GitLab

* disable test

* disable profiling

* Update test_onnx.py

* remove changes that induced BC

* Re-enable tests that have been disabled

* Remove outdated comment

* Remove outdated comment

Rename with_bias() to bias(), and output_channels() to out_channels() in C++ conv layer options usage (#1576)

* enable faster rcnn test

* flake8

* smaller image size

* set min/max

* Simlify and organize test_ops.

We perform the following:

- Simplify the functions slow_roi_pooling, slow_ps_roi_pooling, slow_ps_roi_align and bilinear_interpolate (including finding and removing a semi-bug in slow_ps_roi_pooling, which used bin_w instead of bin_h);
- Wrote a slow_roi_align function, that was missing;
- Create a base class testing all combinations of forward/backward, cpu/cuda, contiguous/non-contiguous;
- Organize all testing inside the base class with _test_forward and _test_backward (which can be easily overriden if a parciular op needs something different); an Op class then only needs to implement fn, get_script_fn, and expected_fn.

A few points:
- We are using the same inputs for all tests, and not trying all possible inputs in the domain of a given operation. One improvement would be to test more diverse inputs, and to personalize the inputs for some ops (e.g. different inputs for pooling ops and align ops).
- Running all tests is quite slow (~1 min only for CPU tests), so that can possibly be improved.

* Reduce input size used in gradcheck.

gradcheck can be quite costly, and it was causing OOM errors and making
the tests slow. By reducing the size of the input, the test speed is
down to 3 seconds for the CPU tests.

Other points:
- We remove an unused namedtuple;
- We inherit from object for better Python 2 compatibility;
- We remove a hardcoded pool_size from the TorchScript functions, and
add it as a parameter instead.

* Replace Tensor by torch.Tensor in type annotations.

This should fix lint errors.

* Fix inconsistent NMS implementation

* Improve tests for NMS

* Remove unnecessary using statement

* Add Scriptable Transform: Grayscale

* add scriptable transforms: rgb_to_grayscale

* add scriptable transform: rgb_to_grayscale

* add scriptable transform: rgb_to_grayscale

* add scriptable transform: rgb_to_grayscale

* update code: rgb_to_grayscale

* add test: rgb_to_grayscale

* update parameters: rgb_to_grayscale

* add scriptable transform: rgb_to_grayscale

* update rgb_to_grayscale

* update rgb_to_grayscale

* quantizable googlenet

* Minor improvements

* Rename basic_conv2d with conv_block plus additional fixes

* More renamings and fixes

* Bugfix

* Fix missing import for mypy

* Add pretrained weights

* Disable C++ models from being compiled without explicitly being asked for

* Fix import in tests, which are already disabled

* Unify video metadata in VideoClips

* Bugfix

* Make tests a bit more robust

* Improve readability of affine transformation code

* Make shear transformation area preserving

The previous shear implementation did not preserve area, and we
implement a version that does.

The formula used was verified with the following sympy code:

from sympy import Matrix, cos, sin, tan, simplify
from sympy.abc import x, y, phi

Xs = Matrix(
        [[1, -tan(x)],
         [0, 1]]
        )

Ys = Matrix(
        [[1, 0],
         [-tan(y), 1]]
        )

R = Matrix(
        [[cos(phi), -sin(phi)],
         [sin(phi), cos(phi)]]
        )

RSS = Matrix(
        [[cos(phi - y)/cos(y), -cos(phi - y)*tan(x)/cos(y) - sin(phi)],
         [sin(phi - y)/cos(y), -sin(phi - y)*tan(x)/cos(y) + cos(phi)]])

print(simplify(R * Ys * Xs - RSS))

One thing that is not clear (and could be tested) is whether avoiding
the explicit products and calculations in _get_inverse_affine_matrix
really gives performance benefits - compared to doing the explicit
calculation done in _test_transformation.

* Use np.matmul instead of @

The @ syntax is not supported in Python 2.

* Support Exporting Mask Rcnn to ONNX

* update tetst

* add control flow test

* fix

* update test and fix img_shape

* add quantized models

* Modify mobilenet.py documentation and clean up comments
Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Move fuse_model method to QuantizableInvertedResidual and clean up args documentation
Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Restore relu settings to default in resnet.py
Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix missing return in forward
Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix missing return in forwards
Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Change pretrained -> pretrained_float_models
Replace InvertedResidual with block

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Update tests to follow similar structure to test_models.py, allowing for modular testing

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Replace forward method with simple function assignment

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix error in arguments for resnet18

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* pretrained_float_model argument missing for mobilenet

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* reference script for quantization aware training and post training quantization

* reference script for quantization aware training and post training quantization

* set pretrained_float_model as False and explicitly provide float model

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Address review comments:
1. Replace forward with _forward
2. Use pretrained models in reference train/eval script
3. Modify test to skip if fbgemm is not supported

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix lint errors.
Use _forward for common code between float and quantized models
Clean up linting for reference train scripts
Test over all quantizable models

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Update default values for args in quantization/train.py

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Update models to conform to new API with quantize argument
Remove apex in training script, add post training quant as an option
Add support for separate calibration data set.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix minor errors in train_quantization.py

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Remove duplicate file

* Bugfix

* Minor improvements on the models

* Expose print_freq to evaluate

* Minor improvements on train_quantization.py

* Ensure that quantized models are created and run on the specified backends
Fix errors in test only mode

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Add model urls

* Fix errors in quantized model tests.
Speedup creation of random quantized model by removing histogram observers

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Move setting qengine prior to convert.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix lint error

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Add readme.md

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Readme.md

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix lint

* Add adjustment operations for RGB Tensor Images.

Right now, we have operations on PIL images, but we want to have a version of the opeartions that act directly on Tensor images.

Here, we add such operations for adjust_brightness, adjust_contrast and adjust_saturation.

In PIL, those functions are implemented by generating an degenerate image from the first, and then interpolating them together.
- https://github.com/python-pillow/Pillow/blob/master/src/PIL/ImageEnhance.py
- https://github.com/python-pillow/Pillow/blob/master/src/libImaging/Blend.c

A few caveats:
* Since PIL operates on uint8, and the tensor operations might be on float, we can get slightly different values because of int truncation.
* We assume here the images are RGB; in particular, to handle an alpha channel, we need to check whether it is present, in which case we copy it to the final image.

* Keep dtype and use broadcast in adjust operations

- We make our operations have input.dtype == output.dtype, at the cost of
adding a few type checks and branches.

- By using Tensor broadcast, we can simplify the calls to _blend.

* Use is_floating_point to check dtype.

* Remove unpacking in tuple

It seems Python 2 does not support this type of unpacking, so it broke
Python 2 builds. This should fix it.

* Add from __future__ import division for Python 2

* Initial version of README for classification reference scripts

* More context

* test: Updated asserts in test_utils

Updated all raw asserts to corresponding unittest.TestCase.assert. See #1483

* style: Fixed lint check

* Removed unnecessary class variables.

* The integrity of dataset files is now being checked right after the download finished. Thus making sure that a corrupt file is not being extracted. In case of corruption we throw a RuntimeError.

* Added missing md5 hashes to MNIST, FashionMNIST, KMNIST, EMNIST and QMNIST datasets.

* Removed printing of error message when integrity check failed.
Reformulated error message.

* Reformatted code to be lint conform.

* Fixed formatting in utils.py

* Unify video backend interfaces

* Remove reference cycle

* Make functions private and enable tests on OSX

* Disable test if video_reader backend not available

* Lint

* Fix import after refactoring

* Fix lint

* test: Updated asserts in test_onnx

Updated all raw asserts to corresponding unittest.TestCase.assert. See #1483

* test: Refactored AssertionError

Opted for cleaner raise to avoid error type casting and string conversion

* extend DistributedSampler to support group_size

* Fix lint

* correctness test implemented with old test architecture

* reverted an unneeded change, ran flake8

* moving to relative tolerance of 1 part in 10k for classification correctness checks

* going down to 1 part in 1000 for correctness checks bc architecture differences

* one percent relative tolerance

* Try vs2019 toolchain

* ver num

* pass env var

* fix search logic

* Support both VS2017 and VS2019

* Some small fixes

* remove download process

* address comments

* fix logic error

* bug fixes

* removed unused import

* add docstrings

* flake8

* remove download BC

* fix test

* removed unused code

* flake 8

* add MD5 verification before extraction

* add mock to test

* * unify _verify_archive() method and function
* remove force flag for parse_*_archive functions
* cleanup

* flake8

* test: Updated asserts in test_io

Updated all raw asserts to corresponding unittest.TestCase.assert. See #1483

* test: Fixed test_io

In order to use unittest.TestCase for asserts in temp_video, the object has to be passed itself to temp_video.

* test: Updated asserts to AssertionError

Switched initial raw assert to AssertionError.

* test: Changed AssertionError for ValueError

Updated all raw asserts to corresponding unittest.TestCase.assert. See #1483