{"id":30290674,"url":"https://github.com/saptakbhoumik/tinyvision","last_synced_at":"2025-08-16T23:50:57.079Z","repository":{"id":306685542,"uuid":"1026936890","full_name":"SaptakBhoumik/TinyVision","owner":"SaptakBhoumik","description":"TinyVision is an evolving project focused on designing ultra-lightweight image classification models with minimal parameter counts. The goal is to explore what’s actually necessary for fundamental vision tasks by combining handcrafted feature preprocessing with highly efficient CNN architectures.","archived":false,"fork":false,"pushed_at":"2025-08-06T15:45:29.000Z","size":11005,"stargazers_count":10,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-08-06T17:44:42.719Z","etag":null,"topics":["computer-vision","machine-learning","python","python3","pytorch","vision"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"cc-by-sa-4.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SaptakBhoumik.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,"zenodo":null}},"created_at":"2025-07-27T00:16:10.000Z","updated_at":"2025-08-03T17:55:30.000Z","dependencies_parsed_at":"2025-07-27T03:53:26.624Z","dependency_job_id":null,"html_url":"https://github.com/SaptakBhoumik/TinyVision","commit_stats":null,"previous_names":["saptakbhoumik/tinyvision"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/SaptakBhoumik/TinyVision","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SaptakBhoumik%2FTinyVision","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SaptakBhoumik%2FTinyVision/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SaptakBhoumik%2FTinyVision/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SaptakBhoumik%2FTinyVision/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SaptakBhoumik","download_url":"https://codeload.github.com/SaptakBhoumik/TinyVision/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SaptakBhoumik%2FTinyVision/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":270786434,"owners_count":24644563,"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":"2025-08-16T02:00:11.002Z","response_time":91,"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":["computer-vision","machine-learning","python","python3","pytorch","vision"],"created_at":"2025-08-16T23:50:52.414Z","updated_at":"2025-08-16T23:50:57.021Z","avatar_url":"https://github.com/SaptakBhoumik.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🧠 TinyVision: Compact Vision Models with Minimal Parameters\n\n**TinyVision** is an evolving research project focused on designing **ultra-lightweight image classification models** with minimal parameter counts. The goal is to explore what’s *actually necessary* for fundamental vision tasks by combining **handcrafted feature preprocessing** with highly efficient CNN architectures.\n\n📦 **Current Release**: [v2.0.0](https://github.com/SaptakBhoumik/TinyVision/releases/tag/v2.0.0) \n\n🔖 **Zenodo DOI**: [10.5281/zenodo.16467349](https://doi.org/10.5281/zenodo.16467349)  \n\n📁 **Latest Results \u0026 Code**: See the `cat_vs_dog_classifier/final/v2` directory\n\n\u003e ⚠️ This release **does not include a paper**, but focuses on the **codebase**, experiment results, and reproducible training scripts. A deeper analysis and formal documentation will come in future updates.\n\n---\n\n## 🚧 Project Status\n\n- ✅ **Cat vs Dog Classification**  \n  First completed task using a 25,000-image dataset with handcrafted preprocessing + compact CNNs.  \n  - Achieved **up to 86.87% test accuracy** with models under **12.5k parameters**\n  - Several models under **5k parameters** reached over **83% accuracy**, showcasing strong efficiency-performance trade-offs.\n  - 📂 Final results and code for this task are in the `cat_vs_dog_classifier/final/v2` directory.\n\n---\n\n## 🧪 What's Coming Next\n\n- 📊 Add thorough **performance analysis** of model architectures to understand why something works while others don't\n- 🧩 Explore new **vision tasks** (edge detection, object detection, etc.) with compact models\n- 📖 Expand **documentation**, architecture diagrams, and visualizations\n- 🧠 Log and reflect on **failed or inconclusive experiments** critical for understanding design boundaries\n\n---\n\n## 🤝 Contributing\n\nThis project is currently personal and tracks my ongoing experiments.  \nI’m **not accepting pull requests**, but you're welcome to:\n\n- 📬 Open an [issue](https://github.com/SaptakBhoumik/TinyVision/issues) for discussion or feedback  \n- 💌 Reach me at: `saptakbhoumik.acad@gmail.com`\n- 📢 Follow me on [X](https://x.com/saptakbhoumik)\n\n---\n\n## 💡 Philosophy\n\n\u003e Small models aren't just about speed—they’re a design challenge.  \n\u003e *How much can we cut before it breaks? What’s essential? What’s fluff?*\n\nTinyVision is my attempt to find those answers.\n\n---\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsaptakbhoumik%2Ftinyvision","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsaptakbhoumik%2Ftinyvision","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsaptakbhoumik%2Ftinyvision/lists"}