Industrial-Copper-Modeling

Introduction

This project leverages machine learning to predict the selling price and status of copper. After cleaning the data, filling missing values, and addressing skewness and outliers, I conducted feature engineering and correlation analysis. I developed a Random Forest regression model for predicting selling prices and an Extra Trees classification model for predicting status. Additionally, I created a Streamlit app that allows users to input data and obtain interactive predictions, displaying the results clearly.

Domain : Manufacturing

Technology and Skills Takeaway

-> Python
-> Numpy
-> Pandas
-> Scikit-Learn
-> Pickle
-> Streamlit
-> Data Preprocessing
-> EDA

Packages and Libraries

!pip install numpy
!pip install pandas
!pip install scikit-learn
!pip install xgboost
!pip install matplotlib
!pip install seaborn
!pip install streamlit

Overview

Data Preprocessing

-> Loaded the copper CSV into a DataFrame.
-> Cleaned and filled missing values, addressed outliers, and adjusted data types.
-> Analyzed data distribution and treated skewness

Feature Engineering

-> Assessed feature correlation to identify potential multicollinearity

Modeling

-> Built a regression model for selling price prediction.
-? Built a classification model for status prediction.
-> Encoded categorical features and optimized hyperparameters.
-> Pickled the trained models for deployment.

Streamlit Application

-> Developed a user interface for interacting with the models.
-> Predicted selling price and status based on user input.

Contact

LINKEDIN: https://www.linkedin.com/in/nithesh-goutham-m-0b0514205/
WEBSITE: https://digital-cv-using-streamlit.onrender.com/
EMAIL : nitheshgoutham2000@gmail.com