Skip to content

Latest commit

Β 

History

2 Commits

Folders and files

NameName
Last commit message
Last commit date
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

🚦 FlowCast AI

Enterprise Smart Traffic Prediction Platform using Machine Learning, Deep Learning & Explainable AI

Python Streamlit XGBoost TensorFlow License


πŸ“Œ Overview

FlowCast AI is an enterprise-grade intelligent traffic prediction and analytics platform that combines Machine Learning, Deep Learning, Explainable AI, and interactive dashboards to forecast traffic conditions and support data-driven traffic management.

The platform performs complete end-to-end traffic analytics, including:

  • Traffic Data Analysis
  • Feature Engineering
  • Machine Learning Prediction
  • Deep Learning Forecasting
  • Explainable AI (SHAP & LIME)
  • Interactive Dashboard
  • Upload Custom Dataset
  • Real-Time Traffic Insights

πŸš€ Features

πŸ“Š Analytics Dashboard

  • Enterprise KPI Dashboard
  • Interactive Plotly Visualizations
  • Traffic Trends
  • Congestion Analysis
  • Weather Analytics
  • Road-wise Traffic Analysis
  • Vehicle Distribution
  • Correlation Heatmap
  • Download Reports

πŸ€– Prediction Engine

Supports multiple prediction models:

  • Random Forest
  • XGBoost
  • Best ML Model
  • LSTM Deep Learning
  • Ensemble Prediction

Prediction includes:

  • Traffic Volume Forecast
  • Congestion Detection
  • Confidence Score
  • Delay Estimation
  • AI Recommendation
  • Traffic Gauge
  • Model Comparison
  • Prediction History

🧠 Explainable AI

Model Interpretability using

  • SHAP Feature Importance
  • SHAP Waterfall Plot
  • SHAP Summary Plot
  • SHAP Dependence Plot
  • LIME Explanation
  • Global Feature Importance
  • Local Prediction Explanation
  • AI Decision Summary

πŸ“‚ Upload Dataset

Users can upload their own traffic dataset for prediction.

Supported format:

  • CSV

Validation includes:

  • Missing Values
  • Duplicate Records
  • Dataset Preview
  • Statistics
  • Download Clean Data

πŸ“· Dashboard Preview

🏠 Home

  • dashboard/screenshots/2_home_page

πŸ“Š Analytics

  • dashboard/screenshots/3_analytics_page

πŸ€– Prediction

  • dashboard/screenshots/4_prediction_page

🧠 AI Insights

  • dashboard/screenshots/5_ai_insights

πŸ“‚ Upload Dataset

  • dashboard/screenshots/6_upload_data

πŸ›  Project Structure

FlowCast-AI
β”‚
β”œβ”€β”€ dashboard/
β”‚   β”œβ”€β”€ app.py
β”‚   β”œβ”€β”€ assets/
β”‚   β”œβ”€β”€ pages/
β”‚   β”‚   β”œβ”€β”€ 01_Home.py
β”‚   β”‚   β”œβ”€β”€ 02_Analytics.py
β”‚   β”‚   β”œβ”€β”€ 03_Prediction.py
β”‚   β”‚   β”œβ”€β”€ 04_AI_Insights.py
β”‚   β”‚   └── 05_Upload_Data.py
β”‚   └── screenshots/
β”‚
β”œβ”€β”€ data/
β”‚   β”œβ”€β”€ raw/
β”‚   └── processed/
β”‚
β”œβ”€β”€ docs/
β”‚
β”œβ”€β”€ models/
β”‚   β”œβ”€β”€ artifacts/
β”‚   β”œβ”€β”€ deep_learning/
β”‚   β”œβ”€β”€ best_model.pkl
β”‚   └── xgboost_model.pkl
β”‚
β”œβ”€β”€ notebooks/
β”‚
β”œβ”€β”€ reports/
β”‚
β”œβ”€β”€ requirements.txt
β”‚
└── README.md

πŸ“ Machine Learning Pipeline

Raw Traffic Data
        β”‚
        β–Ό
Data Cleaning
        β”‚
        β–Ό
Preprocessing
        β”‚
        β–Ό
EDA
        β”‚
        β–Ό
Feature Engineering
        β”‚
        β–Ό
Random Forest
        β”‚
        β–Ό
XGBoost
        β”‚
        β–Ό
LSTM Deep Learning
        β”‚
        β–Ό
Model Evaluation
        β”‚
        β–Ό
Explainable AI
        β”‚
        β–Ό
Interactive Dashboard

πŸ“Š Models Used

Model Purpose
Random Forest Baseline Prediction
XGBoost High Accuracy Prediction
LSTM Sequential Traffic Forecasting
Ensemble Final Prediction

🧰 Technology Stack

Programming

  • Python

Data Processing

  • Pandas
  • NumPy

Machine Learning

  • Scikit-Learn
  • XGBoost

Deep Learning

  • TensorFlow
  • Keras

Explainable AI

  • SHAP
  • LIME

Visualization

  • Plotly
  • Matplotlib

Dashboard

  • Streamlit

πŸ“ˆ Workflow

Traffic Dataset
      ↓
Cleaning
      ↓
EDA
      ↓
Feature Engineering
      ↓
Model Training
      ↓
Model Evaluation
      ↓
Explainable AI
      ↓
Traffic Prediction Dashboard

▢️ Installation

Clone the repository

git clone https://github.com/aparna190417/FlowCast-AI.git

Move into project

cd FlowCast-AI

Install requirements

pip install -r requirements.txt

Run Streamlit

streamlit run dashboard/app.py

πŸ“Œ Future Improvements

  • Live Traffic API Integration
  • Google Maps Visualization
  • Real-Time Prediction
  • IoT Sensor Support
  • Cloud Deployment
  • Automated Model Retraining
  • REST API Support

πŸ‘©β€πŸ’» Developer

Aparna Patel

Data Analyst | Machine Learning | Deep Learning | Explainable AI

GitHub

https://github.com/aparna190417


⭐ If you like this project

Please consider giving it a ⭐ on GitHub.


πŸ“œ License

Β© 2026 Aparna Patel

About

Enterprise AI Traffic Prediction Platform using Machine Learning, Deep Learning, SHAP, LIME and Streamlit.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages