{"id":30848334,"url":"https://github.com/karan-sharma-aiml/future_ml_01","last_synced_at":"2026-04-29T22:36:29.136Z","repository":{"id":312050374,"uuid":"1044428662","full_name":"karan-sharma-aiml/FUTURE_ML_01","owner":"karan-sharma-aiml","description":" Hands-on Machine Learning project built with Python, Streamlit, Prophet \u0026 TensorFlow | Showcasing AIML skills through time-series forecasting \u0026 real-world data insights.","archived":false,"fork":false,"pushed_at":"2025-08-28T09:36:23.000Z","size":1205,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-08-28T14:28:23.740Z","etag":null,"topics":["aiml","data-visualisation","machine-learning","project","prophet","python","streamlit","tensorflow","timreactnative"],"latest_commit_sha":null,"homepage":"https://www.linkedin.com/in/karan-sharma-167957271","language":"Jupyter Notebook","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/karan-sharma-aiml.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":"2025-08-25T17:06:54.000Z","updated_at":"2025-08-28T09:53:00.000Z","dependencies_parsed_at":"2025-08-28T14:28:56.479Z","dependency_job_id":"6bcfc372-3846-4f86-82f1-cb2417669f95","html_url":"https://github.com/karan-sharma-aiml/FUTURE_ML_01","commit_stats":null,"previous_names":["karan-sharma-aiml/future_ml_01"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/karan-sharma-aiml/FUTURE_ML_01","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/karan-sharma-aiml%2FFUTURE_ML_01","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/karan-sharma-aiml%2FFUTURE_ML_01/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/karan-sharma-aiml%2FFUTURE_ML_01/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/karan-sharma-aiml%2FFUTURE_ML_01/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/karan-sharma-aiml","download_url":"https://codeload.github.com/karan-sharma-aiml/FUTURE_ML_01/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/karan-sharma-aiml%2FFUTURE_ML_01/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":273990234,"owners_count":25203301,"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":"2025-09-07T02:00:09.463Z","response_time":67,"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":["aiml","data-visualisation","machine-learning","project","prophet","python","streamlit","tensorflow","timreactnative"],"created_at":"2025-09-07T03:08:42.865Z","updated_at":"2026-04-29T22:36:29.111Z","avatar_url":"https://github.com/karan-sharma-aiml.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🚀 AI-Powered Sales Forecasting Dashboard — Advanced\n\n[![Python](https://img.shields.io/badge/Python-3.8+-blue.svg)](https://python.org)\n[![Streamlit](https://img.shields.io/badge/Streamlit-1.28+-red.svg)](https://streamlit.io)\n[![Prophet](https://img.shields.io/badge/Prophet-1.1+-green.svg)](https://facebook.github.io/prophet/)\n[![TensorFlow](https://img.shields.io/badge/TensorFlow-2.12+-orange.svg)](https://tensorflow.org)\n\n\u003e **Advanced Time Series Forecasting Dashboard with Multi-Model Ensemble, Uncertainty Quantification, and Explainable AI**\n\nA production-ready forecasting system that combines **Prophet**, **LSTM Neural Networks**, and **XGBoost** with comprehensive model explainability using **SHAP**.\n\n---\n\n## 📊 Dashboard Preview\n\n### 🏠 Main Dashboard  \n\n![Dashboard Overview](./screenshots/1.Dashbard%20section/01_dashbard_overview.png)\n\nUploaded sales data preview in real time\n\n---\n\n### 🔮 Forecasting Models  \n![Prophet Forecast](screenshots/2.Forecasting%20Section/04_prophet_forecast.png)  \n*Prophet model prediction with seasonality components*  \n\n![LSTM Forecast](screenshots/2.Forecasting%20Section/06_lstm_forecast.png)  \n*LSTM Neural Network with MC Dropout based uncertainty*  \n\n![Combined Models](screenshots/2.Forecasting%20Section/07_combined_prophet_lstm.png)  \n*Comparison of Prophet \u0026 LSTM predictions*\n\n---\n\n### 📈 Model Evaluation \u0026 Explainability  \n![Rolling Backtest](screenshots/3.Evaluation%20\u0026%20Explainability/08_rolling_bracket_table.png)  \n*Rolling-origin cross-validation metrics (MAE, RMSE, MAPE)*  \n\n###  SHAP Bar Chart  \n![SHAP Bar Chart](./screenshots/3.Evaluation%20%26%20Explainability/09_Sharp_bar_chart.png)\n\n### Feature Importance\n![Feature Importance](screenshots/3.Evaluation%20\u0026%20Explainability/10_feature_importance.png)  \n*XGBoost feature contributions visualized*\n\n```markdown\n## 📂 Project Structure - FUTURE_ML_01\n\n📂 FUTURE_ML_01/ - Root project folder\n├─ 📂 data/ - Contains dataset files\n│  └─ 📄 sample_sales.csv - Sample sales dataset\n├─ 📂 notebook/ - Jupyter notebooks for analysis\n│  ├─ 📓 eda.ipynb - Exploratory Data Analysis\n│  └─ 📓 model_experiments.ipynb - Model experimentation\n├─ 📂 screenshots/ - Screenshots of dashboard and results\n│  ├─ 📂 1.Dashboard Section/ - Dashboard UI images\n│  │  ├─ 🖼️ 01_dashboard_overview.png - Overview of dashboard\n│  │  ├─ 🖼️ 02_sidebar_controls.png - Sidebar controls screenshot\n│  │  └─ 🖼️ 03_data_preview.png - Sample data preview\n│  ├─ 📂 2.Forecasting Section/ - Forecasting model outputs\n│  │  ├─ 🔮 04_prophet_forecast.png - Prophet forecast chart\n│  │  ├─ 📅 05_prophet_components_main.png - Prophet main components\n│  │  ├─ 📆 05_prophet_yearly.png - Yearly trend components\n│  │  ├─ 🤖 06_lstm_forecast.png - LSTM forecast chart\n│  │  └─ ⚡ 07_combined_prophet_lstm.png - Combined forecast\n│  └─ 📂 3.Evaluation \u0026 Explainability/ - Evaluation visuals\n│     ├─ 📋 08_rolling_bracket_table.png - Rolling bracket table\n│     ├─ 📊 09_Sharp_bar_chart.png - Sharpe ratio chart\n│     └─ 🌟 10_feature_importance.png - Feature importance chart\n├─ 📂 src/ - Source code for models and utilities\n│  ├─ ⚙️ backtesting.py - Backtesting logic\n│  ├─ ⚙️ data_processing.py - Data cleaning and preprocessing\n│  ├─ ⚙️ lstm_model.py - LSTM model implementation\n│  ├─ ⚙️ prophet_model.py - Prophet model implementation\n│  ├─ ⚙️ utils.py - Utility functions\n│  └─ ⚙️ xgb_baseline.py - XGBoost baseline model\n├─ ⚙️ .gitignore - Git ignore file\n├─ 📜 LICENSE - License file\n├─ 📘 README.md - Project documentation\n├─ 📦 requirements.txt - Python dependencies\n└─ 🌐 streamlit_sales_forecast.py - Main Streamlit app\n\n\n## 🌟 Key Features\n- Upload custom sales CSV files (time series with date and sales columns)  \n- Automatic feature engineering: lag features, rolling averages, and seasonality indicators  \n- Multi-model sales forecasting using Prophet, LSTM (with MC dropout uncertainty), and XGBoost  \n- Visualize model forecasts with confidence intervals and trend/seasonality decomposition  \n- Rolling-origin cross-validation with performance metrics such as MAE, RMSE, MAPE, and coverage  \n- Explainability using SHAP for model interpretability  \n- Interactive Streamlit dashboard with real-time parameter tuning\n\n### 🔬 **Advanced ML Models**\n- **Facebook Prophet** with automatic seasonality detection and uncertainty intervals\n- **LSTM Neural Networks** with Monte Carlo Dropout for uncertainty quantification  \n- **XGBoost Baseline** with feature importance and SHAP explainability\n- **Multi-model ensemble** with rolling-origin cross-validation\n\n### 📊 **Professional Dashboard**\n- **Interactive Streamlit interface** with real-time parameter tuning\n- **Dynamic visualization** with Plotly charts\n- **Comprehensive data preview** with automated feature engineering\n- **Export functionality** for predictions and plots\n\n### 🎯 **Production-Ready Architecture**\n- **Modular codebase** with separation of concerns\n- **Testing suite** with pytest\n- **Config management** with YAML\n- **Professional docs** \u0026 clean organization\n\n---\n\n## 🛠️ Technical Architecture\n\n### **Pipeline**\n```\n\nData Ingestion → Feature Engineering → Model Training →\nPrediction Generation → Model Validation → Explainability Analysis\n\n````\n\n### **Tech Stack**\n- **Backend**: Python 3.8+ (Pandas, NumPy, scikit-learn ecosystem)\n- **ML Frameworks**: TensorFlow/Keras, Prophet, XGBoost\n- **Explainability**: SHAP\n- **Visualization**: Plotly\n- **UI**: Streamlit\n\n---\n\n## 🚀 Quick Start\n\n```bash\n# Clone repo\ngit clone https://github.com/karan-sharma-aiml/FUTURE_ML_01.git\ncd FUTURE_ML_01\n\n# Setup env\npython -m venv venv\nsource venv/bin/activate   # Windows: venv\\Scripts\\activate\n\n# Install dependencies\npip install -r requirements.txt\n\n# Run dashboard\nstreamlit run streamlit_sales_forecast.py\n````\n\n---\n\n## 📊 Model Performance (Example)\n\n| Model   | MAE   | RMSE  | MAPE  | Coverage% | Training Time |\n| ------- | ----- | ----- | ----- | --------- | ------------- |\n| Prophet | 0.00  | 0.00  | 0.00% | 95.2%     | 2.3s          |\n| LSTM    | 0.00  | 0.00  | 0.00% | 92.8%     | 45.7s         |\n| XGBoost | 70.03 | 85.21 | 12.4% | -         | 1.8s          |\n\n---\n\n## 🎯 Business Applications\n\n* **Retail Forecasting**: Inventory \u0026 demand planning\n* **Revenue Forecasting**: Budget allocation\n* **Resource Planning**: Workforce scheduling\n* **Supply Chain Optimization**: Procurement \u0026 logistics\n\n---\n\n## 🤝 Contributing\n\nContributions are welcome!\n\n1. Fork the repo\n2. Create a feature branch\n3. Commit changes + add docs/tests\n4. Submit PR 🎉\n\n---\n\n## 👨‍💻 Author\n\n**Karan Sharma**\n🎓 B.Tech CSE (AI/ML) Student @ CGC University, Mohali\n\n💡 Focus: Time Series | Deep Learning | Explainable AI\n📧 Email: [karan.sharma@email.com](mailto:karan.sharma@email.com)\n🔗 [LinkedIn](https://www.linkedin.com/in/karan-sharma-167957271)\n🐙 [GitHub](https://github.com/karan-sharma-aiml)\n\n---\n\n\u003cdiv align=\"center\"\u003e\n\n⭐ **Star this repo if you find it useful!** ⭐\n\n*Built with ❤️ for the AI/ML community*\n\n\u003c/div\u003e\n```\n\n---\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkaran-sharma-aiml%2Ffuture_ml_01","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkaran-sharma-aiml%2Ffuture_ml_01","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkaran-sharma-aiml%2Ffuture_ml_01/lists"}