This project aims to train a machine learning model to predict the fuel efficiency (miles per gallon) of vehicles based on various attributes such as cylinders, displacement, horsepower, and weight. Also check out this repository for how I deployed the model with Flask and Heroku: https://github.com/Abhishek-Balram/ml-flask-app
The dataset used in this project is the "Auto MPG" dataset from the UCI Machine Learning Repository. It contains 398 instances and 8 attributes, including the target variable "mpg".
The project uses a supervised learning approach, specifically a regression task. The following steps were performed:
- Loading the dataset
- Exploratory Data Analysis (EDA) to understand the dataset and visualize the relationships between features.
- Data Preparation to handle missing values in the dataset and to create new informative features
- Training and evaluation of various regression models, including Linear Regression, Decision Tree, and Random Forest.
- Hyperparameter tuning using techniques like Grid Search to find the best model configuration.
- Evaluating the final model
- Saving the trained model in binary format for future use and deployment.
- See this repository for how I deployed the model with Flask and Heroku: https://github.com/Abhishek-Balram/ml-flask-app