End-to-End-Machine-Learning-MLflow-Project

Introduction About the Data :

The dataset The goal is to predict price of given car (Regression Analysis).

There are 19 independent variables:

• Make : Company of the car
• Year : Manufacturing Year of the car
• Kilometer : Total kilometers Driven
• Fuel Type : Fuel type of the car
• Transmission : Gear transmission of the car
• Color : Color of the car
• Owner : Number of previous owners
• Seller Type : Tells if car is sold by individual or dealer
• Engine : Engine capacity of the car in cc
• Drivetrain : AWD/RWD/FWD
• Length : Length of the car in mm
• Width : Width of the car in mm
• Height : Height of the car in mm
• Seating Capacity : Maximum people that can fir in a car
• Fuel Tank Capacity : Maximum fuel capacity of the car in litres
• TorquePower : Torque power of the car
• TorquePowerRPM : Torque power RPM of the car
• HorsePower : Horse power of the car
• HorsePowerRPM : Horse power RPM of the car

Target variable:

• price: Price of the given car.

Dataset Source Link : https://www.kaggle.com/datasets/nehalbirla/vehicle-dataset-from-cardekho?select=car+details+v4.csv

Screenshot of UI

Screenshot of MLflow

How to run?

STEPS:

Clone the repository

https://github.com/Apheironn/End-to-end-Machine-Learning-Project-with-MLflow

STEP 01- Create a conda environment after opening the repository

conda create -n mlproj python=3.11 -y

conda activate mlproj

STEP 02- install the requirements

pip install -r requirements.txt

# Finally run the following command
python app.py

Now,

# Open up you local host and port
localhost:8080

First go to,

# You have to train your data first on:
localhost:8080/train

Finally,

# Go back to main address and enter values to estimate the price of the vehicle:
localhost:8080/

Project Stages

1. Create a Repository In Github Account:

   • Created a new GitHub repository to host the project, making version control and collaboration easier.

2. Create Structure Using template.py:

   • Organized the project structure using a Template.py file, establishing a clear layout for code and resources.

3. Implementing setup.py:

   • Implemented a setup.py file to define project dependencies and metadata, simplifying package installation and distribution.

4. Logging Implementation:

   • Integrated logging mechanisms throughout the project, allowing effective tracking of code execution and error debugging.

5. Data Ingestion:

   • Developed data ingestion routines to fetch data from various sources (databases, files, APIs) and prepare it for further processing.

6. Data Validation:

   • Implemented data validation steps to ensure the integrity and quality of incoming data, reducing the risk of incorrect inputs.

7. Data Transformation:

   • Applied data transformation techniques, such as feature engineering and preprocessing, to prepare data for model training.

8. Model Trainer:

   • Developed the model training phase, including selecting an appropriate algorithm, tuning hyperparameters, and fitting the model to the training data.

9. Prediction Pipeline:

   • Created a prediction pipeline that takes in new data, applies the necessary transformations, and generates predictions using the trained model.

10. Set MLflow and Docker:

    • Integrated MLflow to track experiments, model versions, and performance metrics, enhancing model management and reproducibility.
    • Set up Docker to containerize the application, ensuring consistency in deployment environments.

11. Deployment In EC2 with App Runner on AWS CI/CD:

    • Utilized AWS EC2 instances for deployment, providing scalable and customizable infrastructure for hosting the application.
    • Employed AWS App Runner to streamline the deployment process, automating application scaling and management.
    • Implemented Continuous Integration/Continuous Deployment (CI/CD) pipelines to automate testing, building, and deploying updates to the application on AWS, ensuring a smoother development and deployment workflow.

dagshub

dagshub

MLflow

• Its Production Grade
• Trace all of your expriements
• Logging & tagging your model

Documentation