Added ensemble HNN model weights

v.1.4

Added option to train ensemble of HNN models

Model fitted for linear interpolation

Model fitted with 94% accuracy on linearly interpolated data, with 500 observations per task.

Updated model weights, 95% accuracy

System specs:
OS: Ubuntu Cinnamon 22.04.3 LTS x86
CPU: AMD Ryzen 7 5700G with Radeon G
GPU: NVIDIA GeForce GTX 1080 (8GB vram) + NVIDIA GeForce RTX 2080 (6GB vram)
Memory: 31950MiB

epoch,step,accuracy,loss
500,0,0.9354838728904724,0.9684239625930786

Trained model weights, with validation and test

Fixed project structure

Updated gitignore

Initial model fit on Sara data

v.1.0

Updated README.md