{"id":20826830,"url":"https://github.com/arkaung/cardenity","last_synced_at":"2026-04-26T14:31:08.053Z","repository":{"id":70134938,"uuid":"192373573","full_name":"ArkAung/cardenity","owner":"ArkAung","description":"Classify car manufacturers and models from images","archived":false,"fork":false,"pushed_at":"2023-07-06T21:41:11.000Z","size":3059,"stargazers_count":0,"open_issues_count":1,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-01-18T17:50:41.714Z","etag":null,"topics":["classification-model","deep-learning"],"latest_commit_sha":null,"homepage":null,"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/ArkAung.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":"2019-06-17T15:33:28.000Z","updated_at":"2019-06-17T15:56:49.000Z","dependencies_parsed_at":"2024-11-17T23:10:19.534Z","dependency_job_id":"39f7e31a-16a9-4fa5-a5b7-ebe0b419cd9b","html_url":"https://github.com/ArkAung/cardenity","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ArkAung%2Fcardenity","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ArkAung%2Fcardenity/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ArkAung%2Fcardenity/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ArkAung%2Fcardenity/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ArkAung","download_url":"https://codeload.github.com/ArkAung/cardenity/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243174244,"owners_count":20248272,"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":["classification-model","deep-learning"],"created_at":"2024-11-17T23:10:12.593Z","updated_at":"2025-12-24T15:08:02.859Z","avatar_url":"https://github.com/ArkAung.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"Cardenity\n=========\n\nYou can view the notebooks without downloading any dataset or running anything.\nClone repo and run `jupyter notebook` in command line/terminal.\n\n- Run `./download_dataset.sh`\n    * This will download the dataset and unzip in `datasets` directory\n- Open notebook: `Data Analysis and Data Genration.ipynb`\n    * This is where you can check out data analysis steps. Commentaries are in the notebook.\n- Open notebook: `Model Training.ipynb`\n    * This is where you can check model training procedures. Commentaries are in the notebook.\n- Open notebook: `Model Testing.ipynb`\n    * This is where you can check model being evaluated on test set and qualitative results on test set.\n     Commentaries are in the notebook.\n\n\n## Ideas\n\n- **Label separation training**: Original labels are further broken down for the model\nto treat learning make and model as multi-class multiple prediction problem.\n(Details in `Data Analysis and Data Generation` Notebook)\n\n- **Synthetic Data Generation**: The original dataset is quite small and unbalanced. Data\ncan be balanced by synthetically generating more data from original images.\n(Details in `Data Analysis and Data Generation` Notebook)\n\n- **Data Augmentation**: Every car has a character and even seeing A-pillar and B-pillar\nwill allow experts and enthusiasts to identify the make, model and year of the car\npretty accurately. As a car enthusiast myself, I have tried doing that exercise\nto get an idea of human baseline on identifying the vehicle just from parts of\nit. Understanding the nature of data very well before actually building a deep learning\nmodel is absolutely necessary and domain knowledge can help a lot as well. Thus,\nthe images will be cropped, flipped, rotated, saturation, hue, brightness changed.\nRandom cropping should help a lot.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Farkaung%2Fcardenity","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Farkaung%2Fcardenity","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Farkaung%2Fcardenity/lists"}