Efficiently Learning Relative Similarity Embeddings with Crowdsourcing

Efficiently Learning Relative Similarity Embeddings with Crowdsourcing - Published in JOSS (2023)

https://github.com/stsievert/salmon

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 4 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

active-learning crowdsourcing embedding machine-learning triplet-loss triplets

Scientific Fields

Computer Science Computer Science - 84% confidence

Mathematics Computer Science - 84% confidence

Last synced: 4 months ago · JSON representation ·

Repository

A tool to collect triplet queries

Basic Info

Host: GitHub
Owner: stsievert
License: bsd-3-clause
Language: Python
Default Branch: master
Homepage: https://docs.stsievert.com/salmon/
Size: 119 MB

Statistics

Stars: 8
Watchers: 1
Forks: 2
Open Issues: 0
Releases: 78

Topics

active-learning crowdsourcing embedding machine-learning triplet-loss triplets

Created about 6 years ago · Last pushed over 1 year ago

Metadata Files

Readme Contributing License Citation

README.md

Salmon

Salmon is a tool for efficiently generating ordinal embeddings. It relies on "active" machine learning algorithms to choose the most informative queries for humans to answer.

Documentation

This documentation is available at these locations:

Primary source: https://docs.stsievert.com/salmon/
Secondary source: as a raw PDF (and as a slower loading PDF).
Secondary source: as zipped HTML directory, which requires unzipping the directory then opening up index.html.

Please file an issue if you can not access the documentation.

Running Salmon offline

Visit the documentation at https://docs.stsievert.com/salmon/offline.html. Briefly, this should work:

shell $ cd path/to/salmon $ conda env create -f salmon.lock.yml $ conda activate salmon (salmon) $ pip install -e .

The documentation online mentions more about how to generate an embedding offline: https://docs.stsievert.com/salmon/offline.html#generate-embeddings

With this, it's also possible to create a script that uses and imports Salmon:

``` python from salmon.triplets.samplers import TSTE import numpy as np

n, d = 85, 2 sampler = TSTE(n=n, d=d)

eminit = np.array([[i, -i] for i in range(n)]) sampler.opt.initialize(embedding=eminit)

queries, scores, meta = sampler.getqueries(num=10000) ```

This script allows the data scientist to score queries for an embedding they specify.

Owner

Name: Scott Sievert
Login: stsievert
Kind: user

Website: https://stsievert.com
Repositories: 45
Profile: https://github.com/stsievert

JOSS Publication

Efficiently Learning Relative Similarity Embeddings with Crowdsourcing

Published

April 17, 2023

DOI

10.21105/joss.04517

Volume 8, Issue 84, Page 4517

Authors

Scott Sievert

University of Wisconsin--Madison

Robert Nowak
University of Wisconsin--Madison

Timothy Rogers

University of Wisconsin--Madison

Editor

Andrew Stewart

Citation (CITATION.cff)

cff-version: "1.2.0"
authors:
- family-names: Sievert
  given-names: Scott
  orcid: "https://orcid.org/0000-0002-4275-3452"
- family-names: Nowak
  given-names: Robert
- family-names: Rogers
  given-names: Timothy
  orcid: "https://orcid.org/0000-0001-6304-755X"
doi: 10.5281/zenodo.7832431
message: If you use this software, please cite our article in the
  Journal of Open Source Software.
preferred-citation:
  authors:
  - family-names: Sievert
    given-names: Scott
    orcid: "https://orcid.org/0000-0002-4275-3452"
  - family-names: Nowak
    given-names: Robert
  - family-names: Rogers
    given-names: Timothy
    orcid: "https://orcid.org/0000-0001-6304-755X"
  date-published: 2023-04-17
  doi: 10.21105/joss.04517
  issn: 2475-9066
  issue: 84
  journal: Journal of Open Source Software
  publisher:
    name: Open Journals
  start: 4517
  title: Efficiently Learning Relative Similarity Embeddings with
    Crowdsourcing
  type: article
  url: "https://joss.theoj.org/papers/10.21105/joss.04517"
  volume: 8
title: Efficiently Learning Relative Similarity Embeddings with
  Crowdsourcing

GitHub Events

Total

Watch event: 1

Last Year

Watch event: 1

Committers

Last synced: 5 months ago

All Time

Total Commits: 163
Total Committers: 3
Avg Commits per committer: 54.333
Development Distribution Score (DDS): 0.012

Past Year

Commits: 0
Committers: 0
Avg Commits per committer: 0.0
Development Distribution Score (DDS): 0.0

Top Committers

Name	Email	Commits
Scott Sievert	s****t	161
Jason Sicotte	s**n@g**m	1
Christopher Cox	c**c@g**m	1

Issues and Pull Requests

Last synced: 4 months ago

All Time

Total issues: 31
Total pull requests: 69
Average time to close issues: about 2 months
Average time to close pull requests: 5 days
Total issue authors: 1
Total pull request authors: 1
Average comments per issue: 1.29
Average comments per pull request: 0.46
Merged pull requests: 68
Bot issues: 0
Bot pull requests: 0

Past Year

Issues: 0
Pull requests: 0
Average time to close issues: N/A
Average time to close pull requests: N/A
Issue authors: 0
Pull request authors: 0
Average comments per issue: 0
Average comments per pull request: 0
Merged pull requests: 0
Bot issues: 0
Bot pull requests: 0

View more stats

Top Authors

Issue Authors

stsievert (31)

Pull Request Authors

stsievert (70)

Top Labels

Issue Labels

enhancement (2) bug (1)

Pull Request Labels

Dependencies

.github/workflows/docs.yml actions

actions/checkout v2 composite
actions/setup-python v2 composite
peaceiris/actions-gh-pages v3 composite

.github/workflows/python-publish.yml actions

actions/checkout v3 composite
actions/setup-python v3 composite
pypa/gh-action-pypi-publish 27b31702a0e7fc50959f5ad993c78deac1bdfc29 composite

.github/workflows/test.yml actions

actions/checkout v2 composite
actions/setup-python v2 composite
docker/setup-buildx-action v1 composite

.github/workflows/test_offline.yml actions

actions/checkout v2 composite
actions/setup-python v2 composite

.github/workflows/test_pip.yml actions

actions/checkout v2 composite
actions/setup-python v2 composite
docker/setup-buildx-action v1 composite

Dockerfile docker

continuumio/miniconda3 4.10.3 build

docker/redismonitor/Dockerfile docker

redislabs/rejson latest build

docker-compose.yml docker

prom/prometheus latest
redislabs/rejson latest

requirements.txt pypi

Cython *
aiofiles *
altair *
autodoc_pydantic *
blosc *
bokeh ==2.0.1
cloudpickle *
cytoolz *
dask >=2021.02.0
dask-ml *
distributed >=2021.02.0
fastapi *
fastparquet *
gunicorn *
httpx *
ipykernel *
jinja2 <3.1.0
jupyter-server-proxy *
lz4 *
matplotlib *
numpy >=1.18.0
numpydoc *
pandas >=1.0.1
pyarrow *
pytest *
python-multipart *
pyyaml *
rejson *
scikit-learn *
scipy *
seaborn *
skorch >=0.8.0
sphinx >=4.0.0
sphinx_rtd_theme *
starlette-prometheus *
torch *
ujson *

setup.py pypi

for *

docs/requirements.txt pypi

numpydoc *

Efficiently Learning Relative Similarity Embeddings with Crowdsourcing

Science Score: 98.0%

Keywords

Scientific Fields

Repository

Basic Info

Statistics

Topics

Metadata Files

README.md

Salmon

Documentation

Running Salmon offline

Owner

JOSS Publication

Efficiently Learning Relative Similarity Embeddings with Crowdsourcing

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

Dependencies