{"id":18021352,"url":"https://github.com/2mol/gump","last_synced_at":"2025-09-08T11:10:21.220Z","repository":{"id":66291354,"uuid":"163205479","full_name":"2mol/gump","owner":"2mol","description":null,"archived":false,"fork":false,"pushed_at":"2019-11-09T15:16:39.000Z","size":1873,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":5,"default_branch":"master","last_synced_at":"2025-04-04T17:26:36.075Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Haskell","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/2mol.png","metadata":{"files":{"readme":"Readme.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","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":"2018-12-26T18:08:33.000Z","updated_at":"2020-01-20T02:06:55.000Z","dependencies_parsed_at":"2023-02-22T04:46:14.804Z","dependency_job_id":null,"html_url":"https://github.com/2mol/gump","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/2mol/gump","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/2mol%2Fgump","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/2mol%2Fgump/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/2mol%2Fgump/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/2mol%2Fgump/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/2mol","download_url":"https://codeload.github.com/2mol/gump/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/2mol%2Fgump/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":274174271,"owners_count":25235203,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-09-08T02:00:09.813Z","response_time":121,"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":[],"created_at":"2024-10-30T06:09:36.595Z","updated_at":"2025-09-08T11:10:21.188Z","avatar_url":"https://github.com/2mol.png","language":"Haskell","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Gump\n\nan attempt to learn decision trees and random forests by implementing them myself.\n\n## explanation\n\nbasically, decision trees try to answer the question \"which feature makes the biggest difference\"\n\n## approach\n\nmy understanding of the algorithm(s). numbers like 1. are a necessary step, while bulletpoints are different choices we could make.\n\n\n1. read dataframe into a list of feature vectors and one target vector.\n2. look at each feature and measure information gain **if** we'd split on that one.\n  - variable can be continuous or categorical, changing the way we split.\n    - on categorical variables we can split on every value **or** do a binary split.\n    - on continuous variables we can find an optimal binary split value **or** we can discretize values, meaning we kinda get categorical buckets.\n3. find the feature with the highest value split, and perform that split.\n4. for each branch of that split (2 if binary, n if fully categorical) we get a filtered down subset of the measurement data above.\n5. repeat recursively.\n\nNote: make sure to have a stopping condition, so that splits don't result in too few datapoints.\n\n## open questions\n\n- does the target always have to be categorical?\n- bagging seems to be a bit of a choice, i.e. how fine-grained do we want to make the buckets\n  - can this choice be optimized as well?\n  - seems non-trivial, because a greedy algorithm might way overfit at this step\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F2mol%2Fgump","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F2mol%2Fgump","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F2mol%2Fgump/lists"}