- First-Neural-Network.py: Neural network built in Python without using any ML libraries like Keras or PyTorch. Based on Tariq Rashid's book "Make Your Own Neural Network".
- First-Keras-Network.py: Neural network built using Keras, following Jason Brownlee's tutorial.
- Keras-Initials.py: Neural network trained to identify handwritten letters A-Z.
- First-Pytorch-Network.py: Neural network built using PyTorch, implementing the same as
Keras-Initials.py. - Image_to_CSV.py: Utility for converting images to CSV format.
- Optuna.py: Hyperparameter optimization using Optuna.
Located in the SantaFeTrail-CNN/ directory:
- SantaFe_CNN.py: Main implementation of the Santa Fe Trail environment and CNN agent.
- CSNN-Expansion.py: Expansion of the convolutional SNN for the Santa Fe Trail.
- Plotter.py: Visualization and plotting utilities.
- episode_stats.json: Example output file containing episode statistics.
Located in the SantaFeTrail-Snntorch/src/ directory and root:
- Snntorch-Spike-Encoding.py: Tutorial 1 - Spike Encoding from the Snntorch docs.
- Snntorch-Leaky-Integrate-Fire-Neuron.py: Tutorial 2 - The Leaky Integrate-and-Fire Neuron.
- Feedforward-SNN.py: Tutorial 3 - Feedforward SNN.
- 2nd-Order-SNMs.py: Tutorial 4 - Second-order Spiking Neuron Models.
- Training-SNNs.py: Tutorial 5 - Training SNNs.
- Convolutional-SNN.py: Tutorial 6 - Convolutional SNNs.
- Tonic-Neuromorphic.py: Tutorial 7 - Tonic Neuromorphic Datasets.
- First-Neural-Networks/: Contains all first neural network scripts and utilities.
- SantaFeTrail-CNN/: Contains Santa Fe Trail CNN and related scripts.
- SantaFeTrail-Snntorch/: Contains SNN implementations using snntorch.
- SnnTorch-Tutorials/: Additional snntorch tutorials and experiments.
To do:
- Add individual files based off of
First-Neural-Network.pyfor XOR function, vertical/horizontal line classification - Add more frequent comments