autocluster

KO | EN

gitlite — search

#hacktoberfest #typescript #llm #rust #linux #docker #nodejs #windows #awesome-list #security #pytorch #ios #openai #open-source

(★ 61)

AutoML for clustering models in sklearn.

File Explorer

Download Latest Version (.zip)

- - blobs_dataset.ipynb
  - compare_convergence_rate.ipynb
  - digits_dataset.ipynb
- - __init__.py
  - log_helper.py
- - .gitignore
- - __init__.py
  - clusterutils.py
  - constants.py
  - logutils.py
  - metafeatures.py
  - plotting.py
  - stringutils.py
- .gitignore
- __init__.py
- algorithms.py
- autocluster.py
- build_config_space.py
- decode_logs.py
- evaluators.py
- preprocess_data.py
- random_sampling_optimizer.py
- warmstarter.py
- - .gitignore
  - a1.csv
  - a1.json
  - a2.csv
  - a2.json
  - a3.csv
  - a3.json
  - birch1.csv
  - birch1.json
  - birch2.csv
  - birch2.json
  - birch3.csv
  - birch3.json
  - dim032.csv
  - dim032.json
  - dim064.csv
  - dim064.json
  - dim10.csv
  - dim10.json
  - dim1024.csv
  - dim1024.json
  - dim11.csv
  - dim11.json
  - dim12.csv
  - dim12.json
  - dim128.csv
  - dim128.json
  - dim13.csv
  - dim13.json
  - dim14.csv
  - dim14.json
  - dim15.csv
  - dim15.json
  - dim2.csv
  - dim2.json
  - dim256.csv
  - dim256.json
  - dim3.csv
  - dim3.json
  - dim4.csv
  - dim4.json
  - dim5.csv
  - dim5.json
  - dim512.csv
  - dim512.json
  - dim6.csv
  - dim6.json
  - dim7.csv
  - dim7.json
  - dim8.csv
  - dim8.json
  - dim9.csv
  - dim9.json
  - g2-128-10.csv
  - g2-128-10.json
  - g2-128-100.csv
  - g2-128-100.json
  - g2-128-20.csv
  - g2-128-20.json
  - g2-128-30.csv
  - g2-128-30.json
  - g2-128-40.csv
  - g2-128-40.json
  - g2-128-50.csv
  - g2-128-50.json
  - g2-128-60.csv
  - g2-128-60.json
  - g2-128-70.csv
  - g2-128-70.json
  - g2-128-80.csv
  - g2-128-80.json
  - g2-128-90.csv
  - g2-128-90.json
  - g2-16-10.csv
  - g2-16-10.json
  - g2-16-100.csv
  - g2-16-100.json
  - g2-16-20.csv
  - g2-16-20.json
  - g2-16-30.csv
  - g2-16-30.json
  - g2-16-40.csv
  - g2-16-40.json
  - g2-16-50.csv
  - g2-16-50.json
  - g2-16-60.csv
  - g2-16-60.json
  - g2-16-70.csv
  - g2-16-70.json
  - g2-16-80.csv
  - g2-16-80.json
  - g2-16-90.csv
  - g2-16-90.json
  - g2-2-10.csv
  - g2-2-10.json
  - g2-2-100.csv
  - g2-2-100.json
  - g2-2-20.csv
  - g2-2-20.json
  - g2-2-30.csv
  - g2-2-30.json
  - g2-2-40.csv
  - g2-2-40.json
  - g2-2-50.csv
  - g2-2-50.json
  - g2-2-60.csv
  - g2-2-60.json
  - g2-2-70.csv
  - g2-2-70.json
  - g2-2-80.csv
  - g2-2-80.json
  - g2-2-90.csv
  - g2-2-90.json
  - g2-256-10.csv
  - g2-256-10.json
  - g2-256-100.csv
  - g2-256-100.json
  - g2-256-20.csv
  - g2-256-20.json
  - g2-256-30.csv
  - g2-256-30.json
  - g2-256-40.csv
  - g2-256-40.json
  - g2-256-50.csv
  - g2-256-50.json
  - g2-256-60.csv
  - g2-256-60.json
  - g2-256-70.csv
  - g2-256-70.json
  - g2-256-80.csv
  - g2-256-80.json
  - g2-256-90.csv
  - g2-256-90.json
  - g2-32-10.csv
  - g2-32-10.json
  - g2-32-100.csv
  - g2-32-100.json
  - g2-32-20.csv
  - g2-32-20.json
  - g2-32-30.csv
  - g2-32-30.json
  - g2-32-40.csv
  - g2-32-40.json
  - g2-32-50.csv
  - g2-32-50.json
  - g2-32-60.csv
  - g2-32-60.json
  - g2-32-70.csv
  - g2-32-70.json
  - g2-32-80.csv
  - g2-32-80.json
  - g2-32-90.csv
  - g2-32-90.json
  - g2-4-10.csv
  - g2-4-10.json
  - g2-4-100.csv
  - g2-4-100.json
  - g2-4-20.csv
  - g2-4-20.json
  - g2-4-30.csv
  - g2-4-30.json
  - g2-4-40.csv
  - g2-4-40.json
  - g2-4-50.csv
  - g2-4-50.json
  - g2-4-60.csv
  - g2-4-60.json
  - g2-4-70.csv
  - g2-4-70.json
  - g2-4-80.csv
  - g2-4-80.json
  - g2-4-90.csv
  - g2-4-90.json
  - g2-512-10.csv
  - g2-512-10.json
  - g2-512-100.csv
  - g2-512-100.json
  - g2-512-20.csv
  - g2-512-20.json
  - g2-512-30.csv
  - g2-512-30.json
  - g2-512-40.csv
  - g2-512-40.json
  - g2-512-50.csv
  - g2-512-50.json
  - g2-512-60.csv
  - g2-512-60.json
  - g2-512-70.csv
  - g2-512-70.json
  - g2-512-80.csv
  - g2-512-80.json
  - g2-512-90.csv
  - g2-512-90.json
  - g2-64-10.csv
  - g2-64-10.json
  - g2-64-100.csv
  - g2-64-100.json
  - g2-64-20.csv
  - g2-64-20.json
  - g2-64-30.csv
  - g2-64-30.json
  - g2-64-40.csv
  - g2-64-40.json
  - g2-64-50.csv
  - g2-64-50.json
  - g2-64-60.csv
  - g2-64-60.json
  - g2-64-70.csv
  - g2-64-70.json
  - g2-64-80.csv
  - g2-64-80.json
  - g2-64-90.csv
  - g2-64-90.json
  - g2-8-10.csv
  - g2-8-10.json
  - g2-8-100.csv
  - g2-8-100.json
  - g2-8-20.csv
  - g2-8-20.json
  - g2-8-30.csv
  - g2-8-30.json
  - g2-8-40.csv
  - g2-8-40.json
  - g2-8-50.csv
  - g2-8-50.json
  - g2-8-60.csv
  - g2-8-60.json
  - g2-8-70.csv
  - g2-8-70.json
  - g2-8-80.csv
  - g2-8-80.json
  - g2-8-90.csv
  - g2-8-90.json
  - s1.csv
  - s1.json
  - s2.csv
  - s2.json
  - s3.csv
  - s3.json
  - s4.csv
  - s4.json
  - unbalance.csv
  - unbalance.json
  - Untitled.ipynb
- - 0.csv
  - 1.csv
  - 2.csv
  - 3.csv
  - comments.txt
  - hand_gestures_classification_merged.csv
- - .gitignore
  - banana.json
  - bank-additional-full.json
  - breast_cancer.json
  - column_2C_weka.json
  - column_3C_weka.json
  - Development Index.json
  - fertility.json
  - fifa19.json
  - Financial_Distress.json
  - glass.json
  - hand_gestures_classification_merged.json
  - heart.json
  - income_evaluation.json
  - iris.json
  - loan_data_set.json
  - mushrooms.json
  - nasa.json
  - nba.games.stats.json
  - pd_speech_features.json
  - pulsar_stars.json
  - Seed_Data.json
  - Social_Networks_Ads.json
  - spanish_rail_ticket.json
  - telco_customer_churn.json
  - titanic.json
  - Toddler Autism dataset July 2018.json
  - weatherAUS.json
  - xray.json
- - .gitignore
- .gitignore
- avocado.csv
- banana.csv
- bank-additional-full
- bank-additional-full.csv
- breast_cancer.csv
- column_2C_weka.csv
- column_3C_weka.csv
- columns.csv
- Development_Index.csv
- fertility.csv
- fifa19.csv
- fifa19_comments.txt
- Financial_Distress.csv
- glass.csv
- hand_gestures_classification_merged.csv
- heart.csv
- income_evaluation.csv
- iris.csv
- loan_data_set.csv
- mushrooms.csv
- nasa.csv
- nba.games.stats.csv
- pd_speech_features.csv
- pulsar_stars.csv
- pulsar_stars_comments.txt
- responses.csv
- Seed_Data.csv
- Social_Network_Ads.csv
- spanish_rail_ticket.csv
- telco_customer_churn.csv
- titanic.csv
- Toddler_Autism_dataset_July_2018.csv
- Untitled.ipynb
- weatherAUS.csv
- winequality_red.csv
- xray.csv
- - .gitignore
  - benchmark_silhouette_metafeatures_table.csv
- - run_metaknowledge.sh
  - run_metaknowledge_benchmark_silhouette.sh
  - run_metalearning.sh
  - run_metalearning_benchmark_1.sh
  - run_metalearning_benchmark_1_5.sh
  - run_metalearning_benchmark_2.sh
  - run_test_autocluster_on_benchmark_1.sh
  - run_test_autocluster_on_benchmark_2.sh
  - run_test_autocluster_on_benchmark_3.sh
  - run_test_autocluster_on_benchmark_4.sh
- metaknowledge.py
- metalearning.py
- test_autocluster_on_benchmark.py
- test_evaluator_combinations.py
- .keep
- clustering_algorithms.png
- clustering_result_dim128.png
- clustering_result_s2.png
- dim_reduction_algorithms.png
- autocluster_ppt.pdf
- autocluster_report.pdf
.gitattributes
.gitignore
LICENSE
README.md
requirements.txt
setup.cfg
setup.py

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

## autocluster
``autocluster`` is an automated machine learning (AutoML) toolkit for performing clustering tasks.

Report and presentation slides can be found [here](https://github.com/wywongbd/autocluster/blob/master/reports/autocluster_report.pdf) and [here](https://github.com/wywongbd/autocluster/blob/master/reports/autocluster_ppt.pdf).

## Prerequisites
- Python 3.5 or above
- Linux OS, or [Windows WSL](https://docs.microsoft.com/en-us/windows/wsl/about) is also possible

## How to get started?
1. First, install [SMAC](https://automl.github.io/SMAC3/stable/installation.html):
  - ``sudo apt-get install build-essential swig``
  - ``conda install gxx_linux-64 gcc_linux-64 swig``
  - ``pip install smac==0.8.0``
2. ``pip install autocluster``

## How it works?
- ``autocluster`` automatically optimizes the *configuration* of a clustering problem. By *configuration*, we mean 
    - choice of dimension reduction algorithm
    - choice of clustering model
    - setting of dimension reduction algorithm's hyperparameters
    - setting of clustering model's hyperparameters  
  
- ``autocluster`` provides 3 different approaches to optimize the configuration (with increasing complexity): 
    - random optimization
    - bayesian optimization 
    - bayesian optimization + meta-learning (warmstarting)

## Algorithms/Models supported
- List of dimension reduction algorithms in ``sklearn`` supported by ``autocluster``'s optimizer.
<img src="images/dim_reduction_algorithms.png" width="600">

- List of clustering models in ``sklearn`` supported by ``autocluster``'s optimizer.
<img src="images/clustering_algorithms.png" width="600">

## Examples
Examples are available in these [notebooks](/autocluster/examples/).

## Experimental results
- This dataset comprises of 16 Gaussian clusters in 128-dimensional space with ``N = 1024`` points. The optimal configuration obtained by ``autocluster`` (SMAC + Warmstarting) consists of a Truncated SVD dimension reduction model + Birch clustering model.
<img src="images/clustering_result_dim128.png" width="600">

- This dataset comprises of 15 Gaussian clusters in 2-dimensional space with ``N = 5000 points``. The optimal configuration obtained by ``autocluster`` (SMAC + Warmstarting) consists of a TSNE dimension reduction model + Agglomerative clustering model.
<img src="images/clustering_result_s2.png" width="600">

## Links  
- [Link](https://pypi.org/project/autocluster/) to pypi. 
- Great [writeup](http://krasserm.github.io/2018/03/21/bayesian-optimization/) by Martin Krasser on Bayesian Optimization

## Disclaimer
The project is experimental and still under development.

autocluster

File Explorer

Today's picks

Was this content helpful?

Recent Feedback

autocluster

File Explorer

Today's picks

Was this content helpful?

Recent Feedback

Download README