1.10.0 - 2024-08-21

Fix

• Fix compatibility with TF 2.17.0.
• Fix MacOS build.

1.9.2 - 2024-07-04

Fix

• Fix compatibility with TF 2.16.2.
• Fix build.

Known issues

• The MacOS build is currently not available due to an issue with a dependency.

Fix

• Solve dependency collision of YDF Proto between PYDF and TF-DF.

1.9.0 - 2024-03-12

Fix

• Fix max_depth, early stopping parameter documentation.
• Fix plotting contains conditions.

Features

• Compatibility with TensorFlow 2.16.0.
• Expose new parameter sparse_oblique_max_num_projections.
• Using tf_keras instead tf.keras in examples, documentation.
• Support NAConditions for fast engine.
• Faster model loading for models with many features and dense oblique
  conditions.

Documentation

• Clarified documentation of parameters for oblique splits.

1.9.0rc0 - 2024-02-26

Fix

• Fix max_depth, early stopping parameter documentation.
• Fix plotting contains conditions.

Features

• Compatibility with TensorFlow 2.16.0rc0.
• Compatibility with YDF 1.9.0
• Using tf_keras instead tf.keras in examples, documentation.
• Support NAConditions for fast engine.

1.8.1 - 2023-11-17

Fix

• Regression with Mean Squared Error loss and Mean Average error loss
  incorrectly clamped the gradients, leading to incorrect predictions.

1.8.0 - 2023-11-15

Features

• Compatibility with TensorFlow 2.15.0.
• Inspector support for uplifting.

Fix

• Less warnings emitted when loading a model.

1.6.0 2023-09-27

Breaking Changes

• TF-DF no longer supports Python 3.8 since Tensorflow dropped its support.

Features

• Compatibility with Tensorflow 2.14.0
• Contrib: Training preprocessing jointly on the input features, labels and
  weights

Fix

• Incorrect model predictions for models without features
• Data race for model resources

1.5.0 - 2023-07-06

Features

• Compatibility with Tensorflow 2.13.0

Features

• Support for multi-task learning.
• New tutorial for TF-DF <--> TF.js
• Support for uplift modeling in the model inspector.
• New tutorial for Uplift modeling.
• Bump Bazel version to 6.1.0.

Fix

• Regex to generate Bazel workspace.
• Remove warning when converting Keras -> YDF.
• Fixed default hyperparameter issue Github #172.
• Various documentation issues fixed.

Features

• Check learner parameters during the model construction.
• Fix discretized numerical features for regression task.
• Allow for float32 values to be fed as categorical features.
• Add new / improved tutorials for ranking and visualization.
• Compatibility with Tensorflow 2.12.0. Unfortunately, this means dropping
  support for Python 3.7.

Fix

• Fix crashes when using ranking with very large groups.
• Add option to set the port used by YDF in TF-DF distributed training.
• Improve logging robustness.

Features

• Add support for distributed training and distributed hyper-parameter tuning
  in the OSS build. See
  https://www.tensorflow.org/decision_forests/distributed_training
• Setting "subsample" is enough enable random subsampling (to need to also set
  "sampling_method=RANDOM").
• Add "min_vocab_frequency" argument in "FeatureUsage" to control the minimum
  frequency of categorical items.
• Add "override_global_imputation_value" argument in "FeatureUsage" to
  override the value used for global imputation of missing value by the
  global-imputation algorithm.
• The Tuner argument "use_predefined_hps" automatically configures the set of
  hyper-parameters to explore during automatic hyper-parameter tuning.
• Replaces the MEAN_MIN_DEPTH variable importance with INV_MEAN_MIN_DEPTH.
• Add option to forbid model inference with the slow inference engine.

Fix

• Automatic documentation generation for RandomForestModel and other classes.

1.1.0 - 2022-11-18

Features

• Native support for TensorFlow Decision Forests in TensorFlow Serving.
• Add support for zipped Yggdrasil Decision Forests model for
  yggdrasil_model_to_keras_model.
• Added model prediction tutorial.
• Prevent premature stopping of GBT training through new parameter
  early_stopping_initial_iteration.

Fix

• Using loaded datasets with TF-DF no longer fails (Github #131).
• Automatically infer the semantic of int8 values as numerical (was
  categorical before).
• Build script fixed
• Model saving no longer fails when using invalid feature names.
• Added keyword to pandas dataset drop (Github #135).

Features

• Support for Tensorflow Serving APIs.
• Add support for zipped Yggdrasil Decision Forests model for yggdrasil_model_to_keras_model.
• Added model prediction tutorial.
• Prevent premature stopping of GBT training through new parameter early_stopping_initial_iteration.

Fix

• Using loaded datasets with TF-DF no longer fails (Github #131).
• Automatically infer the semantic of int8 values as numerical (was categorical before).
• Build script fixed
• Model saving no longer fails when using invalid feature names.
• Added keyword to pandas dataset drop (Github #135).

Nightly build of TensorFlow Serving 2.11.
TensorFlow Serving >=2.11 supports natively TensorFlow Decision Forests models.

Build instructions:

git clone https://github.com/tensorflow/serving.git
docker run -it -v ${PWD}/..:/working_dir -w /working_dir/serving tensorflow/serving:nightly-devel bash
bazel build //tensorflow_serving/model_servers:tensorflow_model_server

Experimental TF-DF Release for MacOS

This pre-release is designed to help testing a release for TF-DF 1.0.1 with different MacOS versions.

Make sure you pick a version corresponding to your MacOS version and Python version.

TensorFlow Decision Forests 1.0.1

With this release, TensorFlow Decision Forests finally reaches its first major release 🥳

With this milestone we want to communicate more broadly that TensorFlow Decision Forests has become a more stable and mature library. In particular, we established more comprehensive testing to make sure that TF-DF is ready for professional environments.

Features

• Add customization of the number of IO threads when using fit_on_dataset_path.

Fix

• Improved documentation
• Improved testing and stability
• Issue in the application of auditwheel

Fix

• Improve documentation

Features

• Multithreading of the oblique splitter for gradient boosted tree models.
• Support for pure serving model i.e. model containing only serving data.
• Add "edit_model" cli tool.

Fix

• Remove bias toward low outcome in uplift modeling.

TF-DF with TF-Serving binary for Tensorflow 2.9.1