Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[pyclustering.cluster.clique] CLIQUE: Grid-Based Subspace Clustering #381

Closed
annoviko opened this issue Oct 27, 2017 · 4 comments
Closed
Assignees
Labels
Enhancement Tasks related to enhancement and development Good First Issue Tasks that can be easily done by contributors

Comments

@annoviko
Copy link
Owner

annoviko commented Oct 27, 2017

Introduction
CLIQUE is a density-based and grid-based subspace clustering algorithm. Python implementation of the algorithm is required in pyclustering.

Resources

  1. Lecture: https://www.coursera.org/learn/cluster-analysis/lecture/AAHTA/5-6-clique-grid-based-subspace-clustering
  2. Brief description on slides: https://www.slideshare.net/sk_klms/clique
  3. Article: Automatic subspace clustering of high dimensional data for data mining applications. In Proc. of 1998 ACM SIGMOD

Description
File source location: pyclustering/cluster/clique.py
Class name: clique
Interface of the class:

class clique:
    def __init__(<input parameters of the algorithm>):
        # initialization
    
    def process():
        # algorithm implementation
    
    def get_clusters():
        # return clustering result as list of indexes, for example
        # [ [0, 1, 2, 3], [4, 5, 6, 7] ].

Some specific for the algorithm information should be also displayed via additional methods.

Requirements

  • Algorithm implementation;
  • Whole code should have doxygen comments;
  • Switch on the source code to document generation (docs/doxygen_conf_pyclustering);
  • Unit-tests for the algorithm should be written (pyclustering/cluster/tests/unit/ut_clique.py);
  • Examples of clustering algorithm should be written (pyclustering/cluster/examples/clique_examples.py);

Optional (investigation is required before implementation):
Specific visualizer for the algorithm to display clustering results as follows (image from article):

class clique_visualizer:
    def visualize():
        # cluster visualization for 2D and 3D

clique

@annoviko annoviko added the Enhancement Tasks related to enhancement and development label Oct 27, 2017
@annoviko annoviko added this to the 0.7 (release point) milestone Oct 27, 2017
@annoviko annoviko removed this from the 0.7 (release point) milestone Feb 13, 2018
@annoviko annoviko added the Good First Issue Tasks that can be easily done by contributors label Nov 14, 2018
@Datamined
Copy link

Hello, thanks for your fantastic package.
Any progress with this method? Any ETA of release date?

@annoviko
Copy link
Owner Author

Hello, @Datamined ,
It is planned to be in release 0.9.0 that will be at the end of February - beginning of March.

@annoviko annoviko assigned annoviko and unassigned Polladin Jan 15, 2019
annoviko added a commit that referenced this issue Jan 27, 2019
annoviko added a commit that referenced this issue Jan 29, 2019
annoviko added a commit that referenced this issue Jan 29, 2019
…er implementation. Test template is prepared.
annoviko added a commit that referenced this issue Jan 31, 2019
annoviko added a commit that referenced this issue Feb 2, 2019
annoviko added a commit that referenced this issue Feb 6, 2019
@annoviko annoviko closed this as completed Feb 6, 2019
@farzanullah
Copy link

In clique, whenever i try to run code with 9 or more dimensions, it gives the error:

OSError: [WinError -529697949] Windows Error 0xe06d7363

What could be the reason for this?

@annoviko
Copy link
Owner Author

Hi @farzanullah ,

Could you please provide version of the library and code example how do you use this algorithm? Does it work with data with 8 dimensions?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Enhancement Tasks related to enhancement and development Good First Issue Tasks that can be easily done by contributors
Projects
None yet
Development

No branches or pull requests

4 participants