{"id":27249515,"url":"https://github.com/karanjangid8656/neuro-forge","last_synced_at":"2026-04-11T10:33:03.657Z","repository":{"id":287269128,"uuid":"964175684","full_name":"KaranJangid8656/neuro-forge","owner":"KaranJangid8656","description":"NeuroForge is a powerful, intuitive, and modern drag-and-drop platform to design, train, and export neural networks—all in your browser. ","archived":false,"fork":false,"pushed_at":"2025-04-10T20:18:26.000Z","size":133,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-11T00:17:54.103Z","etag":null,"topics":["css","framer-motion","javascript","nextjs","reactjs","shadcn","tailwind","tensorflow","typescript"],"latest_commit_sha":null,"homepage":"https://neuro-forge.vercel.app/","language":"TypeScript","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/KaranJangid8656.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":"2025-04-10T20:02:03.000Z","updated_at":"2025-04-10T20:18:30.000Z","dependencies_parsed_at":"2025-04-10T21:22:46.634Z","dependency_job_id":"e5c6d893-9c19-407c-848c-bd583158e46a","html_url":"https://github.com/KaranJangid8656/neuro-forge","commit_stats":null,"previous_names":["karanjangid8656/neuro-forge"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KaranJangid8656%2Fneuro-forge","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KaranJangid8656%2Fneuro-forge/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KaranJangid8656%2Fneuro-forge/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KaranJangid8656%2Fneuro-forge/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/KaranJangid8656","download_url":"https://codeload.github.com/KaranJangid8656/neuro-forge/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248317721,"owners_count":21083530,"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":["css","framer-motion","javascript","nextjs","reactjs","shadcn","tailwind","tensorflow","typescript"],"created_at":"2025-04-11T00:17:58.484Z","updated_at":"2025-12-30T20:02:35.675Z","avatar_url":"https://github.com/KaranJangid8656.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🧠 NeuroForge\n\n**NeuroForge** is a powerful, intuitive, and modern drag-and-drop platform to design, train, and export neural networks—all in your browser. Built with cutting-edge web technologies, it brings the full power of machine learning and neural network visualization to users with no need for backend services.\n\n\u003e 🚀 Visual. Interactive. Intelligent. Open-source.\n\n---\n\n## 🌐 Live Demo\n\n👉 [Try it out now](https://neuro-forge.vercel.app/)\n\n---\n\n## 📦 Tech Stack\n\n| Layer              | Technology                     |\n|--------------------|---------------------------------|\n| Frontend           | Next.js, React, TypeScript      |\n| ML Engine          | TensorFlow.js                   |\n| UI Styling         | Tailwind CSS, shadcn/ui         |\n| Animation          | Framer Motion                   |\n| Graph Visualization| ReactFlow                       |\n| Charting           | Recharts                        |\n| Icons              | Lucide React                    |\n| Tooling            | ESLint, Prettier                |\n\n---\n\n## 🔧 Core Features\n\n### ⚙️ Drag-and-Drop Model Builder\n- Visual interface to build neural networks\n- Supports core, convolutional, and recurrent layers\n- Layer types: Input, Dense, Dropout, Conv2D, MaxPooling2D, Flatten, LSTM, Activation\n\n### 🧩 Layer Configuration Panel\n- Dynamic UI for setting layer-specific parameters\n- Mobile-responsive layout\n- Smart validation for layer compatibility\n\n### 📊 Real-Time Training\n- In-browser training using TensorFlow.js\n- Live metrics for accuracy and loss\n- Supports MNIST and XOR datasets\n\n### 📈 Training Visualization\n- Real-time charts using Recharts\n- Displays training/validation accuracy \u0026 loss\n- Responsive and interactive\n\n### 🧠 Code Generator\n- Export trained models as:\n  - Python (TensorFlow/Keras)\n  - JavaScript (TensorFlow.js)\n- One-click copy \u0026 download\n\n### 💡 AI-Powered Suggestions\n- Automatic tips for:\n  - Model architecture improvements\n  - Hyperparameter tuning\n  - Regularization techniques\n\n### 🧾 Live Console Panel\n- Timestamped training logs\n- Real-time error and success messages\n- Auto-scroll and log filtering\n\n---\n\n## 🗺 Architecture Overview\n\n### 🧭 Data Flow\n\n1. **User Interaction** → UI event triggers callback  \n2. **State Update** → ModelBuilder.tsx updates core state  \n3. **Re-render** → Components reflect new data  \n4. **Training** → TensorFlow.js runs training loop  \n5. **Metrics Callback** → UI updated in real-time  \n6. **Visualization** → Charts + logs reflect new progress\n\n### 🔑 Key State Objects\n\n- `layers`: Layer configurations  \n- `selectedLayer`: Layer being edited  \n- `modelConfig`: Training settings  \n- `trainingMetrics`: Accuracy/loss values  \n- `logs`: Live console messages  \n- `model`: TensorFlow.js model instance  \n\n---\n\n## 🚨 Challenges \u0026 Solutions\n\n| Challenge                     | Solution |\n|------------------------------|----------|\n| In-browser training perf.    | Async TF.js ops with UI syncing |\n| Visualizing complex models   | Simplified layer cards + mobile layouts |\n| TensorFlow compatibility     | Feature detection + fallback support |\n| State complexity             | Centralized state in ModelBuilder.tsx |\n\n---\n\n## 🔮 Advanced Capabilities\n\n- ✅ **Intelligent Layer Validation**\n- ✅ **Dataset-Adaptive Training**\n- ✅ **Progressive Enhancement**\n- ✅ **Dynamic Code Rendering**\n- ✅ **Custom Mobile Navigation**\n\n---\n\n## 🛠️ Installation\n\n```bash\ngit clone https://github.com/your-username/neuroforge.git\ncd neuroforge\nnpm install\nnpm run dev\n```\n\n\n\n## 👥 Contributors\n\nMade with ❤️ by **Karan Suthar**  \nWant to contribute? PRs are welcome!\n\n---\n\n## 📜 License\n\n**MIT License** — feel free to use, modify, and share 🌐\n\n---\n\n## 🤖 Future Plans\n\n- ✅ Support for more datasets (CIFAR-10, IMDB)  \n- ✅ Export to ONNX  \n- ✅ Training history downloads  \n- ✅ Real-time collaboration (multi-user)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkaranjangid8656%2Fneuro-forge","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkaranjangid8656%2Fneuro-forge","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkaranjangid8656%2Fneuro-forge/lists"}