A highly-configurable tool that enables thorough evaluation of deep metric learning algorithms.
Project description
Powerful Benchmarker
Documentation
A Metric Learning Reality Check
This library was used for A Metric Learning Reality Check. See the documentation for supplementary material.
Benchmark results:
Benefits of this library
- Highly configurable:
- Yaml files for organized configuration
- A powerful command line syntax that allows you to merge, override, swap, apply, and delete config options.
- Customizable:
- Benchmark your own losses, miners, datasets etc. with a simple function call.
- Easy hyperparameter optimization:
- Append the ~BAYESIAN~ flag to the names of hyperparameters you want to optimize.
- Extensive logging:
- View experiment data in tensorboard, CSV and SQLite format.
- Reproducible:
- Config files are saved with each experiment and are easily reproduced.
- Trackable changes:
- Keep track of changes to an experiment's configuration.
Installation
pip install powerful-benchmarker
Citing the benchmark results
If you'd like to cite the benchmark results, please cite this paper:
@misc{musgrave2020metric,
title={A Metric Learning Reality Check},
author={Kevin Musgrave and Serge Belongie and Ser-Nam Lim},
year={2020},
eprint={2003.08505},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
Citing the code
If you'd like to cite the powerful-benchmarker code, you can use this bibtex:
@misc{Musgrave2019,
author = {Musgrave, Kevin and Lim, Ser-Nam and Belongie, Serge},
title = {Powerful Benchmarker},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/KevinMusgrave/powerful-benchmarker}},
}
Acknowledgements
Thank you to Ser-Nam Lim at Facebook AI, and my research advisor, Professor Serge Belongie. This project began during my internship at Facebook AI where I received valuable feedback from Ser-Nam, and his team of computer vision and machine learning engineers and research scientists.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for powerful-benchmarker-0.9.32.dev1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | b80b75dcdfc32ec31640cea6549777431b4f685586ec6633a3b552b1fad93a20 |
|
MD5 | 5e4999ef3459bcd26c18327c727db652 |
|
BLAKE2b-256 | 6b2a3f739edf6db25fcc03aae065241b86c17f061f75ac378d1dcb7a062ffcf4 |
Hashes for powerful_benchmarker-0.9.32.dev1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | ea37ea1340c63b01085228616cc9413e7c38a93eadbe7a47c032b1bbc3f6bc45 |
|
MD5 | b93b9f848c0ace3059b340c499ed3149 |
|
BLAKE2b-256 | 372c01f05f536956d967795cdf86246f87fdac107ec1bfe775ac27fc75dc42ee |