V1.1.0

V1.1.0

Improved performance slightly by removing redundant calculation #9
Added accuracy calculator for cuda
Added train test split in Train function
Added debug and release path for cuda dll
Fixed predictions and accuracy calculations while training on cuda

V1.0.0

V1.0.0

• feed forward neural network: 1 input -> x hidden -> 1 output layer
• xavier initialisation
• cuda/cpu training
• simple and scalable
• easy to use
• easy to extend