{"id":19936139,"url":"https://github.com/proafxin/object-detection","last_synced_at":"2026-04-12T18:54:29.383Z","repository":{"id":204909450,"uuid":"712946756","full_name":"proafxin/object-detection","owner":"proafxin","description":"Implementation of different object detection algorithms.","archived":false,"fork":false,"pushed_at":"2023-12-16T22:21:38.000Z","size":625,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"develop","last_synced_at":"2025-01-12T02:39:26.233Z","etag":null,"topics":["coverage","object-detection","poetry","pytest","python","pytorch","tox","yolo"],"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/proafxin.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":"2023-11-01T14:29:29.000Z","updated_at":"2023-11-10T06:33:21.000Z","dependencies_parsed_at":null,"dependency_job_id":"a139ba00-6208-4a0f-b482-833714448531","html_url":"https://github.com/proafxin/object-detection","commit_stats":null,"previous_names":["proafxin/object-detection"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/proafxin%2Fobject-detection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/proafxin%2Fobject-detection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/proafxin%2Fobject-detection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/proafxin%2Fobject-detection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/proafxin","download_url":"https://codeload.github.com/proafxin/object-detection/tar.gz/refs/heads/develop","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":241362371,"owners_count":19950552,"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","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":["coverage","object-detection","poetry","pytest","python","pytorch","tox","yolo"],"created_at":"2024-11-12T23:23:50.310Z","updated_at":"2025-11-24T23:02:24.079Z","avatar_url":"https://github.com/proafxin.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Object Detection\n\nLibrary of different object detection algorithms. You need a CUDA compatible GPU for this to work.\n\n[![Trunk Check CI](https://github.com/proafxin/object-detection/actions/workflows/trunk.yml/badge.svg)](https://github.com/proafxin/object-detection/actions/workflows/trunk.yml)\n\n[![Upload Code Coverage CI](https://github.com/proafxin/object-detection/actions/workflows/ci.yml/badge.svg)](https://github.com/proafxin/object-detection/actions/workflows/ci.yml)\n\n[![codecov](https://codecov.io/gh/proafxin/object-detection/graph/badge.svg?token=ZM9kss9LrH)](https://codecov.io/gh/proafxin/object-detection)\n\n## Development environment\n\nThis project uses trunk for linting and formatting. Check \u003chttps://trunk.io/products/check\u003e\n\nFor dependency, `poetry` is used: \u003chttps://python-poetry.org/\u003e\n\nGo to root directory and run `poetry install`. But first make sure you have CUDA in your system along with all necessary packages. You can follow \u003chttps://github.com/proafxin/cuda-gpu-scripts\u003e for CUDA related instructions. Make sure CUDA can interact with your GPU. The best way to ensure that is to go to python shell and run\n\n```python\nimport torch\ntorch.cuda.is_available()\n```\n\nYou should see `True` as output. Finally, for virtual environment management, automated testing and task management, `tox` is used: \u003chttps://tox.wiki/\u003e\n\nInstall `tox` in your system using `python3 -m pip install -U tox` and then run `tox` from the project directory.\n\n## Design philosophy for YOLO model implementation\n\nThe idea behind the structure in this package is that we want to make the model reusable and easily configurable.\nWe will say, a model consists of some `units`. Each such unit can consist of `layers` or `blocks` or `cells`.\nCell is an individual component such as Convolution, MaxPool, ReLU or other activation function, etc.\nA block consists of some cells. Sometimes a single component is repeated multiple times.\nTo make it easy for us to specify the model, instead of adding the repeated components individually, we simply specify a repeition factor.\nThe same idea goes for a layer. A layer consists of some blocks which may be repeated.\nThe idea of `unit` may be confusing at first. First of all, it completely depends on the user. However, a good idea to make a separation would be at where we take maxpool or connected layers. See the following figure:\n![Yolov1 architecture](https://i.imgur.com/sIfCQKf.png)\nHere, there is a maxpool after the first convolution. So that will be our first unit: convolution combined with maxpool. Same for second unit.\nIn third unit, we have 4 convolution and then maxpool.\nIn fourth unit, we have two convolutions which are repeated, then two convolutions and a maxpool layer, and so on.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fproafxin%2Fobject-detection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fproafxin%2Fobject-detection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fproafxin%2Fobject-detection/lists"}