BlackBIRDS

BlackBIRDS: Black-Box Inference foR Differentiable Simulators - Published in JOSS (2023)

https://github.com/arnauqb/blackbirds

Science Score: 98.0%

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 7 DOI reference(s) in README and JOSS metadata
✓
Academic publication links
Links to: joss.theoj.org
○
Committers with academic emails
○
Institutional organization owner
✓
JOSS paper metadata
Published in Journal of Open Source Software

Keywords

bayesian-inference differentiable-programming pytorch

Scientific Fields

Artificial Intelligence and Machine Learning Computer Science - 40% confidence

Last synced: 4 months ago · JSON representation ·

Repository

Black-Box Inference foR Differentiable Simulators

Basic Info

Host: GitHub
Owner: arnauqb
License: mit
Language: Python
Default Branch: main
Homepage: http://www.arnau.ai/blackbirds/
Size: 15 MB

Statistics

Stars: 20
Watchers: 4
Forks: 1
Open Issues: 6
Releases: 4

Topics

bayesian-inference differentiable-programming pytorch

Created over 2 years ago · Last pushed about 1 year ago

Metadata Files

Readme Contributing License Citation

BlackBIRDS is a Python package consisting of generically applicable, black-box inference methods for differentiable simulation models. It facilitates both (a) the differentiable implementation of simulation models by providing a common object-oriented framework for their implementation in PyTorch, and (b) the use of a variety of gradient-assisted inference procedures for these simulation models, allowing researchers to easily exploit the differentiable nature of their simulator in parameter estimation tasks. The package consists of both Bayesian and non-Bayesian inference methods, and relies on well-supported software libraries (e.g. normflows, Stimper et al. (2023)) to provide this broad functionality.

1. Installation

The easiest way to install Birds is to install it from the PyPI repository

pip install blackbirds

To get the latest development version, you can install it directly from git

pip install git+https://github.com/arnauqb/blackbirds

2. Usage

Refer to the docs for examples and specific API usage. Here is a basic example:

```python import torch

from blackbirds.models.randomwalk import RandomWalk from blackbirds.infer.vi import VI from blackbirds.posteriorestimators import TrainableGaussian from blackbirds.simulate import simulateandobserve_model

random walk model

rw = RandomWalk(n_timesteps=100)

generate synthetic data to fit to

truep = torch.logit(torch.tensor(0.25)) truedata = rw.observe(rw.run(torch.tensor([true_p])))

define loss to minimize

class L2Loss: def init(self, model): self.model = model self.lossfn = torch.nn.MSELoss() def _call_(self, params, data): observedoutputs = simulateandobservemodel(self.model, params) return self.lossfn(observed_outputs[0], data[0])

initialize generalized variational inference

posteriorestimator = TrainableGaussian([0.], 1.0) prior = torch.distributions.Normal(truep + 0.2, 1) optimizer = torch.optim.Adam(posterior_estimator.parameters(), 1e-2) loss = L2Loss(rw)

vi = VI(loss, posteriorestimator=posteriorestimator, prior=prior, optimizer=optimizer, w = 0) # no regularization

train the estimator

vi.run(truedata, 1000, maxepochswithoutimprovement=100)

```

3. Tests

To run the unit tests of the code, you need to have pytest installed,

bash pip install pytest pytest-cov

and run the command

bash pytest test

4. Contributing

See CONTRIBUTING.md for the contribution guidelines.

5. Citation

@article{Quera-Bofarull2023, doi = {10.21105/joss.05776}, url = {https://doi.org/10.21105/joss.05776}, year = {2023}, publisher = {The Open Journal}, volume = {8}, number = {89}, pages = {5776}, author = {Arnau Quera-Bofarull and Joel Dyer and Anisoara Calinescu and J. Doyne Farmer and Michael Wooldridge}, title = {BlackBIRDS: Black-Box Inference foR Differentiable Simulators}, journal = {Journal of Open Source Software} }

Owner

Name: Arnau Quera-Bofarull
Login: arnauqb
Kind: user
Location: Oxford

Website: arnau.ai
Repositories: 42
Profile: https://github.com/arnauqb

Astrophysicist now doing research on agent-based modeling at Oxford.

JOSS Publication

BlackBIRDS: Black-Box Inference foR Differentiable Simulators

Published

September 28, 2023

DOI

10.21105/joss.05776

Volume 8, Issue 89, Page 5776

Authors

Arnau Quera-Bofarull

Department of Computer Science, University of Oxford, UK, Institute for New Economic Thinking, University of Oxford, UK

Joel Dyer

Department of Computer Science, University of Oxford, UK, Institute for New Economic Thinking, University of Oxford, UK

Anisoara Calinescu

Department of Computer Science, University of Oxford, UK

J. Doyne Farmer

Institute for New Economic Thinking, University of Oxford, UK, Mathematical Institute, University of Oxford, UK, Santa Fe Institute, USA

Michael Wooldridge

Department of Computer Science, University of Oxford, UK

Editor

Rachel Kurchin

Citation (CITATION.cff)

cff-version: "1.2.0"
authors:
- family-names: Quera-Bofarull
  given-names: Arnau
  orcid: "https://orcid.org/0000-0001-5055-9863"
- family-names: Dyer
  given-names: Joel
  orcid: "https://orcid.org/0000-0002-8304-8450"
- family-names: Calinescu
  given-names: Anisoara
  orcid: "https://orcid.org/0000-0003-2082-734X"
- family-names: Farmer
  given-names: J. Doyne
  orcid: "https://orcid.org/0000-0001-7871-073X"
- family-names: Wooldridge
  given-names: Michael
  orcid: "https://orcid.org/0000-0002-9329-8410"
contact:
- family-names: Quera-Bofarull
  given-names: Arnau
  orcid: "https://orcid.org/0000-0001-5055-9863"
- family-names: Dyer
  given-names: Joel
  orcid: "https://orcid.org/0000-0002-8304-8450"
doi: 10.5281/zenodo.8377044
message: If you use this software, please cite our article in the
  Journal of Open Source Software.
preferred-citation:
  authors:
  - family-names: Quera-Bofarull
    given-names: Arnau
    orcid: "https://orcid.org/0000-0001-5055-9863"
  - family-names: Dyer
    given-names: Joel
    orcid: "https://orcid.org/0000-0002-8304-8450"
  - family-names: Calinescu
    given-names: Anisoara
    orcid: "https://orcid.org/0000-0003-2082-734X"
  - family-names: Farmer
    given-names: J. Doyne
    orcid: "https://orcid.org/0000-0001-7871-073X"
  - family-names: Wooldridge
    given-names: Michael
    orcid: "https://orcid.org/0000-0002-9329-8410"
  date-published: 2023-09-28
  doi: 10.21105/joss.05776
  issn: 2475-9066
  issue: 89
  journal: Journal of Open Source Software
  publisher:
    name: Open Journals
  start: 5776
  title: "BlackBIRDS: Black-Box Inference foR Differentiable Simulators"
  type: article
  url: "https://joss.theoj.org/papers/10.21105/joss.05776"
  volume: 8
title: "BlackBIRDS: Black-Box Inference foR Differentiable Simulators"

GitHub Events

Total

Issues event: 2
Watch event: 6
Issue comment event: 6
Push event: 3
Pull request event: 2
Fork event: 1
Create event: 1

Last Year

Issues event: 2
Watch event: 6
Issue comment event: 6
Push event: 3
Pull request event: 2
Fork event: 1
Create event: 1

Committers

Last synced: 5 months ago

All Time

Total Commits: 220
Total Committers: 3
Avg Commits per committer: 73.333
Development Distribution Score (DDS): 0.2

Past Year

Commits: 7
Committers: 2
Avg Commits per committer: 3.5
Development Distribution Score (DDS): 0.429

Top Committers

Name	Email	Commits
Arnau Quera-Bofarull	a**q@p**m	176
joelnmdyer	j**r@g**m	28
arnauq	a**a@g**m	16

Issues and Pull Requests

Last synced: 4 months ago

All Time

Total issues: 29
Total pull requests: 23
Average time to close issues: 13 days
Average time to close pull requests: about 13 hours
Total issue authors: 5
Total pull request authors: 3
Average comments per issue: 1.0
Average comments per pull request: 0.78
Merged pull requests: 22
Bot issues: 0
Bot pull requests: 0

Past Year

Issues: 1
Pull requests: 2
Average time to close issues: 1 day
Average time to close pull requests: about 2 hours
Issue authors: 1
Pull request authors: 1
Average comments per issue: 1.0
Average comments per pull request: 3.0
Merged pull requests: 2
Bot issues: 0
Bot pull requests: 0

View more stats

Top Authors

Issue Authors

joelnmdyer (14)
arnauqb (8)
rajeshrinet (2)
benjaminpatrickevans (1)

Pull Request Authors

arnauqb (12)
joelnmdyer (11)
ayushchopra96 (1)

Top Labels

Issue Labels

documentation (3) enhancement (2) invalid (1) bug (1)

Pull Request Labels

Packages

Total packages: 1
Total downloads:
- pypi 23 last-month

Total dependent packages: 0
Total dependent repositories: 0
Total versions: 4
Total maintainers: 1

pypi.org: blackbirds

Black-Box Inference foR Differentiable Simulators.

Homepage: https://github.com/arnauqb/blackbirds
Documentation: https://blackbirds.readthedocs.io/
License: MIT License
Latest release: 1.0.0
published over 2 years ago

Versions: 4
Dependent Packages: 0
Dependent Repositories: 0
Downloads: 23 Last month

Rankings

Dependent packages count: 7.2%

Average: 24.3%

Dependent repos count: 41.4%

Maintainers (1)

arnauq

Last synced: 4 months ago

BlackBIRDS

Science Score: 98.0%

Keywords

Scientific Fields

Repository

Basic Info

Statistics

Topics

Metadata Files

README.md

1. Installation

2. Usage

random walk model

generate synthetic data to fit to

define loss to minimize

initialize generalized variational inference

train the estimator

3. Tests

4. Contributing

5. Citation

Owner

JOSS Publication

BlackBIRDS: Black-Box Inference foR Differentiable Simulators

Authors

Editor

Tags

Citation (CITATION.cff)

GitHub Events

Total

Last Year

Committers

All Time

Past Year

Top Committers

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: blackbirds

Rankings

Maintainers (1)