v1.0.7

• Fixed integration with MSVC.
• Changed cosine score function in CPU and CUDA.

v1.0.6

Improved PyTorch integration

The step and predict/__call__ functions, GBRL functions in GBRL now work directly with PyTorch tensors, eliminating the need to clone data to the CPU and convert it to NumPy.

v1.0.2

• Removed the limitation of using only two optimizers.
• Changed naming of GradientBoostingTrees class to GBRL.

v1.0.1

Added SHAP values

v1.0.0

Initial codebase release