TensorFlow Lattice v2.1.1

Changes:

• Bug fixes for tutorial imports.
• Bug fixes for lattice layer build/load.

PyPI Release:

• Generic package for py3 and TF 2.x.

TensorFlow Lattice v2.1.0

Changes:

• Updating keras imports to use tf_keras (for TF >= 2.16 tf.keras defaults to Keras-3).
• Deprecating Estimators and Visualization library (removed in TF >= 2.16).
• All docs and tutorials converted to use Keras premade models instead of Estimators.

PyPI Release:

• Generic package for py3 and TF 2.x.

TensorFlow Lattice 2.0.13

Changes:

• Updates to the visualization library to support newer versions of matplotlib.

PyPI Release:

• Generic package for py3 and TF 2.x.

TensorFlow Lattice 2.0.12

Changes:

• Newly added conditional-CDF layer: A CDF function with derived parameters. This is similar to conditional-PWL, but the parameters are the shifts in the input space.
• Updates to model save/load functions to match changes in Keras lib.
• Removing Sonnet PWL module to simplify dependencies. The layer library code can still be wrapped in Sonnet modules by the user as needed.

PyPI Release:

• Generic package for py3 and TF 2.x.

TensorFlow Lattice 2.0.11

Changes:

• Updating code, tests and tutorials to support changes to tf.keras.optimizers.
• Documentation updates.
• Minor bug fixes.

PyPI Release:

• Generic package for py3 that should work for TF 1.15 or TF 2.x.

TensorFlow Lattice 2.0.10

Changes:

• Support for weighted quantiles for Estimators and Premade.
• Helper functions for computing quantiles in premade_lib
• Documentation updates.
• Minor bug fixes.

PyPI Release:

• Generic package for py3 that should work for TF 1.15 or TF 2.x.

TensorFlow Lattice 2.0.9

Changes:

• (experimental) Cumulative Distribution Function (CDF) Layer that supports projection free monotonicity.
• 'input_keypoints_type' parameter for PWLCalibration integration with Premade/Estimator models.
• Estimator support for tf.data.Dataset inputs.
• General tutorial/code cleanup.
• Typo fixes.
• Bug fixes.

PyPI Release:

• Generic package for py3 that should work for TF 1.15 or TF 2.x.

TensorFlow Lattice 2.0.8

Changes:

• (experimental) Parameterization option for Premade/Estimators that enables the use of both normal tfl.layers.Lattice layers ('all_vertices') and tfl.layers.KroneckerFactoredLattice layers ('kronecker_factored').
• (experimental) KroneckerFactoredLattice layer visualization support for Estimators.
• (experimental) KroneckerFactoredLattice bound constraints.
• 'input_keypoints_type' parameter for PWLCalibration layers that enables learned input keypoints ('learned_interior') or the original fixed keypoints ('fixed').
• General tutorial/code cleanup
• Typo fixes
• Bug fixes

PyPI Release:

• Generic package for py3 that should work for TF 1.15 or TF 2.x.

TensorFlow Lattice 2.0.7

Changes:

• (experimental) KroneckerFactoredLattice initialization now sorts on kernel axis 1 such that we sort each term individually.
• (experimental) KroneckerFactoredLattice initialization defaults to [0.5, 1.5] instead of [0,1].
• (experimental) KroneckerFactoredLattice custom_reduce_prod in interpolation for faster gradient computations.
• Update bound and trust projection algorithms to compute violations for each unit separately.
• 'loss_fn' option for estimators to use custom loss without having to define a custom head.
• Enable calibrators to return a list of outputs per unit.
• Enable RTL layer to return non-averaged outputs.
• General tutorial/code cleanup
• Typo fixes
• Bug fixes

PyPI Release:

• Generic package for py3 that should work for TF 1.15 or TF 2.x.

TensorFlow Lattice 2.0.6 [Note: last py2 compatible release]

TensorFlow is dropping py2 support, so we will be dropping support as well in our future releases. This is the last release that will support py2.

Changes:

• New (experimental) KroneckerFactoredLattice Layer, which introduces a new parameterization of our Lattice layer with linear space/time complexity.
• rtl_lib.py helper functions for RTL Layer.
• Utils module with useful helper functions for all layers.
• 'rtl_layer' option for CalibratedLatticeEnsemble Premade Models and Canned Estimators, which uses an RTL Layer for the underlying ensemble. Can potentially give a speed-boost for models with a large number of lattices.
• General code cleanup
• Typo fixes
• Bug fixes

PyPI release:

• Generic package for py2/py3 that should work for TF 1.15 or TF 2.x.