Aging-Aware Training for Printed Neuromorphic Circuits

This github repository is for the paper at ICCAD'22 - Aging-Aware Training for Printed Neuromorphic Circuits

cite as

Aging-Aware Training for Printed Neuromorphic Circuits
Zhao, H.; Hefenbrock, M.; Beigl, M.; Tahoori, M.
2022 International Conference on Computer-Aided Design (ICCAD), October, 2022 IEEE/ACM.

Usage of the code:

1. Training of printed neural networks

$ sh experiment_ICCAD_2022.sh

Alternatively, the experiments can be conducted by running command lines in experiment_ICCAD_2022.sh separately, e.g.,

$ python3 experiment.py --DATASET 0  --SEED 0  --MODE nominal --projectname ICCAD_2022
$ python3 experiment.py --DATASET 0  --SEED 1  --MODE nominal --projectname ICCAD_2022
...

2. After training printed neural networks, the trained networks are in ./ICCAD_2022/model/, the log files for training can be found in ./ICCAD_2022/log/. If there is still files in ./ICCAD_2022/temp/, you should run the corresponding command line to train the networks further. Note that, each training is limited to 48 hours, you can change this time limitation in configuration.py

3. Evaluation can be done by running the evaluation_ICCAD_2022.sh in ./ICCAD_2022/ folder with

$ sh evaluation_ICCAD_2022.sh

Of course, each line in this file can be run separately as in step 1.

4. For visualization, run

$ python3 visualization.py