of rgb and lab conversion (#2153)
* modify comments of rgb and lab conversion
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* add __repr__ and __getitem__ to vector (#2163)
* add __repr__ and __getitem__ to vector
* silly mistake fix
* Augmentation Base Refactor (#2117)
* refactor
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Updated augmentation base.
* Removed label & start apply/inverse
* Finished augmentaion base refactor
* container refactoring
* Added missing files
* Added ops
* Update sequential ops
* Almost there
* Fixed computation matrix computation
* Fixed randomcrop
* Fixed erasing
* almost almost
* finished
* Added missing file
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Bug and typing fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Added list typing
* fixed test base
* Fixed typing
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* bug fix
* Update kornia/augmentation/_2d/geometric/crop.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fixed build-docs
* Fixed bfloat16 issue on torch1.13.1
* Revert the last commit
* Fixed typing
* Fixed typos
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Typo fix
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* unpin `sphinx-autodoc-typehints` (#2166)
* Fix adalam-config (#2170)
* Fix adalam-config https://github.com/kornia/kornia/issues/2161
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Fix docs of `boxes`, `MultiResolutionDetector` and `apply colormap` (#2167)
* fix boxes table
* fix `MultiResolutionDetector` doc
* fix apply colormap doc indexing
* add exception test for se2 + small bug fix (#2160)
* add exception test for se2, small bug fix
* Update test/geometry/liegroup/test_se2.py
* rem asserts
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] pre-commit suggestions (#2174)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/PyCQA/docformatter: v1.5.1 → v1.6.0.rc1](https://github.com/PyCQA/docformatter/compare/v1.5.1...v1.6.0.rc1)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/feature/integrated.py
Just to rerun pre-commit CI
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix MobileViT (#2172)
* Fix output types of augmentations on autocast regions (#2168)
* update `.type` casts on base augmentation class
- remove all casts around the bases augmentations classes
- add cast on `transform_inputs` of `_AugmentationBase` when autocast region is enabled
* add autocast test to `GeometricAugmentationBase2D`
* add `is_autocast_enabled` with old torch compat
- add to docs the map_location_to_cpu
* add aug 2d and 3d base tests
* Fix autocast for geometric 2d
* update `eye_like` to cast type after construction
* add test sequential container 2D
* fix autocast for AugmentationSequential container
* fix AugmentationSequential container
- add `.type` method to Keypoints and Boxes
* add test for `VideoSequential`, `PatchSequential`
* Update `is_autocast_enabled`
* revert typechecking statement on is_autocast_enabled
* Fix planckian jitter for cuda (#2177)
* fix test
* fix planckian jitter coefs
- add lazyload to the coefs calculus
- move coef tensor to input device
* fix typing
* [enhance] improve flipping and cropping speed (#2179)
* init
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* restore wrong file
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Replace jit test method in favor of dynamo in `BaseTester` (#2120)
* Replace jit test with dynamo test in `BaseTester`
* remove macOS of nightly tests -- it can be added back when https://github.com/pytorch/pytorch/pull/89262 be merged
* downgrade docformatter to 1.5.1 (#2176)
* downgrade docformatter to 1.5.1
* Update .pre-commit-config.yaml
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Default resample method to None, if not in flags (#2185)
* [pre-commit.ci] pre-commit suggestions (#2194)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/PyCQA/isort: 5.11.4 → 5.12.0](https://github.com/PyCQA/isort/compare/5.11.4...5.12.0)
--------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Move padding_size to device (#2197)
* Return boxes tensor directly if no boxes (#2196)
* Return boxes tensor directly if no boxes
* Add more cases to bbox test
* The isinstance checks order are inverted (#2192)
* Skip `x` tests for `torch=1.12.1` and `accelerate` not available (#2178)
* Skip `x` tests if torch=1.12.1 and accelerate not available
* Drop some ci job (#2191)
* Drop quantity of jobs for scheduled CI
* Drop quantity of jobs for PR CI
* drop CI for fp16
* Make value an attribute of RandomErasing instances as it used to be (#2195)
* TensorWrapper bug fix + add __radd__, __rmul__, __rsub__ (#2190)
* TensorWrapper bug fix + add __radd__, __rnum__, __rsub__
* add tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fp64 tests just on ubuntu and nightly just dynamo test (#2199)
* Small refactor on `filters` module: Dropping JIT support (#2187)
* add suport to Tensor for sigmas
- Removed the functions `gaussian_blur2d_t`, `get_gaussian_kernel1d_t`, `get_gaussian_kernel2d_t`, and
`get_gaussian_kernel3d_t` in favor of support sigmas as `Tensors`, floats, or tuple of floats on the functions without
the suffix `_t`
* Remove crashing JIT tests related to filters
* fix typing
* add support to kernel size as tuple or integer
* update laplacian test
* update DTYPES atol and rtol
- added bfloat16
- use same values for BaseTester class and the function
* add device and dtype for default gaussian
* add device and dtype for discrete gaussian by erf
* add device, dtype, batched for discrete gaussian
- add support to compute batched kernerl for discrete gaussian by bessel functions
- add support to pass device and dtype when sigma is a float
* add device and dtype for laplacian kernels
* add device and dtype for binary and box kernels
* add new typping annotations
- by adding `from __future__ import annotations` we can use the pyupgrade
to update the typing annotations to be like on python 3.10
* add device and dtype for static kernels (sobel)
* add dtype and device for canny
- fix gaussian_blur2d sigma shape
* add KORNIA_CHECK API to filters
* fix typing and docs
* fix laplacian module kernel_size typing
* remove unsqueeze in favor of pythonic slicing
* update kernel size canny module
* add depreciation for `*_t` functions
* update blur tests
* add noncontiguous tests to blur
* update canny tests
- add BaseTester
- add Dynamo test
* update gaussian tests
- Add basetester
- add dynamo tests
* update hanning tests
* update laplacian tests
- Add BaseTester
- Add dynamo test
* update median tests
- add BaseTester
- add dynamo test
* remove border parameter
- this should be tested just on `filter2d`, `filter3d`, etc
* update motion tests
- Add BaseTester
- Add dynamo
- Add tests for 3D
* update sobel tests
- Add basetester
- add dynamo tests
* update unsharp mask test
- add basetester
- add dynamo test
* update filter 2d and 3d tests
- add basetester
- add dynamo
* fix typing
* remove Pyr down and up jit tests
* update atol for fp64 based on old _DTYPE_PRECISIONS values
* add TODO note about dtype precision for fp64
* skip filter3d with reflect border for < 1.9 torch
* small fix
* Bump accelerate from 0.15.0 to 0.16.0 (#2202)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.15.0 to 0.16.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.15.0...v0.16.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2203)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/psf/black: 22.12.0 → 23.1.0](https://github.com/psf/black/compare/22.12.0...23.1.0)
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Add `integral_image` and `integral_tensor` (#1779)
* Integral of a tensor
* Integral of tensor
* Integral changes
* Changes
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update integral.py
* implement tests
* downgrade docformatter to 1.5.1
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* fix typing
* fix typing
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: Dmytro Mishkin <ducha.aiki@gmail.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* move kornia check api to kornia.core.logger (#2143)
* move kornia check api to kornia.core.logger
* add initial tests
* Fix/repr bug (#2207)
* changed all __repr__ to fstring
* fixed test fails
---------
Co-authored-by: Neyaz Basheer <neyaz.basheer@appliedagi.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Replace `assert_allclose` by `assert_close` (#2210)
- The `torch.testing.assert_allclose` is deprecated in favor of `torch.testing.assert_close`
* Fix random crop for keypoints on CUDA device (#2209)
- The Keypoints pad and unpad requires the padding size to be on the same device. Before this patch the tests above
are failling on cuda:
- test/augmentation/test_container.py::TestAugmentationSequential::test_random_crops
- test/augmentation/test_container.py::TestAugmentationSequential::test_individual_forward_and_inverse
* Remove outdated augmentation example (#2206)
* DataKey: add 'image' as alias of 'input' (#2193)
* DataKey: add 'image' as alias of 'input'
* Document new 'image' key
* Test image in addition to input
* flake8 fix
* Remove py 3.7 for nightly CI (#2204)
- Deprecation of CUDA 11.6 and Python 3.7 Support - from https://pytorch.org/blog/deprecation-cuda-python-support/
* [Feat] Initiate AutoAugment modules (#2181)
* init
* Added autoaugment
* Added RandAugment
* Added trivial augment
* Added missing files
* Updated docs
* refactoring
* Added missing files
* Added shear and translate functions
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Fixed translate bug
* Added color
* Added autocontrast
* Fix
* Fix doctest
* Fixed typing
* Removed legacy generators
* fix
* fixed cutmix float64 bug
* include more test
* Sequential support first commit
* Added missing files
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* first refactor
* Make it compatible with AugmentationSequential
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Added more tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make ops module lazy loading
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make policy module lazy loading
* Lazy loading auto module
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make augmentation module lazy loading
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Bug fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* final fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Revert review commnets
* Update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix CUDA failling tests of same device (#2215)
* Release 0.6.10 (#2212)
* downgrade docformatter to 1.5.1
* release 0.6.10
* Update setup.cfg
* Update contributing guide (#2217)
* Update contributing guide
* fix branch name
* update typing example for use kornia.core
* review changes
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Apply suggestions from the review
---------
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Fix `RandomGaussianBlur` for sigma as tensor (#2220)
* Separate gradcheck for test_conversions (#2223)
* refactor out gradcheck tests for geometry.conversions
* rename
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* fix
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* MPS tests exclude dynamo
* Drop JIT support for `core.check`, `Boxes`, and others (#2219)
* Drop JIT support for `core.check` API
- Consequently for this, we drop support of JIT on the following items: (in of dynamo)
- enhance
- AdjustSigmoid
- AdjustLog
- AddWeighted
- geometry
- UndistortPoints
- bbox and Boxes - follow up on #2218
- EuclideanDistance
- TransformPoints
- HomographyWarper
- WarpPerspective
- UpscaleDouble
- losses
* Update typing with pyupgrade
* drop all jit related from bbox and boxes
from #2218
* fix/skip failing dynamo tests
* fix loss hd
* fix typing
* PEP621: Move more configuration into pyproject.toml (#2225)
* PEP621: Move more configuration into pyproject.toml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* pre-commit: Add validate-pyproject
* pre-commit: Add https://github.com/tox-dev/pyproject-fmt
* Update pyproject.toml
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update pyproject.toml
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* pyproject-fmt --indent 4
* "4", not 4
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* four space indentation
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* enable `disallow_incomplete_defs` on mypy (#2094)
* enable `disallow_incomplete_defs` on mypy
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix `blur_pool2d` doc
* finish v1: works on torch 1.13.1
- Remove JIT support for Boxes3D
* rip off the np typing
* replace `Size` with `Tuple[int, ...]` on augs
* add `Dtype` to kornia.filters.kernels
* minor fix after rebase
* Remove old torch from typing CI
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Update tensor equality tests to use`assert_close()` (#2233)
* change .all() to assert_close()
* Feat/random median blur (#2234)
* add median_blur.py
* Add class for tests
* add RandomMedianBlur to init files
* Add test median blur
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* updated generate_example_images.py
---------
Co-authored-by: karyna <karyna.volokhatiuk@ucu.edu.ua>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Clean up metadata and fix requirements (#2232)
* Clenaup metadata and fix requirements
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove test files from source
* fix typo
* fix docs exclude pattern
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Feat/random snow (#2229)
* Removed unused variable.
* Delete random_snow.py
* Added class, the instance of ; implemented __init__ and apply_transform.
* Added documentation.
* Added the testing of random snow generation, also added RandomSnow to init files
* Reverted the commit with testing of random snow generation, in order to fix some bugs.
* Added the testing of random snow generation, also added RandomSnow to init files.
* Added comments to the RandomSnow module
* Fixed typing in documentation.
* Slightly modified the existing tests and added several new
* Removed in-place operations; used param_generator; adjusted documentation.
* Blind fix of tensor unmatching.
* Changed condition.
* Fixed an error in the exception test.
* Tensor match fix.
* Adjusted light channel range and snow_coefficient.
* Removed imports.
* Removed unimplemented jit test
* Made parameters as tuples and changed tests accordingly.
* Made parameters as tuples and changed tests accordingly.
* Fixed check.
* Rolled back apply_transform implementation.
* Fixed gradcheck.
* assinged correct channel for operations; now different parameters samples for each image in the batch.
* fixed gradcheck.
* condensced code.
* fixed doctest.
* Added several new test cases to the exception test.
* Removed loop.
* Deleted temp.onnx
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Fixed tests.
---------
Co-authored-by: Severyn12 <s.peleshko@ucu.edu.ua>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2236)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.9.1 → 0.9.2](https://github.com/tox-dev/pyproject-fmt/compare/0.9.1...0.9.2)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* add random snow and median blur to docs (#2238)
* Use more pythonic expressions in rgb to hls (#2235)
* replaced `torch.select` with more pythonic expressions
* replaced `torch.add` with +
* replaced torch arithmetic op with python op
* fix typo on doc of `RandAugment` (#2243)
* Add bilateral filter (#2242)
* Add bilateral filter
* Feature LAF docs fix (#2245)
* fix docstrings in kornia.feature.laf
* fix docsstrings in affine and ori as well
* Apply suggestions from code review
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Drop JIT support for `geometry.subpix` (#2253)
- Remove `kornia.testing.check_is_tensor` in favor of
`kornia.core.check.KORNIA_CHECK_IS_TENSOR`
* Remove unused files (#2251)
* Add Joint Bilateral Filter (#2249)
* initial commit
* add tests
* add checks for image size
* fix mypy typing
* fix docstring
* add docs
* update tests
* add tests against opencv
* add image to doc and minor changes
* update error message in tests
* Remove Re-definition found for builtin input function - Update tests (#2255)
* change input to sample in test/feature folder
* change input to sample in test/filters folder
* change input to sample in test/morphology folder
* change input to sample in test/geometry/subpix/test_spatial_softargmax.py
* change input to sample in test/geometry/transfrom in test_imgwarp3d.py and test_pyramid.py
* change input to input_org in test/geometry/transfrom in test_imgwarp.py
* change input to inpt in test/grad_estimator/test_ste.py
* change input to inpt and sample in test/enhance/ folder
* change input to input_tensor in test/augmentation/test_augmentation_3d.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: oleksadobush <oleksandra.stasiuk@ucu.edu.ua>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Bump pytest from 7.2.1 to 7.2.2 (#2256)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.2.1 to 7.2.2.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.2.1...7.2.2)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* fix extract_patches and consequent bugs (#2262)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* hotfix
* Remove typecheck for 2D kernel size on filters (#2259)
* Bump accelerate from 0.16 to 0.17.0 (#2265)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.16 to 0.17.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.16.0...v0.17.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Feat/dice loss averaging (#2264)
average by default takes "micro" value and leaves the current behavior unchanged. Also, added another option "macro" that calculates Dice Loss for each class separately and then averages between classes and samples.
* added averaging option for dice loss
* added tests for dice loss averaging param
* fixed typing
* more convenient checks
* Update kornia/losses/dice.py
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix SOLD2 matching on CUDA (#2270)
* Fix Normalize with integer inputs (#2269)
* Fix Normalize with integer inputs
* improve typing
---------
Co-authored-by: edgar <edgar.riba@gmail.com>
* Add warnings as error on documentation build (#2273)
* add flag to sphinx build
* move the flag for the CI
* fix docstrings
* remove unsupported analitics
* Bump accelerate from 0.17.0 to 0.17.1 (#2280)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.17.0 to 0.17.1.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.17.0...v0.17.1)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* add batch_squared_norm to kornia/geometry/linalg __all__ (#2279)
* remove numpy conversion (#2277)
* Update release workflow (#2278)
* update the release workflow
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix/diamond square normalize (#2283)
* Add test for normalize_range argument
* Fix error on normalization
* Fix type annotation for normalize_range argument
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Specify device and dtype in expected outputs.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Add Random rain augmentation (#2268)
* Add Random rain augmentation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* New augmentation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Random rain augmentation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Random Rain
* Random rain augmentation
* Refactoring random rain augmentation
* Refactor assert
* Refactor generator
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Add typehints for `kornia.geometry.linalg.inverse_transformation` (#2286)
* add typehints for kornia.geometry.linalg.inverse_transformation
* change torch.Tensor to kornia.core.Tensor
* Bump accelerate from 0.17.1 to 0.18.0 (#2289)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.17.1 to 0.18.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.17.1...v0.18.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Integrating DISK (#2285)
* Re-unify DISK and Disk, add more tests
* Remove the random sampling option in DISK, leaving only NMS keypoints
* Remove the Detector class from DISK implementation; add docstrings; adjust error messages; remove DISK.detect in favor of DISK.forward
* Make DISK.from_pretrained load directly to chosen device; reintroduce a test for loading pretrained models
* added heatmap to the docs and other responces with images, added pad_if_not_divisible
* unhardcode float32
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Dmytro Mishkin <ducha.aiki@gmail.com>
* Add releases 0.6.6 - 0.6.10 to Changelog (#2295)
* fix the DISK padding bug 32 -> 16 (#2294)
* fix the padding bug 32 -> 16
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* expose DISKFeatures structure (#2297)
* added convolution/correlation switch to filter2d (#2271)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Disk features docs (#2299)
* Adjust docstring for DISKFeatures
* Add docstrings for DISKFeatures methods; fix a display issue with DISK.heatmap_and_dense_descriptors docstring
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Remove print factor (#2291)
* make adalam work w/o scale and ori (#2300)
* Bump version for 0.6.11 release (#2296)
* Update __init__.py
* semantic versioning for v0.6.12-dev
https://semver.org/
* Create face_detection.rst (#2308)
* Fix elastic transformation if only partially applied (#2303)
* remove argument description for removed deprecated parameter
* return input if transformation should not be applied for elastic transform
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* use batch prob as index for assertion
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix current typing errors (#2305)
* remove numpy links from image utils (#2306)
* Update io.rst to use `ImageLoadType` (#2309)
* add sub-bin precision to improve local feature orientation (#2310)
- Added subpix interpolation of the angle, improving the the angular consistency
- Tuned constants and fix the bug of not using weighting kernel
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2266)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/psf/black: 23.1.0 → 23.3.0](https://github.com/psf/black/compare/23.1.0...23.3.0)
- [github.com/codespell-project/codespell: v2.2.2 → v2.2.4](https://github.com/codespell-project/codespell/compare/v2.2.2...v2.2.4)
- [github.com/PyCQA/bandit: 1.7.4 → 1.7.5](https://github.com/PyCQA/bandit/compare/1.7.4...1.7.5)
* fix codespell
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* fix: Fix links to CONTRIBUTING.md (#2312)
* Fix readme links (#2311)
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Bump pytest from 7.2.2 to 7.3.0 (#2316)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.2.2 to 7.3.0.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.2.2...7.3.0)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Update nerf test to use `kornia/data` (#2319)
* delete data nerf test
* move nerf test to use kornia/data
* Fix pos_weight in focal loss (#2323)
* fix referenced shape dim on input for check on pos_weight
* fix test for focal loss with pos_weight
* Fix bugs in Bilateral filter tests (#2320)
* remove unused arguments
* fix Tensor sigmas for BilateralBlur module
* use default atol and rtol. update opencv reference
* add gradcheck for sigmas
* Add Guided filter (#2322)
* add tests and API
* add single channel impl
* add fastpath when guidance = input
* updates
- use correct variable names
- change border type to replicate to match OpenCV and original implementation
- add test for grayscale
* revert border mode to reflect. add placeholder for fast guided filter
* add notes about OpenCV test
* implement multi-channel guided filter
* clean up code. all tests pass
* add missing output type
* remove fast bilateral blur. add some docs
* add guided filter bib source
* add fast guided filter
* minor code clean up
* fix type check
* make mypy happy
* add some tests for subsample
* update docs
* always register eps as buffer
* test gradcheck for eps
* minor fixes
* Bump pytest from 7.3.0 to 7.3.1 (#2327)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.3.0 to 7.3.1.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.3.0...7.3.1)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Fix kernel size ordering (#2326)
* fix kernel size order
* fix more kernel ordering
* update url collect_env.py (#2329)
* update url collect_env.py
* modify action too
* update imagegen (#2331)
* add keypoints to the docs (#2330)
* fix missing geometry.keypoints docs (#2332)
* [pre-commit.ci] pre-commit suggestions (#2318)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/PyCQA/docformatter: v1.5.1 → v1.6.0](https://github.com/PyCQA/docformatter/compare/v1.5.1...v1.6.0)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* bump docformatter to `v1.6.1-rc1` and minor fix
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Ensure support to torch 2.0 (#2272)
* Add torch 2.0 to CI
* Update .github/workflows/pr_test_typing.yml
* fix typing
* remove cast and fix type
* skip dynamo crashing tests
* skip onxx export test for RandomRGBShift
* update optimizer setup to use api
* skip dynamo tests for platforms other than linux
* skip dynamo test for color gray on torch 2.0.0
* skip dynamo test for canny on torch 2.0.0
* skip `TestImageRegistrator::test_registration_real` on win for torch 2.0
tests raising "Tensor-likes are not close!"
* mark as xfail the dynamo test of `WarpPerspective` for float64 on ubuntu with torch 2
* skip dynamo test for `LovaszSoftmaxLoss`
* Use separable filter2d for box filter (#2328)
* use separable filter for box filter
* remove normalized. add some tests for separable
* import get_box_kernel1d() to kornia.filters
* update doc
* check input. remove unnecessary if
* add docstring
---------
Co-authored-by: edgar <edgar.riba@gmail.com>
* [feat] add segment anything base (#2315)
* add sam model
* add predictor
* add docstrings TODO note
* update device mapping
* Update the model loader and fix predictor
* fix model device and bbox prompt, add docstring
- add predictor docstrings
- add SamPrediction
* add boxes conversion
* fix datakey typing
* fix init tfs
* fix typping
* add to the docs
* move docs to "models" section and add card
* rm wrong static data on docs
* fix preproc `Keypoints` on `AugmentationSequential`
Co-authored-by: Jian Shi <sj8716643@126.com>
* move to absolute import
* add support to tensor for points and box
* remove model device property
* reorder preprocess method
* move build to be a method of the model
* fix typo docs example
* add p=1 for resize tf
* remove if device
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* update sam forward
Sam now returns only low-res logits, and expects the input image and prompts to be preprocessed
- Remove pre and post processing from sam arch
- Modify sam prediction to use sam forward
* add imageprompter
* update prompter, add tests and ref links
* add prompter tests and update docs
* init `contrib.models`
* `PrompterModelBase`
* fix future annotations
* unify the prompter API to `ImagePrompter`
* remove fp64 tests
* update docs and import levels
* fix path of SamConfig import on example
* rename prompter to `image_prompter`
* move sam to `contrib.models`
* remove squeeze output prompter
* fix docs
* change input image of prompter to range [0-1]
* fix example on doctest without squeeze
* add hardcoded official checkpoint
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: Jian Shi <sj8716643@126.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* make api docs more visible (#2334)
* Fix params name on `ImagePrompter` example (#2340)
* bump version 0.6.12 (#2339)
* fix skipped tests for cuda `TestColorJiggleGen` (#2341)
* remove inplace operation (#2346)
* remove inplace operation
remove inplace update of rotation matrix in angle_axis_to_rotation_matrix
* Update conversions.py
* Replace bandit, flake8, isort, pyupgrade, and yesqa with ruff (#2292)
* Replace bandit, flake8, isort, pyupgrade, and yesqa with ruff
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* fix unused import on `geometry.conversions` (#2357)
* [pre-commit.ci] pre-commit suggestions (#2345)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/tox-dev/pyproject-fmt: 0.9.2 → 0.10.0](https://github.com/tox-dev/pyproject-fmt/compare/0.9.2...0.10.0)
- [github.com/PyCQA/docformatter: v1.6.1-rc1 → v1.6.3](https://github.com/PyCQA/docformatter/compare/v1.6.1-rc1...v1.6.3)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Visual prompter (#2356)
* rename ImagePrompter to VisualPrompter
* add Visual Prompting task page
* remove images
* Update docs/source/applications/visual_prompting.rst
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Update docs/source/contrib.rst
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Integrate vector to liegroups (#2344)
* integrate vector to liegroups
* minor
* integrate vec to lg, add tests
* typing, docs fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/geometry/liegroup/se2.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* add __future__.annotations
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* review changes
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* add from_wxyz and from_qxyz to So3 and Se3 (#2359)
* [pre-commit.ci] pre-commit suggestions (#2360)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/tox-dev/pyproject-fmt: 0.10.0 → 0.11.1](https://github.com/tox-dev/pyproject-fmt/compare/0.10.0...0.11.1)
- [github.com/charliermarsh/ruff-pre-commit: v0.0.259 → v0.0.263](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.259...v0.0.263)
- [github.com/PyCQA/docformatter: v1.6.3 → v1.6.4](https://github.com/PyCQA/docformatter/compare/v1.6.3...v1.6.4)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Fix line lenght
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* RandomGrayscale: add rgb_weights parameter (#2361)
* add support for tuple of types on KORNIA_CHECK_TYPE + bug fix (#2353)
* add support for tuple of types on KORNIA_CHECK_TYPE + bug fix on test_check.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/core/check.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* todo
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Remove examples in favor of `kornia/tutorials` repo (#2366)
All the examples are available in our [tutorials repository](https://github.com/kornia/tutorials)
* add tutorials testing ci (#2367)
* add tutorials ci testing
* fix typo of the info about the workflow
* Update version to 0.6.13-dev (#2368)
* [pre-commit.ci] pre-commit suggestions (#2370)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.11.1 → 0.11.2](https://github.com/tox-dev/pyproject-fmt/compare/0.11.1...0.11.2)
- [github.com/charliermarsh/ruff-pre-commit: v0.0.263 → v0.0.265](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.263...v0.0.265)
- [github.com/PyCQA/docformatter: v1.6.4 → v1.6.5](https://github.com/PyCQA/docformatter/compare/v1.6.4...v1.6.5)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Optional raise in kornia check functions (#2375)
* Add raises args to KORNIA_CHECK
* Add raises arg to KORNIA_CHECK_IS_COLOR
* Add raises arg to KORNIA_CHECK_IS_GRAY
* Add raises arg to KORNIA_CHECK_SHAPE
* Add raises arg to KORNIA_CHECK_TYPE
* Add raises arg to KORNIA_CHECK_SAME_DEVICE
* Add raises arg to KORNIA_CHECK_SAME_DEVICES
* Add missing asserts
* Add raises arg to KORNIA_CHECK_SAME_SHAPE
* Add raises arg to KORNIA_CHECK_IS_TENSOR
* Add consistent error handling to KORNIA_CHECK_IS_LIST_OF_TENSOR
* Add raises arg to KORNIA_CHECK_LAF
* Add raises arg to KORNIA_CHECK_DM_DESC
* Add raises arg to KORNIA_CHECK_IS_COLOR_OR_GRAY
* Fix wrong docstrings and return types
* Run code formatter, fix duplicate test name
* Fix core check doctests
---------
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* remove unused circle ci (#2378)
* Bump accelerate from 0.18.0 to 0.19.0 (#2381)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.18.0 to 0.19.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.18.0...v0.19.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Replace miniconda for `setup-python@v4` on env setup CI (#2380)
* [pre-commit.ci] pre-commit suggestions (#2382)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.265 → v0.0.267](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.265...v0.0.267)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2388)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.267 → v0.0.269](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.267...v0.0.269)
- [github.com/PyCQA/docformatter: v1.6.5 → v1.7.1](https://github.com/PyCQA/docformatter/compare/v1.6.5...v1.7.1)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/feature/adalam/core.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/contrib/vit.py
* Apply suggestions from code review
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Use separable filter for SSIM calculation to speed up. (#2383)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
* Bump pytest-cov from 4 to 4.1.0 (#2392)
Bumps [pytest-cov](https://github.com/pytest-dev/pytest-cov) from 4 to 4.1.0.
- [Changelog](https://github.com/pytest-dev/pytest-cov/blob/master/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest-cov/compare/v4.0.0...v4.1.0)
---
updated-dependencies:
- dependency-name: pytest-cov
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2394)
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.269 → v0.0.270](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.269...v0.0.270)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Batched quaternion to matrix (#2395)
* Test multiple leading dimensions in quaternion_to_rotation_matrix tests: (B*, 4) and not just (B, 4)
* Implement arbitrarily batched quaternion_to_rotation_matrix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix docs
* Test for unbatched quaternion conversion
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: edgar <edgar.riba@gmail.com>
* Batched draw_lines (#2373)
* batched draw
* update tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* docs
* review changes
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Fix AugmentationSequential support for RandomResizedCrop (#2398)
* drop python 3.7 support (#2400)
* drop python 3.7 support
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* add type check
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Bump accelerate from 0.19.0 to 0.20.3 (#2402)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.19.0 to 0.20.3.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.19.0...v0.20.3)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Bump pytest from 7.3.1 to 7.3.2 (#2401)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.3.1 to 7.3.2.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.3.1...7.3.2)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Makefile: Drop Python code formatter yapf in favor of psf/black (#2404)
As discussed at https://github.com/kornia/kornia/pull/2358#issuecomment-1546592237
* https://github.com/search?q=repo%3Akornia%2Fkornia+yapf&type=code
* https://github.com/search?q=repo%3Akornia%2Fkornia+black&type=code
* update the kornia governance page (#2403)
* Update readthedocs.yml with python 3.8 (#2406)
* [pre-commit.ci] pre-commit suggestions (#2408)
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.270 → v0.0.272](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.270...v0.0.272)
- [github.com/PyCQA/docformatter: v1.7.1 → v1.7.2](https://github.com/PyCQA/docformatter/compare/v1.7.1...v1.7.2)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* add collector for CI on PR's and remove unused CI (#2379)
- Add typing tests as a reusable workflow
- Use it on test CPU workflows
- Remove specific typing workflows
- Remove unused changelog CI
- Remove unused test cuda CI
- Rename PyPI package test CI
- Rename PR workflow and PyPI workflow
* remove old artefacts in visual_prompting.rst (#2409)
* use reshape instead view to convert matrix from quaternion (#2413)
* Revamped camera API (introduce kornia.sensors) (#2349)
* init
* add distortion, projection models
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* minor updates
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* updates
* updates
* updates
* api structure experiment
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/sensor/camera/camera.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* review changes, add kornia.image,
* Update kornia/image/__init__.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* finish up pinhole cam, add tests
* mypy fix
* minor updates, add docs etc
* add docs
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* review changes
* add sensor init
* review changes
* docs fix
* mark Unnecessary tests
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] pre-commit suggestions (#2415)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/tox-dev/pyproject-fmt: 0.11.2 → 0.12.0](https://github.com/tox-dev/pyproject-fmt/compare/0.11.2...0.12.0)
- [github.com/codespell-project/codespell: v2.2.4 → v2.2.5](https://github.com/codespell-project/codespell/compare/v2.2.4...v2.2.5)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2418)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.12.0 → 0.12.1](https://github.com/tox-dev/pyproject-fmt/compare/0.12.0...0.12.1)
- [github.com/charliermarsh/ruff-pre-commit: v0.0.272 → v0.0.275](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.272...v0.0.275)
- [github.com/PyCQA/docformatter: v1.7.2 → v1.7.3](https://github.com/PyCQA/docformatter/compare/v1.7.2...v1.7.3)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Bump pytest from 7.3.2 to 7.4.0 (#2416)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.3.2 to 7.4.0.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.3.2...7.4.0)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Fix PLR0124 (#2422)
* Ignore PLR0124
* Circumvent PLR0124
* Add missing docs for warp_grid (#2423)
* Add GPU support apple silicon augmentation 2D (#2425)
* Add GPU support apple silicon
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* init.py fix
* planckian_jitter update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* planckian_jitter fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/utils/helpers.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* added OS and platform check in utils helper
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Add mps to testing init and fixing test cases (#2428)
* Small fix utils init.py
* add mps in test init
* bring back mps and cuda
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* test skip if macos
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update test
* add python version
* small python version fix
* add condition to TestFindHomographyDLTIter
* test fix
* test fix
* test try
* test fix
* used torch_version_le instead
* test
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* test
* test fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update test/geometry/test_homography.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove if else
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* drop JIT support from `geometry.conversions` (#2424)
* drop JIT support from `geometry.conversions`
* mark xfail on cpu `TestDenormalizePixelCoordinates` dynamo
* Update rst(docs) files to support MPS (#2430)
* Update rst file to support MPS
* Update segment_anything.rst
* Update segment_anything.rst
* [pre-commit.ci] pre-commit suggestions (#2429)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.12.1 → 0.13.0](https://github.com/tox-dev/pyproject-fmt/compare/0.12.1...0.13.0)
- https://github.com/charliermarsh/ruff-pre-commit → https://github.com/astral-sh/ruff-pre-commit
- [github.com/astral-sh/ruff-pre-commit: v0.0.275 → v0.0.276](https://github.com/astral-sh/ruff-pre-commit/compare/v0.0.275...v0.0.276)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fix engine arg missing in opening and closing (#2431)
* Added terminal printing (#2407)
* Added terminal printing
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* added docs
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* Update kornia/utils/image_print.py
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Update test/utils/test_print.py
* Update test/utils/test_print.py
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* add support for quaternion in Se3 instantiation (#2433)
* add quat support
* docs
* fix miss assertion test geometry conversion(#2435)
Signed-off-by: Qingpeng Li <qingpeng9802@gmail.com>
* Add RT-DETR and ObjectDetection API (#2363)
* add resnet-d backbone
* fix impl. add small test
* use kornia core names
* add hybrid encoder
* add test for hybrid encoder. fix bugs
* fix sequential.append
* fix sequential.append
* add query selector init
* add anchors
* add more decode logic
* add deformable attention
* rename
* add note about caching
* add the custom decoder
* add forward test
* add some model wrapper
* rename modules so it's easier for renaming weights
* add RepVGG and rename for easier weights conversion
* restructure to make …
cjpurackal
added a commit
to cjpurackal/kornia
that referenced
this pull request
Oct 20, 2023
* Fixe DoG accuracy, add `upscale_double` (#2105)
* Difference of gaussians fixes/improvments
* right upscaling
* DoG test
* removing unused vars
* remove matplotlib
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* removed PIL
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fixing test
* PR comments reflected
* test_dog updated
dog test update
dog test update
dog test update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* removed unused files, better import
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update test/geometry/transform/test_pyramid.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* TestUpscaleDouble: more illustrative checks
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fixing test path
* deleted too heavy-weight test (see https://github.com/kornia/kornia-benchmark/pull/7)
* assert_close
* fixing test
* added tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/geometry/transform/pyramid.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update test/geometry/transform/test_pyramid.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* dtype and device params provided
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update documentation
* TestUpscaleDouble fixed
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update `nightly` labeled condition (#2140)
* update labeled condition
* rerun CI
to test if will run the nightly again
* improve `TestUpscaleDouble` (#2147)
- implement cardinality test
- use `assert_close`
- fix gradcheck test
- enable fast mode
* fix adalam tests (#2145)
* add `fail-fast:false` as default on tests workflow (#2146)
* Added Face detection Interactive demo (#2142)
* Bump pytest from 7.2.0 to 7.2.1 (#2148)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.2.0 to 7.2.1.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.2.0...7.2.1)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* add docs (#2150)
* Explicitly cast output to input type to avoid type mismatch errors (#1842)
* cast transformation matrix and output to input type to avoid type mismatch errors (fixes #1737)
* code style
* extend test and check seed behavior
* also add explicit type conversion in MixAugmentationBase class
* fix autocast test, be more explicit about the desired batch_prob
* select only needed sigma values
* skip test if autocast is not available
* revoke changes in gaussian blur, fix autocast test
* Fix params computation for `LongestMaxSize` and `SmallestMaxSize` (#2131)
* Fix the side resize for `LongestMaxSize`
* Fix the side resize for `SmallestMaxSize`
* Fix param random generator
* torch_version_geq -> torch_version_ge according to todo (#2157)
* torch_version_geq -> torch_version_ge according to todo
* fix doc build - `sphinx-autodoc-typehints==1.21.3` (#2159)
The 1.21.4 version is crashing the documentation build
* ScaleSpaceDetector -> Fast ScaleSpaceDetector (#2154)
* ScaleSpaceDetector ->MultiResolutionDetector
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Improve losses tests, add `TestSSIM3d`, and `BaseTester.gradcheck` (#2152)
* Improve `TestSSIM3DLoss`
- Add BaseTester
- Implement unit test
* add `gradcheck` on `BaseTester`
gradcheck as a method of BaseTester to avoid need to pass `fast_mode=True` on each gradcheck test
* add test for `metrics.SSIM3d`
`kornia.metrics.SSIM3D` was added in #2130, and here are the tests
* improve `TestWelschLoss`
* improve `TestCauchyLoss`
* improve `TestGemanMcclureLossLoss`
- remove unecessary tests from `TestCauchyLoss`
* improve `TestCharbonnierLoss`
* Update docs
* fix typing
* add non uniform shape to smoke test of `TestSSIM3DLoss`
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* modify comments of rgb and lab conversion (#2153)
* modify comments of rgb and lab conversion
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* add __repr__ and __getitem__ to vector (#2163)
* add __repr__ and __getitem__ to vector
* silly mistake fix
* Augmentation Base Refactor (#2117)
* refactor
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Updated augmentation base.
* Removed label & start apply/inverse
* Finished augmentaion base refactor
* container refactoring
* Added missing files
* Added ops
* Update sequential ops
* Almost there
* Fixed computation matrix computation
* Fixed randomcrop
* Fixed erasing
* almost almost
* finished
* Added missing file
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Bug and typing fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Added list typing
* fixed test base
* Fixed typing
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* bug fix
* Update kornia/augmentation/_2d/geometric/crop.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fixed build-docs
* Fixed bfloat16 issue on torch1.13.1
* Revert the last commit
* Fixed typing
* Fixed typos
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Typo fix
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* unpin `sphinx-autodoc-typehints` (#2166)
* Fix adalam-config (#2170)
* Fix adalam-config https://github.com/kornia/kornia/issues/2161
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Fix docs of `boxes`, `MultiResolutionDetector` and `apply colormap` (#2167)
* fix boxes table
* fix `MultiResolutionDetector` doc
* fix apply colormap doc indexing
* add exception test for se2 + small bug fix (#2160)
* add exception test for se2, small bug fix
* Update test/geometry/liegroup/test_se2.py
* rem asserts
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] pre-commit suggestions (#2174)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/PyCQA/docformatter: v1.5.1 → v1.6.0.rc1](https://github.com/PyCQA/docformatter/compare/v1.5.1...v1.6.0.rc1)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/feature/integrated.py
Just to rerun pre-commit CI
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix MobileViT (#2172)
* Fix output types of augmentations on autocast regions (#2168)
* update `.type` casts on base augmentation class
- remove all casts around the bases augmentations classes
- add cast on `transform_inputs` of `_AugmentationBase` when autocast region is enabled
* add autocast test to `GeometricAugmentationBase2D`
* add `is_autocast_enabled` with old torch compat
- add to docs the map_location_to_cpu
* add aug 2d and 3d base tests
* Fix autocast for geometric 2d
* update `eye_like` to cast type after construction
* add test sequential container 2D
* fix autocast for AugmentationSequential container
* fix AugmentationSequential container
- add `.type` method to Keypoints and Boxes
* add test for `VideoSequential`, `PatchSequential`
* Update `is_autocast_enabled`
* revert typechecking statement on is_autocast_enabled
* Fix planckian jitter for cuda (#2177)
* fix test
* fix planckian jitter coefs
- add lazyload to the coefs calculus
- move coef tensor to input device
* fix typing
* [enhance] improve flipping and cropping speed (#2179)
* init
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* restore wrong file
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Replace jit test method in favor of dynamo in `BaseTester` (#2120)
* Replace jit test with dynamo test in `BaseTester`
* remove macOS of nightly tests -- it can be added back when https://github.com/pytorch/pytorch/pull/89262 be merged
* downgrade docformatter to 1.5.1 (#2176)
* downgrade docformatter to 1.5.1
* Update .pre-commit-config.yaml
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Default resample method to None, if not in flags (#2185)
* [pre-commit.ci] pre-commit suggestions (#2194)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/PyCQA/isort: 5.11.4 → 5.12.0](https://github.com/PyCQA/isort/compare/5.11.4...5.12.0)
--------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Move padding_size to device (#2197)
* Return boxes tensor directly if no boxes (#2196)
* Return boxes tensor directly if no boxes
* Add more cases to bbox test
* The isinstance checks order are inverted (#2192)
* Skip `x` tests for `torch=1.12.1` and `accelerate` not available (#2178)
* Skip `x` tests if torch=1.12.1 and accelerate not available
* Drop some ci job (#2191)
* Drop quantity of jobs for scheduled CI
* Drop quantity of jobs for PR CI
* drop CI for fp16
* Make value an attribute of RandomErasing instances as it used to be (#2195)
* TensorWrapper bug fix + add __radd__, __rmul__, __rsub__ (#2190)
* TensorWrapper bug fix + add __radd__, __rnum__, __rsub__
* add tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fp64 tests just on ubuntu and nightly just dynamo test (#2199)
* Small refactor on `filters` module: Dropping JIT support (#2187)
* add suport to Tensor for sigmas
- Removed the functions `gaussian_blur2d_t`, `get_gaussian_kernel1d_t`, `get_gaussian_kernel2d_t`, and
`get_gaussian_kernel3d_t` in favor of support sigmas as `Tensors`, floats, or tuple of floats on the functions without
the suffix `_t`
* Remove crashing JIT tests related to filters
* fix typing
* add support to kernel size as tuple or integer
* update laplacian test
* update DTYPES atol and rtol
- added bfloat16
- use same values for BaseTester class and the function
* add device and dtype for default gaussian
* add device and dtype for discrete gaussian by erf
* add device, dtype, batched for discrete gaussian
- add support to compute batched kernerl for discrete gaussian by bessel functions
- add support to pass device and dtype when sigma is a float
* add device and dtype for laplacian kernels
* add device and dtype for binary and box kernels
* add new typping annotations
- by adding `from __future__ import annotations` we can use the pyupgrade
to update the typing annotations to be like on python 3.10
* add device and dtype for static kernels (sobel)
* add dtype and device for canny
- fix gaussian_blur2d sigma shape
* add KORNIA_CHECK API to filters
* fix typing and docs
* fix laplacian module kernel_size typing
* remove unsqueeze in favor of pythonic slicing
* update kernel size canny module
* add depreciation for `*_t` functions
* update blur tests
* add noncontiguous tests to blur
* update canny tests
- add BaseTester
- add Dynamo test
* update gaussian tests
- Add basetester
- add dynamo tests
* update hanning tests
* update laplacian tests
- Add BaseTester
- Add dynamo test
* update median tests
- add BaseTester
- add dynamo test
* remove border parameter
- this should be tested just on `filter2d`, `filter3d`, etc
* update motion tests
- Add BaseTester
- Add dynamo
- Add tests for 3D
* update sobel tests
- Add basetester
- add dynamo tests
* update unsharp mask test
- add basetester
- add dynamo test
* update filter 2d and 3d tests
- add basetester
- add dynamo
* fix typing
* remove Pyr down and up jit tests
* update atol for fp64 based on old _DTYPE_PRECISIONS values
* add TODO note about dtype precision for fp64
* skip filter3d with reflect border for < 1.9 torch
* small fix
* Bump accelerate from 0.15.0 to 0.16.0 (#2202)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.15.0 to 0.16.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.15.0...v0.16.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2203)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/psf/black: 22.12.0 → 23.1.0](https://github.com/psf/black/compare/22.12.0...23.1.0)
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Add `integral_image` and `integral_tensor` (#1779)
* Integral of a tensor
* Integral of tensor
* Integral changes
* Changes
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update kornia/enhance/integral.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Update integral.py
* implement tests
* downgrade docformatter to 1.5.1
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* fix typing
* fix typing
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: Dmytro Mishkin <ducha.aiki@gmail.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* move kornia check api to kornia.core.logger (#2143)
* move kornia check api to kornia.core.logger
* add initial tests
* Fix/repr bug (#2207)
* changed all __repr__ to fstring
* fixed test fails
---------
Co-authored-by: Neyaz Basheer <neyaz.basheer@appliedagi.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Replace `assert_allclose` by `assert_close` (#2210)
- The `torch.testing.assert_allclose` is deprecated in favor of `torch.testing.assert_close`
* Fix random crop for keypoints on CUDA device (#2209)
- The Keypoints pad and unpad requires the padding size to be on the same device. Before this patch the tests above
are failling on cuda:
- test/augmentation/test_container.py::TestAugmentationSequential::test_random_crops
- test/augmentation/test_container.py::TestAugmentationSequential::test_individual_forward_and_inverse
* Remove outdated augmentation example (#2206)
* DataKey: add 'image' as alias of 'input' (#2193)
* DataKey: add 'image' as alias of 'input'
* Document new 'image' key
* Test image in addition to input
* flake8 fix
* Remove py 3.7 for nightly CI (#2204)
- Deprecation of CUDA 11.6 and Python 3.7 Support - from https://pytorch.org/blog/deprecation-cuda-python-support/
* [Feat] Initiate AutoAugment modules (#2181)
* init
* Added autoaugment
* Added RandAugment
* Added trivial augment
* Added missing files
* Updated docs
* refactoring
* Added missing files
* Added shear and translate functions
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Fixed translate bug
* Added color
* Added autocontrast
* Fix
* Fix doctest
* Fixed typing
* Removed legacy generators
* fix
* fixed cutmix float64 bug
* include more test
* Sequential support first commit
* Added missing files
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* first refactor
* Make it compatible with AugmentationSequential
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Added more tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make ops module lazy loading
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make policy module lazy loading
* Lazy loading auto module
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make augmentation module lazy loading
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Bug fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* final fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Revert review commnets
* Update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix CUDA failling tests of same device (#2215)
* Release 0.6.10 (#2212)
* downgrade docformatter to 1.5.1
* release 0.6.10
* Update setup.cfg
* Update contributing guide (#2217)
* Update contributing guide
* fix branch name
* update typing example for use kornia.core
* review changes
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Apply suggestions from the review
---------
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Fix `RandomGaussianBlur` for sigma as tensor (#2220)
* Separate gradcheck for test_conversions (#2223)
* refactor out gradcheck tests for geometry.conversions
* rename
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
* fix
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* MPS tests exclude dynamo
* Drop JIT support for `core.check`, `Boxes`, and others (#2219)
* Drop JIT support for `core.check` API
- Consequently for this, we drop support of JIT on the following items: (in of dynamo)
- enhance
- AdjustSigmoid
- AdjustLog
- AddWeighted
- geometry
- UndistortPoints
- bbox and Boxes - follow up on #2218
- EuclideanDistance
- TransformPoints
- HomographyWarper
- WarpPerspective
- UpscaleDouble
- losses
* Update typing with pyupgrade
* drop all jit related from bbox and boxes
from #2218
* fix/skip failing dynamo tests
* fix loss hd
* fix typing
* PEP621: Move more configuration into pyproject.toml (#2225)
* PEP621: Move more configuration into pyproject.toml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* pre-commit: Add validate-pyproject
* pre-commit: Add https://github.com/tox-dev/pyproject-fmt
* Update pyproject.toml
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update pyproject.toml
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* pyproject-fmt --indent 4
* "4", not 4
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* four space indentation
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* enable `disallow_incomplete_defs` on mypy (#2094)
* enable `disallow_incomplete_defs` on mypy
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix `blur_pool2d` doc
* finish v1: works on torch 1.13.1
- Remove JIT support for Boxes3D
* rip off the np typing
* replace `Size` with `Tuple[int, ...]` on augs
* add `Dtype` to kornia.filters.kernels
* minor fix after rebase
* Remove old torch from typing CI
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Update tensor equality tests to use`assert_close()` (#2233)
* change .all() to assert_close()
* Feat/random median blur (#2234)
* add median_blur.py
* Add class for tests
* add RandomMedianBlur to init files
* Add test median blur
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* updated generate_example_images.py
---------
Co-authored-by: karyna <karyna.volokhatiuk@ucu.edu.ua>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Clean up metadata and fix requirements (#2232)
* Clenaup metadata and fix requirements
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove test files from source
* fix typo
* fix docs exclude pattern
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Feat/random snow (#2229)
* Removed unused variable.
* Delete random_snow.py
* Added class, the instance of ; implemented __init__ and apply_transform.
* Added documentation.
* Added the testing of random snow generation, also added RandomSnow to init files
* Reverted the commit with testing of random snow generation, in order to fix some bugs.
* Added the testing of random snow generation, also added RandomSnow to init files.
* Added comments to the RandomSnow module
* Fixed typing in documentation.
* Slightly modified the existing tests and added several new
* Removed in-place operations; used param_generator; adjusted documentation.
* Blind fix of tensor unmatching.
* Changed condition.
* Fixed an error in the exception test.
* Tensor match fix.
* Adjusted light channel range and snow_coefficient.
* Removed imports.
* Removed unimplemented jit test
* Made parameters as tuples and changed tests accordingly.
* Made parameters as tuples and changed tests accordingly.
* Fixed check.
* Rolled back apply_transform implementation.
* Fixed gradcheck.
* assinged correct channel for operations; now different parameters samples for each image in the batch.
* fixed gradcheck.
* condensced code.
* fixed doctest.
* Added several new test cases to the exception test.
* Removed loop.
* Deleted temp.onnx
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Fixed tests.
---------
Co-authored-by: Severyn12 <s.peleshko@ucu.edu.ua>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2236)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.9.1 → 0.9.2](https://github.com/tox-dev/pyproject-fmt/compare/0.9.1...0.9.2)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* add random snow and median blur to docs (#2238)
* Use more pythonic expressions in rgb to hls (#2235)
* replaced `torch.select` with more pythonic expressions
* replaced `torch.add` with +
* replaced torch arithmetic op with python op
* fix typo on doc of `RandAugment` (#2243)
* Add bilateral filter (#2242)
* Add bilateral filter
* Feature LAF docs fix (#2245)
* fix docstrings in kornia.feature.laf
* fix docsstrings in affine and ori as well
* Apply suggestions from code review
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Drop JIT support for `geometry.subpix` (#2253)
- Remove `kornia.testing.check_is_tensor` in favor of
`kornia.core.check.KORNIA_CHECK_IS_TENSOR`
* Remove unused files (#2251)
* Add Joint Bilateral Filter (#2249)
* initial commit
* add tests
* add checks for image size
* fix mypy typing
* fix docstring
* add docs
* update tests
* add tests against opencv
* add image to doc and minor changes
* update error message in tests
* Remove Re-definition found for builtin input function - Update tests (#2255)
* change input to sample in test/feature folder
* change input to sample in test/filters folder
* change input to sample in test/morphology folder
* change input to sample in test/geometry/subpix/test_spatial_softargmax.py
* change input to sample in test/geometry/transfrom in test_imgwarp3d.py and test_pyramid.py
* change input to input_org in test/geometry/transfrom in test_imgwarp.py
* change input to inpt in test/grad_estimator/test_ste.py
* change input to inpt and sample in test/enhance/ folder
* change input to input_tensor in test/augmentation/test_augmentation_3d.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: oleksadobush <oleksandra.stasiuk@ucu.edu.ua>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Bump pytest from 7.2.1 to 7.2.2 (#2256)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.2.1 to 7.2.2.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.2.1...7.2.2)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* fix extract_patches and consequent bugs (#2262)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* hotfix
* Remove typecheck for 2D kernel size on filters (#2259)
* Bump accelerate from 0.16 to 0.17.0 (#2265)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.16 to 0.17.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.16.0...v0.17.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Feat/dice loss averaging (#2264)
average by default takes "micro" value and leaves the current behavior unchanged. Also, added another option "macro" that calculates Dice Loss for each class separately and then averages between classes and samples.
* added averaging option for dice loss
* added tests for dice loss averaging param
* fixed typing
* more convenient checks
* Update kornia/losses/dice.py
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix SOLD2 matching on CUDA (#2270)
* Fix Normalize with integer inputs (#2269)
* Fix Normalize with integer inputs
* improve typing
---------
Co-authored-by: edgar <edgar.riba@gmail.com>
* Add warnings as error on documentation build (#2273)
* add flag to sphinx build
* move the flag for the CI
* fix docstrings
* remove unsupported analitics
* Bump accelerate from 0.17.0 to 0.17.1 (#2280)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.17.0 to 0.17.1.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.17.0...v0.17.1)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* add batch_squared_norm to kornia/geometry/linalg __all__ (#2279)
* remove numpy conversion (#2277)
* Update release workflow (#2278)
* update the release workflow
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix/diamond square normalize (#2283)
* Add test for normalize_range argument
* Fix error on normalization
* Fix type annotation for normalize_range argument
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Specify device and dtype in expected outputs.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Add Random rain augmentation (#2268)
* Add Random rain augmentation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* New augmentation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Random rain augmentation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Random Rain
* Random rain augmentation
* Refactoring random rain augmentation
* Refactor assert
* Refactor generator
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Add typehints for `kornia.geometry.linalg.inverse_transformation` (#2286)
* add typehints for kornia.geometry.linalg.inverse_transformation
* change torch.Tensor to kornia.core.Tensor
* Bump accelerate from 0.17.1 to 0.18.0 (#2289)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.17.1 to 0.18.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.17.1...v0.18.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Integrating DISK (#2285)
* Re-unify DISK and Disk, add more tests
* Remove the random sampling option in DISK, leaving only NMS keypoints
* Remove the Detector class from DISK implementation; add docstrings; adjust error messages; remove DISK.detect in favor of DISK.forward
* Make DISK.from_pretrained load directly to chosen device; reintroduce a test for loading pretrained models
* added heatmap to the docs and other responces with images, added pad_if_not_divisible
* unhardcode float32
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Dmytro Mishkin <ducha.aiki@gmail.com>
* Add releases 0.6.6 - 0.6.10 to Changelog (#2295)
* fix the DISK padding bug 32 -> 16 (#2294)
* fix the padding bug 32 -> 16
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* expose DISKFeatures structure (#2297)
* added convolution/correlation switch to filter2d (#2271)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Disk features docs (#2299)
* Adjust docstring for DISKFeatures
* Add docstrings for DISKFeatures methods; fix a display issue with DISK.heatmap_and_dense_descriptors docstring
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Remove print factor (#2291)
* make adalam work w/o scale and ori (#2300)
* Bump version for 0.6.11 release (#2296)
* Update __init__.py
* semantic versioning for v0.6.12-dev
https://semver.org/
* Create face_detection.rst (#2308)
* Fix elastic transformation if only partially applied (#2303)
* remove argument description for removed deprecated parameter
* return input if transformation should not be applied for elastic transform
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* use batch prob as index for assertion
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Fix current typing errors (#2305)
* remove numpy links from image utils (#2306)
* Update io.rst to use `ImageLoadType` (#2309)
* add sub-bin precision to improve local feature orientation (#2310)
- Added subpix interpolation of the angle, improving the the angular consistency
- Tuned constants and fix the bug of not using weighting kernel
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2266)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/psf/black: 23.1.0 → 23.3.0](https://github.com/psf/black/compare/23.1.0...23.3.0)
- [github.com/codespell-project/codespell: v2.2.2 → v2.2.4](https://github.com/codespell-project/codespell/compare/v2.2.2...v2.2.4)
- [github.com/PyCQA/bandit: 1.7.4 → 1.7.5](https://github.com/PyCQA/bandit/compare/1.7.4...1.7.5)
* fix codespell
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* fix: Fix links to CONTRIBUTING.md (#2312)
* Fix readme links (#2311)
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* Bump pytest from 7.2.2 to 7.3.0 (#2316)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.2.2 to 7.3.0.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.2.2...7.3.0)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Update nerf test to use `kornia/data` (#2319)
* delete data nerf test
* move nerf test to use kornia/data
* Fix pos_weight in focal loss (#2323)
* fix referenced shape dim on input for check on pos_weight
* fix test for focal loss with pos_weight
* Fix bugs in Bilateral filter tests (#2320)
* remove unused arguments
* fix Tensor sigmas for BilateralBlur module
* use default atol and rtol. update opencv reference
* add gradcheck for sigmas
* Add Guided filter (#2322)
* add tests and API
* add single channel impl
* add fastpath when guidance = input
* updates
- use correct variable names
- change border type to replicate to match OpenCV and original implementation
- add test for grayscale
* revert border mode to reflect. add placeholder for fast guided filter
* add notes about OpenCV test
* implement multi-channel guided filter
* clean up code. all tests pass
* add missing output type
* remove fast bilateral blur. add some docs
* add guided filter bib source
* add fast guided filter
* minor code clean up
* fix type check
* make mypy happy
* add some tests for subsample
* update docs
* always register eps as buffer
* test gradcheck for eps
* minor fixes
* Bump pytest from 7.3.0 to 7.3.1 (#2327)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.3.0 to 7.3.1.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.3.0...7.3.1)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Fix kernel size ordering (#2326)
* fix kernel size order
* fix more kernel ordering
* update url collect_env.py (#2329)
* update url collect_env.py
* modify action too
* update imagegen (#2331)
* add keypoints to the docs (#2330)
* fix missing geometry.keypoints docs (#2332)
* [pre-commit.ci] pre-commit suggestions (#2318)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/PyCQA/docformatter: v1.5.1 → v1.6.0](https://github.com/PyCQA/docformatter/compare/v1.5.1...v1.6.0)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* bump docformatter to `v1.6.1-rc1` and minor fix
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Ensure support to torch 2.0 (#2272)
* Add torch 2.0 to CI
* Update .github/workflows/pr_test_typing.yml
* fix typing
* remove cast and fix type
* skip dynamo crashing tests
* skip onxx export test for RandomRGBShift
* update optimizer setup to use api
* skip dynamo tests for platforms other than linux
* skip dynamo test for color gray on torch 2.0.0
* skip dynamo test for canny on torch 2.0.0
* skip `TestImageRegistrator::test_registration_real` on win for torch 2.0
tests raising "Tensor-likes are not close!"
* mark as xfail the dynamo test of `WarpPerspective` for float64 on ubuntu with torch 2
* skip dynamo test for `LovaszSoftmaxLoss`
* Use separable filter2d for box filter (#2328)
* use separable filter for box filter
* remove normalized. add some tests for separable
* import get_box_kernel1d() to kornia.filters
* update doc
* check input. remove unnecessary if
* add docstring
---------
Co-authored-by: edgar <edgar.riba@gmail.com>
* [feat] add segment anything base (#2315)
* add sam model
* add predictor
* add docstrings TODO note
* update device mapping
* Update the model loader and fix predictor
* fix model device and bbox prompt, add docstring
- add predictor docstrings
- add SamPrediction
* add boxes conversion
* fix datakey typing
* fix init tfs
* fix typping
* add to the docs
* move docs to "models" section and add card
* rm wrong static data on docs
* fix preproc `Keypoints` on `AugmentationSequential`
Co-authored-by: Jian Shi <sj8716643@126.com>
* move to absolute import
* add support to tensor for points and box
* remove model device property
* reorder preprocess method
* move build to be a method of the model
* fix typo docs example
* add p=1 for resize tf
* remove if device
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* update sam forward
Sam now returns only low-res logits, and expects the input image and prompts to be preprocessed
- Remove pre and post processing from sam arch
- Modify sam prediction to use sam forward
* add imageprompter
* update prompter, add tests and ref links
* add prompter tests and update docs
* init `contrib.models`
* `PrompterModelBase`
* fix future annotations
* unify the prompter API to `ImagePrompter`
* remove fp64 tests
* update docs and import levels
* fix path of SamConfig import on example
* rename prompter to `image_prompter`
* move sam to `contrib.models`
* remove squeeze output prompter
* fix docs
* change input image of prompter to range [0-1]
* fix example on doctest without squeeze
* add hardcoded official checkpoint
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: Jian Shi <sj8716643@126.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* make api docs more visible (#2334)
* Fix params name on `ImagePrompter` example (#2340)
* bump version 0.6.12 (#2339)
* fix skipped tests for cuda `TestColorJiggleGen` (#2341)
* remove inplace operation (#2346)
* remove inplace operation
remove inplace update of rotation matrix in angle_axis_to_rotation_matrix
* Update conversions.py
* Replace bandit, flake8, isort, pyupgrade, and yesqa with ruff (#2292)
* Replace bandit, flake8, isort, pyupgrade, and yesqa with ruff
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* fix unused import on `geometry.conversions` (#2357)
* [pre-commit.ci] pre-commit suggestions (#2345)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/tox-dev/pyproject-fmt: 0.9.2 → 0.10.0](https://github.com/tox-dev/pyproject-fmt/compare/0.9.2...0.10.0)
- [github.com/PyCQA/docformatter: v1.6.1-rc1 → v1.6.3](https://github.com/PyCQA/docformatter/compare/v1.6.1-rc1...v1.6.3)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Visual prompter (#2356)
* rename ImagePrompter to VisualPrompter
* add Visual Prompting task page
* remove images
* Update docs/source/applications/visual_prompting.rst
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Update docs/source/contrib.rst
---------
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Integrate vector to liegroups (#2344)
* integrate vector to liegroups
* minor
* integrate vec to lg, add tests
* typing, docs fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/geometry/liegroup/se2.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* add __future__.annotations
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* review changes
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* add from_wxyz and from_qxyz to So3 and Se3 (#2359)
* [pre-commit.ci] pre-commit suggestions (#2360)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/tox-dev/pyproject-fmt: 0.10.0 → 0.11.1](https://github.com/tox-dev/pyproject-fmt/compare/0.10.0...0.11.1)
- [github.com/charliermarsh/ruff-pre-commit: v0.0.259 → v0.0.263](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.259...v0.0.263)
- [github.com/PyCQA/docformatter: v1.6.3 → v1.6.4](https://github.com/PyCQA/docformatter/compare/v1.6.3...v1.6.4)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Fix line lenght
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* RandomGrayscale: add rgb_weights parameter (#2361)
* add support for tuple of types on KORNIA_CHECK_TYPE + bug fix (#2353)
* add support for tuple of types on KORNIA_CHECK_TYPE + bug fix on test_check.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/core/check.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* todo
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Remove examples in favor of `kornia/tutorials` repo (#2366)
All the examples are available in our [tutorials repository](https://github.com/kornia/tutorials)
* add tutorials testing ci (#2367)
* add tutorials ci testing
* fix typo of the info about the workflow
* Update version to 0.6.13-dev (#2368)
* [pre-commit.ci] pre-commit suggestions (#2370)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.11.1 → 0.11.2](https://github.com/tox-dev/pyproject-fmt/compare/0.11.1...0.11.2)
- [github.com/charliermarsh/ruff-pre-commit: v0.0.263 → v0.0.265](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.263...v0.0.265)
- [github.com/PyCQA/docformatter: v1.6.4 → v1.6.5](https://github.com/PyCQA/docformatter/compare/v1.6.4...v1.6.5)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Optional raise in kornia check functions (#2375)
* Add raises args to KORNIA_CHECK
* Add raises arg to KORNIA_CHECK_IS_COLOR
* Add raises arg to KORNIA_CHECK_IS_GRAY
* Add raises arg to KORNIA_CHECK_SHAPE
* Add raises arg to KORNIA_CHECK_TYPE
* Add raises arg to KORNIA_CHECK_SAME_DEVICE
* Add raises arg to KORNIA_CHECK_SAME_DEVICES
* Add missing asserts
* Add raises arg to KORNIA_CHECK_SAME_SHAPE
* Add raises arg to KORNIA_CHECK_IS_TENSOR
* Add consistent error handling to KORNIA_CHECK_IS_LIST_OF_TENSOR
* Add raises arg to KORNIA_CHECK_LAF
* Add raises arg to KORNIA_CHECK_DM_DESC
* Add raises arg to KORNIA_CHECK_IS_COLOR_OR_GRAY
* Fix wrong docstrings and return types
* Run code formatter, fix duplicate test name
* Fix core check doctests
---------
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* remove unused circle ci (#2378)
* Bump accelerate from 0.18.0 to 0.19.0 (#2381)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.18.0 to 0.19.0.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.18.0...v0.19.0)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Replace miniconda for `setup-python@v4` on env setup CI (#2380)
* [pre-commit.ci] pre-commit suggestions (#2382)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.265 → v0.0.267](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.265...v0.0.267)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2388)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.267 → v0.0.269](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.267...v0.0.269)
- [github.com/PyCQA/docformatter: v1.6.5 → v1.7.1](https://github.com/PyCQA/docformatter/compare/v1.6.5...v1.7.1)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/feature/adalam/core.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/contrib/vit.py
* Apply suggestions from code review
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Use separable filter for SSIM calculation to speed up. (#2383)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
* Bump pytest-cov from 4 to 4.1.0 (#2392)
Bumps [pytest-cov](https://github.com/pytest-dev/pytest-cov) from 4 to 4.1.0.
- [Changelog](https://github.com/pytest-dev/pytest-cov/blob/master/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest-cov/compare/v4.0.0...v4.1.0)
---
updated-dependencies:
- dependency-name: pytest-cov
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2394)
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.269 → v0.0.270](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.269...v0.0.270)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Batched quaternion to matrix (#2395)
* Test multiple leading dimensions in quaternion_to_rotation_matrix tests: (B*, 4) and not just (B, 4)
* Implement arbitrarily batched quaternion_to_rotation_matrix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix docs
* Test for unbatched quaternion conversion
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: edgar <edgar.riba@gmail.com>
* Batched draw_lines (#2373)
* batched draw
* update tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* docs
* review changes
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Fix AugmentationSequential support for RandomResizedCrop (#2398)
* drop python 3.7 support (#2400)
* drop python 3.7 support
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* add type check
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Bump accelerate from 0.19.0 to 0.20.3 (#2402)
Bumps [accelerate](https://github.com/huggingface/accelerate) from 0.19.0 to 0.20.3.
- [Release notes](https://github.com/huggingface/accelerate/releases)
- [Commits](https://github.com/huggingface/accelerate/compare/v0.19.0...v0.20.3)
---
updated-dependencies:
- dependency-name: accelerate
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Bump pytest from 7.3.1 to 7.3.2 (#2401)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.3.1 to 7.3.2.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.3.1...7.3.2)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Makefile: Drop Python code formatter yapf in favor of psf/black (#2404)
As discussed at https://github.com/kornia/kornia/pull/2358#issuecomment-1546592237
* https://github.com/search?q=repo%3Akornia%2Fkornia+yapf&type=code
* https://github.com/search?q=repo%3Akornia%2Fkornia+black&type=code
* update the kornia governance page (#2403)
* Update readthedocs.yml with python 3.8 (#2406)
* [pre-commit.ci] pre-commit suggestions (#2408)
updates:
- [github.com/charliermarsh/ruff-pre-commit: v0.0.270 → v0.0.272](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.270...v0.0.272)
- [github.com/PyCQA/docformatter: v1.7.1 → v1.7.2](https://github.com/PyCQA/docformatter/compare/v1.7.1...v1.7.2)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* add collector for CI on PR's and remove unused CI (#2379)
- Add typing tests as a reusable workflow
- Use it on test CPU workflows
- Remove specific typing workflows
- Remove unused changelog CI
- Remove unused test cuda CI
- Rename PyPI package test CI
- Rename PR workflow and PyPI workflow
* remove old artefacts in visual_prompting.rst (#2409)
* use reshape instead view to convert matrix from quaternion (#2413)
* Revamped camera API (introduce kornia.sensors) (#2349)
* init
* add distortion, projection models
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* minor updates
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* updates
* updates
* updates
* api structure experiment
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/sensor/camera/camera.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* review changes, add kornia.image,
* Update kornia/image/__init__.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* finish up pinhole cam, add tests
* mypy fix
* minor updates, add docs etc
* add docs
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* review changes
* add sensor init
* review changes
* docs fix
* mark Unnecessary tests
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] pre-commit suggestions (#2415)
* [pre-commit.ci] pre-commit suggestions
updates:
- [github.com/tox-dev/pyproject-fmt: 0.11.2 → 0.12.0](https://github.com/tox-dev/pyproject-fmt/compare/0.11.2...0.12.0)
- [github.com/codespell-project/codespell: v2.2.4 → v2.2.5](https://github.com/codespell-project/codespell/compare/v2.2.4...v2.2.5)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit suggestions (#2418)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.12.0 → 0.12.1](https://github.com/tox-dev/pyproject-fmt/compare/0.12.0...0.12.1)
- [github.com/charliermarsh/ruff-pre-commit: v0.0.272 → v0.0.275](https://github.com/charliermarsh/ruff-pre-commit/compare/v0.0.272...v0.0.275)
- [github.com/PyCQA/docformatter: v1.7.2 → v1.7.3](https://github.com/PyCQA/docformatter/compare/v1.7.2...v1.7.3)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Bump pytest from 7.3.2 to 7.4.0 (#2416)
Bumps [pytest](https://github.com/pytest-dev/pytest) from 7.3.2 to 7.4.0.
- [Release notes](https://github.com/pytest-dev/pytest/releases)
- [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst)
- [Commits](https://github.com/pytest-dev/pytest/compare/7.3.2...7.4.0)
---
updated-dependencies:
- dependency-name: pytest
dependency-type: direct:development
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* Fix PLR0124 (#2422)
* Ignore PLR0124
* Circumvent PLR0124
* Add missing docs for warp_grid (#2423)
* Add GPU support apple silicon augmentation 2D (#2425)
* Add GPU support apple silicon
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* init.py fix
* planckian_jitter update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* planckian_jitter fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update kornia/utils/helpers.py
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* added OS and platform check in utils helper
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Add mps to testing init and fixing test cases (#2428)
* Small fix utils init.py
* add mps in test init
* bring back mps and cuda
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* test skip if macos
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update test
* add python version
* small python version fix
* add condition to TestFindHomographyDLTIter
* test fix
* test fix
* test try
* test fix
* used torch_version_le instead
* test
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* test
* test fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update test/geometry/test_homography.py
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove if else
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edgar Riba <edgar.riba@gmail.com>
* drop JIT support from `geometry.conversions` (#2424)
* drop JIT support from `geometry.conversions`
* mark xfail on cpu `TestDenormalizePixelCoordinates` dynamo
* Update rst(docs) files to support MPS (#2430)
* Update rst file to support MPS
* Update segment_anything.rst
* Update segment_anything.rst
* [pre-commit.ci] pre-commit suggestions (#2429)
updates:
- [github.com/tox-dev/pyproject-fmt: 0.12.1 → 0.13.0](https://github.com/tox-dev/pyproject-fmt/compare/0.12.1...0.13.0)
- https://github.com/charliermarsh/ruff-pre-commit → https://github.com/astral-sh/ruff-pre-commit
- [github.com/astral-sh/ruff-pre-commit: v0.0.275 → v0.0.276](https://github.com/astral-sh/ruff-pre-commit/compare/v0.0.275...v0.0.276)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fix engine arg missing in opening and closing (#2431)
* Added terminal printing (#2407)
* Added terminal printing
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* added docs
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* Update kornia/utils/image_print.py
* Apply suggestions from code review
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* Update test/utils/test_print.py
* Update test/utils/test_print.py
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: João Gustavo A. Amorim <joaogustavoamorim@gmail.com>
* add support for quaternion in Se3 instantiation (#2433)
* add quat support
* docs
* fix miss assertion test geometry conversion(#2435)
Signed-off-by: Qingpeng Li <qingpeng9802@gmail.com>
* Add RT-DETR and ObjectDetection API (#2363)
* add resnet-d backbone
* fix impl. add small test
* use kornia core names
* add hybrid encoder
* add test for hybrid encoder. fix bugs
* fix sequential.append
* fix sequential.append
* add query selector init
* add anchors
* add more decode logic
* add deformable attention
* rename
* add note about caching
* add the custom decoder
* add forward test
* add some model wrapper
* rename modules so it's easier for renaming weights
* add RepVGG and rename for easier weights conversion
* restructure to make …
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Changes
Type of change
Checklist