{"id":25850660,"url":"https://github.com/axsk/sqra.jl","last_synced_at":"2026-04-28T17:33:13.829Z","repository":{"id":37488595,"uuid":"383432946","full_name":"axsk/Sqra.jl","owner":"axsk","description":"WIP scrapbook, containing way too much.. partially exported to SqraCore.jl, ISOKANN.jl","archived":false,"fork":false,"pushed_at":"2023-01-04T15:23:25.000Z","size":9900,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-09-09T13:28:42.096Z","etag":null,"topics":["julia","sparse-boxes","sqra"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/axsk.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}},"created_at":"2021-07-06T10:42:58.000Z","updated_at":"2025-01-07T07:44:33.000Z","dependencies_parsed_at":"2023-02-02T17:15:33.880Z","dependency_job_id":null,"html_url":"https://github.com/axsk/Sqra.jl","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/axsk/Sqra.jl","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/axsk%2FSqra.jl","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/axsk%2FSqra.jl/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/axsk%2FSqra.jl/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/axsk%2FSqra.jl/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/axsk","download_url":"https://codeload.github.com/axsk/Sqra.jl/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/axsk%2FSqra.jl/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32392294,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-28T14:34:11.604Z","status":"ssl_error","status_checked_at":"2026-04-28T14:32:37.009Z","response_time":56,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["julia","sparse-boxes","sqra"],"created_at":"2025-03-01T12:31:57.207Z","updated_at":"2026-04-28T17:33:13.813Z","avatar_url":"https://github.com/axsk.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Sqra.jl\n\nThis \"package\" grew from the work on and with the 'Square Root Approximation' (SQRA).\nDuring that time it 'evolved' into a collection of different applications.\nThe main use case was the comparison of the *committors* of a *Lennard-Jones cluster* system computed via the *SQRA* on either *SparseBoxes* or\n*Voronoi tesselations*.\n\nIt furthermore contains the code for the convergence analysis of the SQRA in the Paper\n\n## Summary of contents:\n- Lennard-Jones Clusters\n- SQRA\n- SparseBoxes\n- Committors, classically via Q and via neural networks\n- ISOKANN, clasically via power iterations and via residual loss\n- error computation on voronoi/sparsebox discretizations\n\n## Contents:\n\n- permadict.jl - HD memoization implemented by a Dict forwarding read/writes to the hard disk\n- eulermaruyama.jl - An (adaptive!) Euler Maruyama SDE solver with a maximal dy threshold \n- picking.jl - Implementation of the Picking algorithm (by Vedat Durmaz)\n- voronoi_lp.jl - Linear Program to determine the adjacency of Voronoi cells (by Han Cheng Lie)\n- sparseboxes.jl - this was sbmatrix.jl once, now switched to the Dict representation\n- spdict.jl - Representation of SparseBoxes via Dict\n- spdistancs.jl - Calculate $\\int f-g dx$ where $f,g$ are defined on different SparseBox discretizations\n- errors.jl - Calculate $\\int f-g dx$ for $f,g$ Voronoi-Voronoi / Voronoi-SparseBox discretizations\n- sqra_core.jl - The SQRA core algorithm\n- committor.jl - Solver for the committor problems given a generator matrix Q\n- models.jl - TripleWell and Lennard-Jones cluster models (with LJ Normal Form plot)\n- experiment.jl / convergence.jl - comparison of SparseBoxes and Voronoi-Picking SQRA committor results \n- experiment2.jl - simpler rewrite of experiment.jl Experiments\n- mcerror.jl - Adaptive Voronoi MC Volume computation (to a specified confidence) (not yet in VoronoiGraph.jl)\n- mcfinvol.jl - Alternative to SQRA by using the Voronoi MC method to compute the boundary fluxes\n- martin/*.jl - Code for the co\n\nFurther contents:\n- archive/batch.jl - batch computation and comparison of the different discretization resolutions\n- archive/boxtrie.jl - attempt of writing a *Trie* datastructure for the *SparseBoxes*\n- archive/isokann.jl - *ISOKANN*, either classically via the *Fixed Point Iteration* or via *Residual Minimization*\n- archive/lennardjones.jl - first attempt at implementing the Lennard Jones Clusters\n- archive/metasgd.jl - SGD on the SGD learning rate by linear regression on past improvements\n- archive/molly.jl - attempt to run Molecular Dynamics via Molly.jl\n- archive/nestedgrad*.jl - Tests to single out AD problems with nested gradients\n- archive/neurcomm.jl - A custom approach to compute committors via neural networks, inspired by (Khoo et al and Li et al) (actually this was quite interesting, worth looking into again)\n- archive/sbmatrix.jl - the first implementation of sparseboxes\n- archive/trie.jl - and another trie implementation\n\nNotable efforts went into the code for the SparseBoxes (i.e. a structured \"matrix\" approach, then tries, finally the Dict) and their utility functions (cartesiancoords, merge, dists).\nThis was also the birthplace of the VoronoiGraph.jl package, which got outsourced, and a testbed for neural committors (neurcomm.jl) and ISOKANN.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faxsk%2Fsqra.jl","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faxsk%2Fsqra.jl","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faxsk%2Fsqra.jl/lists"}