End to End Machine Learning Project with Deployment using Streamlit, FastAPI and AWS EC2

About Project👨‍💻

The Machine learning project aim to predict the strength of concrete structure based on different dependent variables viz., cement content, blast furnance slag, fly ash content, water content, superplasticizer, coarse aggregate content, fine aggregate content, days dried (a total of 8). The independent varibale is the strength of the concrete.

Description of dependent variables:

• Cement content – Amount of cement mixed

• Blast furnance slag – Amount of blast furnance slag mixed

• Fly ash content - Amount of fly ash content mixed

• Water content – Amount of water content added

• Superplasticizer - Amount of superplasticizer content added

• Coarse aggregate content – Amount of Coarse aggregate content

• Fine aggregate content – Amount of Fine aggregate content

• Age – Days to dry the concrete

The work flow for this project is as below:

• Load the dataset with dependent and independent variables

• Perform Exploratory data Analysis with custom summary function

• Perform Outlier Treatment

• Mulivariate Analysis using Regression

• Multicolinearity test and Correlation Analysis

• PCA Technique to deal with Multicolinearity

• Machine Learning Model Building

• Hyperparamter Tuning

• Cross Valiadation Post Hyperparameter tuning

• K-Means clustering for model accuracy check and improvement

• Understanding Feature importance with XGBoost

• Building learning Curve Analysis

• Pickling the best ML model

• Develop streamlit App

• Create FastAPI

• Deploy the App in Amazon web services(AWS) EC2 server

Concrete Strenth Prediction Web App:

AWS EC2 WEB APPLICATION LINK

Fast API

Liked my Contributions:question:Follow Me👉 Nominate Me for GitHub Stars ⭐ ✨

##For any doubts/queries 🔗 👇