TF Boosted Trees (TFBT)
TF Boosted trees is an implementation of a gradient boosting algorithm with trees used as weak learners.
Folder "examples" demonstrates how TFBT estimators can be used for various problems. Namely, it contains:
- binary_mnist.py - an example on how to use TFBT for binary classification.
- mnist.py - a multiclass example.
- boston.py - a regression example.