v1.1.1

What's Changed

• Introduce TestCase and ensure mixed precision works with all layers by @james77777778 in #137

Full Changelog: v1.1.0...v1.1.1

v1.1.0

What's Changed

• Add ToDType and fix bugs in RandAugment and TrivialAugmentWide by @james77777778 in #136

Full Changelog: v1.0.1...v1.1.0

v1.0.1

What's Changed

• Update guides and README by @james77777778 in #132
• Fix app.py by @james77777778 in #133
• Add p argument to RandAugment and TrivialAugmentWide and fix bug in RandomApply by @james77777778 in #134
• Update app.py by @james77777778 in #135

Full Changelog: v1.0.0...v1.0.1

v1.0.0

Breaking Change

Now, KerasAug is compatible with Keras 3 and tf.data.Dataset with all backends!

Note that the new APIs largly follow torchvision and the correctness of all layers has been verified through unit tests.

v0.5.8

What's Changed

• Fix streamlit requirements by @james77777778 in #124
• Fix tf.where inconsistent dtype bug by @james77777778 in #126

Full Changelog: v0.5.7...v0.5.8

v0.5.7

KerasAug

Major Features and Improvements

• Add support of new released Tensorflow 2.13 and Keras 2.13

What's Changed

• Update benchmarks/ by @james77777778 in #110
• Update streamlit app with segmentation masks support by @james77777778 in #111
• Fix streamlit app by @james77777778 in #113
• Fix streamlit app by @james77777778 in #114
• fix streamlit app by @james77777778 in #115
• Fix streamlit app by @james77777778 in #116
• Upgrade to TF2.13 and fix RandomGenerator bug by @james77777778 in #118
• Streamlit app sticks to tensorflow 2.12 until keras-cv is updated by @james77777778 in #120
• Release v0.5.7 by @james77777778 in #122

Full Changelog: v0.5.6...v0.5.7

v0.5.6

KerasAug

Major Features and Improvements

• Drop KerasCV dependency

What's Changed

• Drop KerasCV dependency by @james77777778 in #107
• Release v0.5.6 by @james77777778 in #109

Full Changelog: v0.5.5...v0.5.6

v0.5.5

KerasAug

Major Features and Improvements

• Add support of the mixed precision
  • keras.mixed_precision.set_global_policy("mixed_float16")
  • keras.mixed_precision.set_global_policy("mixed_bfloat16")

What's Changed

• Fix numerical issuse when using mixed_precision by @james77777778 in #102
• Release v0.5.5 by @james77777778 in #104

Full Changelog: v0.5.4...v0.5.5

v0.5.4

KerasAug

Major Features and Improvements

• Add Streamlit App

Visit https://james77777778-keras-aug-streamlit-appapp-mxd7v1.streamlit.app to try the layers of KerasAug!

What's Changed

• Use OIDC for PyPI by @james77777778 in #85
• Create demo site with streamlit by @james77777778 in #87
• Update README by @james77777778 in #89
• Add position and padding_value to RandomZoomAndCrop by @james77777778 in #92
• Add bounding boxes support to streamlit app by @james77777778 in #93
• Fix streamlit app by @james77777778 in #95
• Fix streamlit app by @james77777778 in #97
• Release v0.5.4 by @james77777778 in #99

Full Changelog: v0.5.3...v0.5.4

v0.5.3

What's Changed

• Fix shape with graph mode by @james77777778 in #73
• Add segmentation masks support to Mosaic by @james77777778 in #75
• Remove tensorflow_probability dependency by @james77777778 in #77
• Migrate pytest, setup.py and setup.cfg to pyproject.toml by @james77777778 in #80
• Allow dense bounding boxes by @james77777778 in #81
• Release v0.5.3 by @james77777778 in #83

Full Changelog: v0.5.2...v0.5.3