This package is a Python wrapper for the DPMMSubClusters.jl Julia package.
Working on a subset of 100K images from ImageNet, containing 79 classes, we have created embeddings using SWAV, and reduced the dimension to 128 using PCA. We have compared our method with the popular scikit-learn GMM and DPGMM with the following results:
| Method | Timing (sec) | NMI (higher is better) |
|---|---|---|
| Scikit-learn's GMM (using EM, and given the True K) | 2523 | 0.695 |
| Scikit-learn's DPGMM | 6108 | 0.683 |
| DPMMpython | 475 | 0.705 |
pip install dpmmpython
If you already have Julia installed, install PyJulia and add the package DPMMSubClusters to your julia installation.
Make sure Julia path is configured correctly, e.g. you should be able to run julia by typing `julia` from the terminal, unless configured properly, PyJulia wont work.
Installation Shortcut for Ubuntu distributions
If you do not have Julia installed, or wish to create a clean installation for the purpose of using this package. after installing (with pip), do the following:
import dpmmpython
dpmmpython.install()
Optional arguments are install(julia_download_path = 'https://julialang-s3.julialang.org/bin/linux/x64/1.4/julia-1.4.0-linux-x86_64.tar.gz', julia_target_path = None), where the former specify the julia download file, and the latter the installation path, if the installation path is not specified, $HOME$/julia will be used.
As the install() command edit your .bashrc path, before using the pacakge, the terminal should either be reset, or modify the current environment according to the julia path you specified ($HOME$/julia/julia-1.4.0/bin by default).
from dpmmpython.dpmmwrapper import DPMMPython
from dpmmpython.priors import niw
import numpy as np
data,gt = DPMMPython.generate_gaussian_data(10000, 2, 10, 100.0)
prior = niw(1,np.zeros(2),4,np.eye(2))
labels,_,results= DPMMPython.fit(data,100,prior = prior,verbose = True, gt = gt)
Iteration: 1 || Clusters count: 1 || Log posterior: -71190.14226686998 || Vi score: 1.990707323192506 || NMI score: 6.69243345834295e-16 || Iter Time:0.004499912261962891 || Total time:0.004499912261962891
Iteration: 2 || Clusters count: 1 || Log posterior: -71190.14226686998 || Vi score: 1.990707323192506 || NMI score: 6.69243345834295e-16 || Iter Time:0.0038819313049316406 || Total time:0.008381843566894531
...
Iteration: 98 || Clusters count: 9 || Log posterior: -40607.39498126549 || Vi score: 0.11887067921133423 || NMI score: 0.9692247699387838 || Iter Time:0.015907764434814453 || Total time:0.5749104022979736
Iteration: 99 || Clusters count: 9 || Log posterior: -40607.39498126549 || Vi score: 0.11887067921133423 || NMI score: 0.9692247699387838 || Iter Time:0.01072382926940918 || Total time:0.5856342315673828
Iteration: 100 || Clusters count: 9 || Log posterior: -40607.39498126549 || Vi score: 0.11887067921133423 || NMI score: 0.9692247699387838 || Iter Time:0.010260820388793945 || Total time:0.5958950519561768
predictions = DPMMPython.predict(results[-1],data)
You can modify the number of processes by using DPMMPython.add_procs(procs_count), note that you can only scale it upwards.
Due to recent issue with the package used as interface between Julia and Python JuliaPy/pyjulia#425 , there might be problems working with Python >= 3.8.
For any questions: dinari@post.bgu.ac.il
Contributions, feature requests, suggestion etc.. are welcomed.
If you use this code for your work, please cite the following:
@inproceedings{dinari2019distributed,
title={Distributed MCMC Inference in Dirichlet Process Mixture Models Using Julia},
author={Dinari, Or and Yu, Angel and Freifeld, Oren and Fisher III, John W},
booktitle={2019 19th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)},
pages={518--525},
year={2019}
}