mirror of https://github.com/tensorflow/addons.git
3380b3ccf9
Keras 3.0 will become default in TF 2.16 (and is currently default in tf-nightly). This breaks this tf-addons package. Here we make minimal changes to return functionality in a backward-compatible way. |
||
|---|---|---|
| .. | ||
| tests | ||
| BUILD | ||
| README.md | ||
| __init__.py | ||
| color_ops.py | ||
| compose_ops.py | ||
| connected_components.py | ||
| cutout_ops.py | ||
| dense_image_warp.py | ||
| distance_transform.py | ||
| distort_image_ops.py | ||
| filters.py | ||
| interpolate_spline.py | ||
| resampler_ops.py | ||
| sparse_image_warp.py | ||
| transform_ops.py | ||
| translate_ops.py | ||
| utils.py | ||
README.md
Addons - Image
Components
https://www.tensorflow.org/addons/api_docs/python/tfa/image
Contribution Guidelines
Standard API
In order to conform with the current API standard, all image ops must:
- Be a standard image processing technique
- Must be impossible to implement in one of the other API standards (Layers, Losses, etc.).
Testing Requirements
- Simple unittests that demonstrate the image op is behaving as expected.
- To run your
tf.functionsin eager mode and graph mode in the tests, you can use the@pytest.mark.usefixtures("maybe_run_functions_eagerly")decorator. This will run the tests twice, once normally, and once withtf.config.run_functions_eagerly(True).
Documentation Requirements
- Update the CODEOWNERS file