Skip to content

Basic FFNN + Trainer

Compare
Choose a tag to compare
@Ithanil Ithanil released this 12 Jul 20:09
· 114 commits to master since this release

v0.1

  • clean and stable basic FFNN with well-separated component class structure
  • includes most common activation functions
  • provides derivatives d/dx, d²/dx², d/dv, d²/dv/dx, d³/dv/dx²
  • is reasonably performance optimized
  • handles least-squares-fitting (training) against target data with minimal hassle for the user
  • allows data split in training/validation/testing and early stopping technique