{"id":19734794,"url":"https://github.com/sanghyukchun/sc_si","last_synced_at":"2026-05-14T06:35:54.298Z","repository":{"id":71717818,"uuid":"115583942","full_name":"SanghyukChun/SC_SI","owner":"SanghyukChun","description":null,"archived":false,"fork":false,"pushed_at":"2020-06-27T04:31:23.000Z","size":16,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-02-28T04:55:40.740Z","etag":null,"topics":["robust"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SanghyukChun.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2017-12-28T04:11:32.000Z","updated_at":"2020-06-27T04:31:25.000Z","dependencies_parsed_at":"2023-05-16T10:45:20.449Z","dependency_job_id":null,"html_url":"https://github.com/SanghyukChun/SC_SI","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/SanghyukChun/SC_SI","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SanghyukChun%2FSC_SI","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SanghyukChun%2FSC_SI/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SanghyukChun%2FSC_SI/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SanghyukChun%2FSC_SI/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SanghyukChun","download_url":"https://codeload.github.com/SanghyukChun/SC_SI/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SanghyukChun%2FSC_SI/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33013277,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-13T13:14:54.681Z","status":"online","status_checked_at":"2026-05-14T02:00:06.663Z","response_time":57,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["robust"],"created_at":"2024-11-12T00:39:35.397Z","updated_at":"2026-05-14T06:35:54.282Z","avatar_url":"https://github.com/SanghyukChun.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# SC_SI\n\nPython implementation of SC_SI (Subspace Clustering with Scalable and Iterative Algorithm).\nSC_SI solves multiple R-PCA (Robust PCA) problem to clustering.\n\nThe algorithm is proposed at my master thesis\n\n`Scalable Iterative Algorithm for Robust Subspace Clustering: Convergence and Initialization`\n\nYou can check full paper in the link:\n- https://sanghyukchun.github.io/home/media/papers/chun2016scsi.pdf\n- http://library.kaist.ac.kr/mobile/book/view.do?bibCtrlNo=649637\n\n## Usage\n\nIt is almost same as scikit-learn K-means [link](http://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html)\n```python\nimport numpy as np\nfrom sc_si.clustering.sc_si import SC_SI, MiniBatchSC_SI\n\nn_data = 10000\ndim_data = 500\nX = np.random.random((n_data, dim_data))\n\nn_clusters = 10\nn_components = 20\n# if you have a large dataset, use `MiniBatchSC_SI` instead.\nmodel = SC_SI(n_clusters=n_clusters, n_components=n_components, alpha=1.0, init='sc_si', n_init=3, max_iter=100, verbose=True)\nlabels = model.fit_predict(X)\n\nn_data2 = 1000\nX_unseen = np.random.random((n_data2, dim_data))\nlabels_unseen = model.predict(X_unseen)\n```\n\n## Optimization Hints\n\n1. alpha handles 'robustness' of the objective function. The objective function is more robust with less alpha.\n2. Use alpha=1.0. Theoretically alpha can be any number between (0, 2] but practically, I recommend to choose alpha as 1\n3. Use default initialization named SC_IN if size of dataset is not too large. Otherwise, use 'random' initialization with large n_init. Or, sampling datasets to initialization\n4. Use default svd_algorithm (subspace iteration). It is much faster and use less memory than exact SVD.\n5. For datasets with less outliers, use large beta (e.g. 10) otherwise, set beta = alpha\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsanghyukchun%2Fsc_si","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsanghyukchun%2Fsc_si","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsanghyukchun%2Fsc_si/lists"}