{"id":19070285,"url":"https://github.com/amogh7joshi/machine-learning-research","last_synced_at":"2026-04-17T06:03:29.987Z","repository":{"id":41449912,"uuid":"303551787","full_name":"amogh7joshi/machine-learning-research","owner":"amogh7joshi","description":"Research and implementations of machine learning algorithms and constructs, and deep learning models.","archived":false,"fork":false,"pushed_at":"2022-06-06T14:44:47.000Z","size":179,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-02-22T03:43:21.778Z","etag":null,"topics":["deep-learning","deep-learning-algorithms","machine-learning","neural-network","research"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/amogh7joshi.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}},"created_at":"2020-10-13T01:08:16.000Z","updated_at":"2022-03-30T23:55:16.000Z","dependencies_parsed_at":"2022-09-08T07:01:34.461Z","dependency_job_id":null,"html_url":"https://github.com/amogh7joshi/machine-learning-research","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/amogh7joshi/machine-learning-research","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amogh7joshi%2Fmachine-learning-research","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amogh7joshi%2Fmachine-learning-research/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amogh7joshi%2Fmachine-learning-research/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amogh7joshi%2Fmachine-learning-research/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/amogh7joshi","download_url":"https://codeload.github.com/amogh7joshi/machine-learning-research/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amogh7joshi%2Fmachine-learning-research/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31917372,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-16T18:22:33.417Z","status":"online","status_checked_at":"2026-04-17T02:00:06.879Z","response_time":62,"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":["deep-learning","deep-learning-algorithms","machine-learning","neural-network","research"],"created_at":"2024-11-09T01:17:54.610Z","updated_at":"2026-04-17T06:03:29.921Z","avatar_url":"https://github.com/amogh7joshi.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Machine Learning Research\n\nThis repository is dedicated to my own machine learning research, in essence implementations of algorithms\nfrom scratch as I attempt to learn them, construction of existing model architectures to attempt to get an understanding\nof the internal functions of them, or experimentation with different algorithms and constructs to see their result.\n\nMost of the implementations here are from scratch, using only NumPy and Matplotlib in order to use basic linear algebra operations \nand visualize data, respectively, but the high-level neural network and other model construct implementations make use of\nthe [Tensorflow](https://github.com/tensorflow/tensorflow) and [PyTorch](https://github.com/pytorch/pytorch) libraries.\n\nI primarily make these implementations in Python, due to not only the number of libraries but also because of the sheer ease that \nis writing code in Python (at least to me). However, sometimes when testing optimization I will add constructs in C++.\n\n![kmeans-example](diagrams/kmeans-example.png)\n\nWhile currently I have been primarily experimenting with supervised learning techniques, e.g. regression and even basic classification\nusing SVM/LDA, I have also tinkered with unsupervised learning, e.g. K-Means clustering (as pictured above), and PCA.\n\n## Navigation\n\nThe directory names are quite literal, so to find different algorithms you can find the directory that contains the name \nand then navigate through its subdirectories if necessary. Some directories that may need explanation (a dynamic list):\n\n- The `nn-basic` directory contains any implementations of neural networks from when I was first working with neural networks. \nFor example, my initial MNIST digit classification network resides there.\n\n## Implementations\n\nWhat use is there in constructing these algorithms if they are not applied to anything? The `implementations` directory contains actual\nusage of selected algorithms I've constructed on actual datasets, alongside a comparison to the [scikit-learn](https://github.com/scikit-learn/scikit-learn)\nimplementation of the same algorithm (on the same dataset). Some implementations to note (a growing list) include:\n\n1.  **Logistic Regression**: Implemented the Logistic Regression algorithm on the [Pima Indians Diabetes](https://www.kaggle.com/uciml/pima-indians-diabetes-database/discussion) \nDataset. On a train/test split of 0.75/0.25, the algorithm achieves an accuracy of 81.8%, compared to the scikit-learn equivalent of 82.4%.\n\n## Experimentation\n\nThe `experimental` directory is a special directory, as anything in there is purely experimental research and often may not even function \nas intended to. In these cases, I try and determine exactly what the issue is that prevents the algorithm from functioning correctly. More \ninformation can be found in the directory of that README.\n\n## License\n\nAll of the algorithms, constructs, and implementations in this library are licensed under the MIT License. You are free to work with them as you desire.\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Famogh7joshi%2Fmachine-learning-research","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Famogh7joshi%2Fmachine-learning-research","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Famogh7joshi%2Fmachine-learning-research/lists"}