https://github.com/prajakta1321/streetml-a-cityscape-traffic-volume-prognostication

StreetML leverages ML learning techniques to revolutionize urban traffic prediction through precise volume prognostication, aiming to enhance cityscape mobility through data-driven insights.
https://github.com/prajakta1321/streetml-a-cityscape-traffic-volume-prognostication

catboostregressor data datavisualisation exploratory-data-analysis lightgbm-regressor linearregression machine-learning machine-learning-algorithms predictive-analytics random-forest-regression xgboost-regression

Last synced: about 1 month ago
JSON representation

StreetML leverages ML learning techniques to revolutionize urban traffic prediction through precise volume prognostication, aiming to enhance cityscape mobility through data-driven insights.

• Host: GitHub
• URL: https://github.com/prajakta1321/streetml-a-cityscape-traffic-volume-prognostication
• Owner: prajakta1321
• Created: 2024-12-12T13:31:57.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2025-01-08T08:28:15.000Z (4 months ago)
• Last Synced: 2025-01-08T08:29:24.432Z (4 months ago)
• Topics: catboostregressor, data, datavisualisation, exploratory-data-analysis, lightgbm-regressor, linearregression, machine-learning, machine-learning-algorithms, predictive-analytics, random-forest-regression, xgboost-regression
• Language: Jupyter Notebook
• Homepage:
• Size: 2.87 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
## MachineHack.The-Great-Indian-Hiring-Hackathon-2024

 StreetML: Cityscape Traffic Volume Prognostication

## 🎯  Project Overview

A regression-based machine learning project focused on predicting urban traffic volumes using ensemble learning techniques. Currently in development as part of The Great Indian Hiring Hackathon 2024.

## ✅ Current Implementation:

✔ Model Evaluation

✔ Regression problem targeting traffic volume prediction

✔ Model performance evaluated using R² (R-squared) value

✔ R² metric chosen to determine goodness of fit and prediction accuracy

✔ Higher R² values indicate better model performance

## ✅ Status

🚧 Work in Progress 

✔ Actively developing and optimizing models

✔ Testing various feature engineering approaches

v [Future updates will include results and performance metrics]

## ✅ Tech Stack

✔ Python

✔ scikit-learn

✔ XGBoost

✔ pandas

✔ numpy

✔ seaborn

✔ matplotlib

# DATASET OVERVIEW :

![image](https://github.com/user-attachments/assets/058ac80f-f5ba-44bd-954d-4d7af8584088)

# FEATURE ENGINEERING :

![image](https://github.com/user-attachments/assets/b32f03a1-0c21-4486-8878-e60739ecaa48)

# NEW FEATURES :

![image](https://github.com/user-attachments/assets/59dc8985-8a66-41ac-978f-5e52079b2dc1)

![image](https://github.com/user-attachments/assets/8a316b05-58eb-4fba-a821-c0e34ab1a509)

# SHAP VISUALIZATION : 

![image](https://github.com/user-attachments/assets/2a69b41a-5299-481b-b8e8-3c7b40ff0011)

# algorithm performance :

![image](https://github.com/user-attachments/assets/d659d672-861f-4058-873e-ce462f5fe6ff)