{"id":20328722,"url":"https://github.com/inborastudio/aigender","last_synced_at":"2025-08-26T00:07:35.101Z","repository":{"id":296651056,"uuid":"852408749","full_name":"InboraStudio/AIGender","owner":"InboraStudio","description":"Gender Detection Network A highly optimized machine learning project that detects gender from photos using a pre-trained neural network model. Fast and accurate, ready to deploy for image-based gender classification tasks.","archived":false,"fork":false,"pushed_at":"2025-07-27T08:45:08.000Z","size":56,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-07-27T10:30:10.639Z","etag":null,"topics":["ai","aimodel","classfication","gender-classification","gender-recognition","neuralnetwork"],"latest_commit_sha":null,"homepage":"https://huggingface.co/spaces/DrChamyoung/AIGender","language":"Python","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/InboraStudio.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,"zenodo":null}},"created_at":"2024-09-04T18:51:21.000Z","updated_at":"2025-07-27T08:45:11.000Z","dependencies_parsed_at":"2025-06-01T14:07:29.608Z","dependency_job_id":"0e282972-8854-40ba-a502-72762b96bf5e","html_url":"https://github.com/InboraStudio/AIGender","commit_stats":null,"previous_names":["inborastudio/aigender"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/InboraStudio/AIGender","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/InboraStudio%2FAIGender","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/InboraStudio%2FAIGender/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/InboraStudio%2FAIGender/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/InboraStudio%2FAIGender/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/InboraStudio","download_url":"https://codeload.github.com/InboraStudio/AIGender/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/InboraStudio%2FAIGender/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":272149880,"owners_count":24882043,"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":"2025-08-25T02:00:12.092Z","response_time":1107,"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":["ai","aimodel","classfication","gender-classification","gender-recognition","neuralnetwork"],"created_at":"2024-11-14T20:07:34.426Z","updated_at":"2025-08-26T00:07:35.092Z","avatar_url":"https://github.com/InboraStudio.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# AI Gender\n\n```A Project Build to Test my New GPU 😊 ```\n\n**AI Gender** is a compact, high-performance gender classification project that uses deep learning techniques to detect gender from facial images. This solution is engineered specifically for environments with **limited computational resources** such as:\n\n* Legacy laptops and desktops\n* Raspberry Pi devices\n* Embedded systems\n* Low-cost cloud VMs\n\nThe project is available as a **Gradio demo on Hugging Face Spaces**, and can be deployed or integrated into lightweight systems easily.\n## About\nThe core of **AI Gender** is a **pre-trained convolutional neural network** model. It accepts an image input, processes it to extract facial features, and classifies the gender as either male or female.\n\nUnlike large-scale models that require dedicated GPUs, this project prioritizes mathematical efficiency and architectural optimization.\n## Mathematical Foundation\n\n* Uses **Convolutional Neural Networks (CNNs)** optimized for small-scale inputs\n* Reduces floating point operations (FLOPs) using smaller kernels and fewer channels\n* Implements **transfer learning** with frozen backbone layers to minimize active computation\n* Model size is trimmed using **quantization-aware training** (QAT)\n* Confidence score output is based on softmax probability distribution\n\nKey equations:\n\n1. Convolution Operation:\n   $(I * K)(x, y) = \\sum_m \\sum_n I(m, n) \\cdot K(x - m, y - n)$\n\n2. Softmax:\n    $\\text{Softmax}(z_i) = \\frac{e^{z_i}}{\\sum_j e^{z_j}}$\n\nThese optimizations ensure the model runs efficiently on minimal hardware.\n\n---\n\n## Low-End Technology Support\n\nTo make this model runnable on **super low-end tech**, we applied the following:\n\n* **Backbone**: MobileNetV2 / ResNet18 (for small size and speed)\n* **RAM Usage**: \u003c 200MB during inference\n* **Disk Footprint**: Model file under 10MB\n* **CPU Only**: No CUDA or GPU dependencies\n* **Lazy Loading**: Loads only necessary components on start\n* **Zero External Heavy Libraries**: Avoids OpenCV, dlib, or other memory-intensive tools\n\nThis makes it suitable for real-time or batch inference on constrained devices.\n\n---\n\n## Demo (Hugging Face)\n\nTry the model online here:\n\n[https://huggingface.co/spaces/your-username/ai-gender](https://huggingface.co/spaces/DrChamyoung/AIGender)\n\n---\n\n## Notes\n\nThis project was created as an accessible AI showcase, proving that **practical deep learning** can be done without expensive hardware. It’s ideal for classrooms, research demonstrations, and offline AI use cases.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Finborastudio%2Faigender","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Finborastudio%2Faigender","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Finborastudio%2Faigender/lists"}