- Model can be composed with the functional Keras API before being trained.
- Makes all the Yggdrasil structural variable importances available.
- Makes getting the variable importance instantaneous.
- Surface the
name
argument in the model classes constructors. - Add a
postprocessing
model constructor argument to easy apply post-processing on the model predictions without relying on the Keras Functional API. - Add
extract_all_trees
method in the model inspector to efficiently exact all the trees. - Add
num_threads
constructor argument to control the number of training threads without using the advanced configuration. - By default, remove the temporary directory used to train the model when the model python object is garbage collected.
- Add the
import_dataspec
constructor argument to the model builder to import the feature definition and dictionaries (instead of relying on automatic discovery).
- When saving a model in a directory already containing a model, only the
assets
directory is entirely removed before the export (instead of the entire model directory).
- Wrong label shape in the model inspector's objective field for pre-integerized labels.
- Add more of characters to the non-recommended list of feature name characters.
- Make the inference op multi-thread compatible.
- Print an explicit error and some instructions when training a model with a Pandas dataframe.
pd_dataframe_to_tf_dataset
can automatically rename feature to make them compatible with SavedModel export signatures.model.save(...)
can override an existing model.- The link function of GBT model can be removed. For example, a binary classification GBT model trained with apply_link_function=False will output logits.
- Add hyper-parameter
sorting_strategy
to disable the computation of the pre-sorted index (slower to train, but consumes less memory). - Format wrapper code for colab help display.
- Raises an error when a feature name is not compatible (e.g. contains a space).
- Raise an error of the number of classes is greater than 100 (can be disabled).
- Raise an error if the model's task does not match the
pd_dataframe_to_tf_dataset
's task.
- Fix failure when input feature contains commas.
- Stop the training when interrupting a colab cell / typing ctrl-c.
model.fit
support training callbacks and a validation dataset.
- Fix failure when there are not input features.
- Register new inference engines.
- Inference engines: QuickScorer Extended and Pred
- Migration to TensorFlow 2.5.0.
- By default, use a pre-compiled version of the OP wrappers.
- Add missing
plotter.js
from Pip package. - Use GitHub version of Yggdrasil Decision Forests by default.
Initial Release of TensorFlow Decision Forests.
- Random Forest learner.
- Gradient Boosted Tree learner.
- CART learner.
- Model inspector: Inspect the internal model structure.
- Model plotter: Plot decision trees.
- Model builder: Create model "by hand".