analogvnn

A fully modular framework for modeling and optimizing analog neural networks

https://github.com/vivswan/analogvnn

Science Score: 77.0%

This score indicates how likely this project is to be science-related based on various indicators:

✓
CITATION.cff file
Found CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
✓
DOI references
Found 8 DOI reference(s) in README
✓
Academic publication links
Links to: arxiv.org
✓
Committers with academic emails
1 of 5 committers (20.0%) from academic institutions
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (11.7%) to scientific vocabulary

Keywords

analog deep-learning framework neural-network photonics pytorch

Last synced: 4 months ago · JSON representation ·

Repository

A fully modular framework for modeling and optimizing analog neural networks

Basic Info

Host: GitHub
Owner: Vivswan
License: other
Language: Python
Default Branch: master
Homepage: https://analogvnn.readthedocs.io
Size: 3.24 MB

Statistics

Stars: 20
Watchers: 2
Forks: 5
Open Issues: 0
Releases: 9

Topics

analog deep-learning framework neural-network photonics pytorch

Created about 3 years ago · Last pushed over 1 year ago

Metadata Files

Readme Changelog License Citation Codeowners

AnalogVNN

Documentation: https://analogvnn.readthedocs.io/

Installation:

Install PyTorch
Install AnalogVNN using pip

```bash # Current stable release for CPU and GPU pip install analogvnn

# For additional optional features pip install analogvnn[full] ```

Usage:

Sample code with AnalogVNN: sample_code.py
Sample code without AnalogVNN: samplecodenon_analog.py
Sample code with AnalogVNN and Logs: samplecodewith_logs.py
Jupyter Notebook: AnalogVNN_Demo.ipynb

Abstract

3 Layered Linear Photonic Analog Neural Network

AnalogVNN is a simulation framework built on PyTorch which can simulate the effects of optoelectronic noise, limited precision, and signal normalization present in photonic neural network accelerators. We use this framework to train and optimize linear and convolutional neural networks with up to 9 layers and ~1.7 million parameters, while gaining insights into how normalization, activation function, reduced precision, and noise influence accuracy in analog photonic neural networks. By following the same layer structure design present in PyTorch, the AnalogVNN framework allows users to convert most digital neural network models to their analog counterparts with just a few lines of code, taking full advantage of the open-source optimization, deep learning, and GPU acceleration libraries available through PyTorch.

AnalogVNN Paper: https://doi.org/10.1063/5.0134156

Citing AnalogVNN

We would appreciate if you cite the following paper in your publications for which you used AnalogVNN:

bibtex @article{shah2023analogvnn, title={AnalogVNN: A fully modular framework for modeling and optimizing photonic neural networks}, author={Shah, Vivswan and Youngblood, Nathan}, journal={APL Machine Learning}, volume={1}, number={2}, year={2023}, publisher={AIP Publishing} }

Or in textual form:

text Vivswan Shah, and Nathan Youngblood. "AnalogVNN: A fully modular framework for modeling and optimizing photonic neural networks." APL Machine Learning 1.2 (2023). DOI: 10.1063/5.0134156

Owner

Name: Vivswan Shah
Login: Vivswan
Kind: user
Company: University of Pittsburgh

Website: vivswan.github.io
Repositories: 5
Profile: https://github.com/Vivswan

PhD Student @ Upitt in Machine Learning and Quantum Computing

Citation (CITATION.cff)

cff-version: 1.2.0
title: 'AnalogVNN: A fully modular framework for modeling and optimizing photonic neural networks'
message: 'If you use this software, please cite it as below.'
preferred-citation:
  type: article
  authors:
    - given-names: Vivswan
      family-names: Shah
      email: vivswanshah@pitt.edu
      affiliation: University of Pittsburgh
    - family-names: Youngblood
      given-names: Nathan
      affiliation: University of Pittsburgh
  doi: "10.1063/5.0134156"
  journal: "APL Machine Learning"
  title: 'AnalogVNN: A fully modular framework for modeling and optimizing photonic neural networks'
  year: 2023
authors:
  - given-names: Vivswan
    family-names: Shah
    email: vivswanshah@pitt.edu
    affiliation: University of Pittsburgh
  - family-names: Youngblood
    given-names: Nathan
    affiliation: University of Pittsburgh
identifiers:
  - type: doi
    value: 10.1063/5.0134156
    description: >-
      The concept DOI for the collection containing
      all versions of the Citation File Format.
repository-code: 'https://github.com/Vivswan/AnalogVNN'
url: 'https://analogvnn.readthedocs.io/'
abstract: >-
  AnalogVNN, a simulation framework built on PyTorch
  which can simulate the effects of optoelectronic
  noise, limited precision, and signal normalization
  present in photonic neural network accelerators. We
  use this framework to train and optimize linear and
  convolutional neural networks with up to 9 layers
  and ~1.7 million parameters, while gaining insights
  into how normalization, activation function,
  reduced precision, and noise influence accuracy in
  analog photonic neural networks. By following the
  same layer structure design present in PyTorch, the
  AnalogVNN framework allows users to convert most
  digital neural network models to their analog
  counterparts with just a few lines of code, taking
  full advantage of the open-source optimization,
  deep learning, and GPU acceleration libraries
  available through PyTorch.
keywords:
  - photonics
  - neural-networks
  - analog-computing
  - deep-learning
license: MPL-2.0

GitHub Events

Total

Watch event: 5
Fork event: 2

Last Year

Watch event: 5
Fork event: 2

Committers

Last synced: almost 3 years ago

All Time

Total Commits: 283
Total Committers: 5
Avg Commits per committer: 56.6
Development Distribution Score (DDS): 0.3

Top Committers

Name	Email	Commits
Vivswan Shah	5**n@u**m	198
Vivswan Shah	s**n@g**m	71
Vivswan Shah	5**n@u**m	10
Vivswan Shah	v**h@p**u	3
Tianyi Zheng	t**2@g**m	1

Committer Domains (Top 20 + Academic)

pitt.edu: 1

Issues and Pull Requests

Last synced: 5 months ago

All Time

Total issues: 3
Total pull requests: 83
Average time to close issues: 4 days
Average time to close pull requests: 3 days
Total issue authors: 2
Total pull request authors: 2
Average comments per issue: 2.33
Average comments per pull request: 0.05
Merged pull requests: 74
Bot issues: 0
Bot pull requests: 0

Past Year

Issues: 1
Pull requests: 21
Average time to close issues: 12 days
Average time to close pull requests: 6 days
Issue authors: 1
Pull request authors: 1
Average comments per issue: 5.0
Average comments per pull request: 0.14
Merged pull requests: 13
Bot issues: 0
Bot pull requests: 0

View more stats

Top Authors

Issue Authors

PierrickPochelu (2)
hatsuka20 (1)

Pull Request Authors

Vivswan (111)
tianyizheng02 (1)

Top Labels

Issue Labels

question (2)

Pull Request Labels

Packages

Total packages: 1
Total downloads:
- pypi 113 last-month

Total dependent packages: 0
Total dependent repositories: 0
Total versions: 16
Total maintainers: 1

pypi.org: analogvnn

A fully modular framework for modeling and optimizing analog/photonic neural networks

Homepage: https://github.com/Vivswan/AnalogVNN
Documentation: https://analogvnn.readthedocs.io/en/latest/
License: Mozilla Public License 2.0 (MPL 2.0)
Latest release: 1.0.8
published over 1 year ago

Versions: 16
Dependent Packages: 0
Dependent Repositories: 0
Downloads: 113 Last month

Rankings

Dependent packages count: 6.6%

Downloads: 16.3%

Forks count: 17.3%

Average: 18.5%

Stargazers count: 21.8%

Dependent repos count: 30.6%

Maintainers (1)

vivswan

Last synced: 5 months ago

Dependencies

requirements.txt pypi

build *
furo *
graphviz *
importlib_metadata *
johnnydep *
matplotlib *
myst_parser *
natsort *
networkx *
numpy >=1.16.5
pillow *
rst-to-myst *
scipy *
seaborn *
setuptools ==65.6.3
sphinx >=4.2.0
sphinx-autoapi *
sphinx-autobuild *
sphinx-copybutton *
sphinx-inline-tabs *
sphinx-notfound-page *
sphinx-rtd-theme *
sphinxcontrib-katex *
sphinxext-opengraph *
tabulate *
tensorboard *
tensorflow >=2.0.0
torch *
torchaudio *
torchvision *
torchviz *

pyproject.toml pypi

dataclasses *
importlib-metadata <5.0.0,>=2.0.0; python_version < '3.8'
networkx *
numpy >=1.16.5
scipy *

requirements/requirements-dev.txt pypi

build * development
johnnydep * development
setuptools >=61.0.0 development

requirements/requirements-docs.txt pypi

furo *
myst_parser *
rst-to-myst *
sphinx >=4.2.0
sphinx-autoapi *
sphinx-autobuild *
sphinx-copybutton *
sphinx-inline-tabs *
sphinx-notfound-page *
sphinx-rtd-theme *
sphinxcontrib-katex *
sphinxext-opengraph *

requirements/requirements-test.txt pypi

flake8 * test
flake8-bugbear * test
flake8-coding * test
flake8-comprehensions * test
flake8-deprecated * test
flake8-docstrings * test
flake8-executable * test
flake8-quotes * test
flake8-return * test

requirements-all.txt pypi

analogvnn

Science Score: 77.0%

Keywords

Repository

Basic Info

Statistics

Topics

Metadata Files

README.md

AnalogVNN

Installation:

Usage:

Abstract

Citing AnalogVNN

Owner

Citation (CITATION.cff)

GitHub Events

Total

Last Year

Committers

All Time

Top Committers

Committer Domains (Top 20 + Academic)

Issues and Pull Requests

All Time

Past Year

Top Authors

Issue Authors

Pull Request Authors

Top Labels

Issue Labels

Pull Request Labels

Packages

pypi.org: analogvnn

Rankings

Maintainers (1)

Dependencies