Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/ajaysub110/tensor_genn
A library for converting models trained using TensorFlow into GeNN
https://github.com/ajaysub110/tensor_genn
Last synced: about 1 month ago
JSON representation
A library for converting models trained using TensorFlow into GeNN
- Host: GitHub
- URL: https://github.com/ajaysub110/tensor_genn
- Owner: ajaysub110
- License: gpl-2.0
- Created: 2019-06-05T05:53:59.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2019-11-16T16:43:46.000Z (about 5 years ago)
- Last Synced: 2024-12-05T17:21:31.843Z (about 1 month ago)
- Language: Python
- Homepage:
- Size: 33.3 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
- License: LICENSE
Awesome Lists containing this project
README
# TensorGeNN
A Python library for converting trained TensorFlow ANNs to GeNN spiking neural networks with low loss in performance using the algorithm proposed by *Diehl et al.* in "***Fast-Classifying, High-Accuracy Spiking Deep Networks Through Weight and Threshold Balancing***"
## Results
Below are the results of conversion experiments on MNIST and CIFAR-10 with and without data-based normalization.
### 1) MNIST
**ANN model**:\
Trained on 60000 examples for 5 epochs.\
Train accuracy: 99.21% on 60000 examples\
Test accuracy: 98.86% on 10000 examples
**GeNN model**:\
Conversion parameters:\
single_example_time = 350\
dense_membrane_capacitance = 1.0\
sparse_membrane_capacitance = 0.2\
neuron_threshold_voltage = -57.0\
neuron_resting_voltage = -60.0
**On three trials**: (all accuracies on test set)\
Note: ANN test set contains 10000 example while the GeNN accuracies have been computed on 500 examples.
| ANN accuracy | GeNN (without normalization) | GeNN (with normalization) |
| ------------- | ---------------------------- | ------------------------- |
| 98.91% | 97.8% | 98.0% |
| 98.88% | 96.8% | 97.2% |
| 98.88% | 96.8% | 96.6% |
### 2) CIFAR-10
**ANN model**:\
Trained on 50000 examples for 10 epochs.\
Train accuracy: 90.99% on 50000 examples\
Test accuracy: 77.29% on 10000 examples (overfitting since I haven't added dropout layers yet)
**GeNN model**:\
Conversion parameters:\
single_example_time = 3750 (though it takes only ~2500 with no weight normalization)\
dense_membrane_capacitance = 0.1\
sparse_membrane_capacitance = 0.01\
neuron_threshold_voltage = -58.0\
neuron_resting_voltage = -60.0
**On one trials**: (all accuracies on test set)\
Note: ANN test set contains 10000 example while the GeNN accuracies have been computed on 500 examples.
| ANN accuracy | GeNN (without normalization) | GeNN (with normalization) |
| ------------- | ---------------------------- | ------------------------- |
| 77.29% | 75.0% | 77.0% |
## Requirements:
- Python 3
- TensorFlow
- PyGeNN (see [here](https://github.com/genn-team/genn/blob/master/pygenn/README.md) for installation instructions)
- Matplotlib
## Installation:
Note: Since work is still in progress, we don't have a standard installation procedure. However, in the meantime, you can follow the steps below.
1. Clone the repository: `git clone https://github.com/genn-team/tensor_genn.git`
2. Change directory: `cd ~/tensor_genn`
3. Create a new file and follow a procedure shown in `conversion_example.py` to convert your TensorFlow model into GeNN.
## References:
- Peter U. Diehl, Daniel Neil, Jonathan Binas, Matthew Cook, Shih-Chii Liu, and Michael Pfeiffer. 2015. Fast-Classifying, High-Accuracy Spiking Deep Networks Through Weight and Threshold Balancing. IJCNN (2015)