https://github.com/nitheshgoutham/industrial-copper-modeling
To Predict the selling price and status
https://github.com/nitheshgoutham/industrial-copper-modeling
data-science data-visualization machine-learning python3 sql streamlit
Last synced: about 2 months ago
JSON representation
To Predict the selling price and status
- Host: GitHub
- URL: https://github.com/nitheshgoutham/industrial-copper-modeling
- Owner: NitheshGoutham
- Created: 2024-07-20T20:05:17.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-07-20T20:24:37.000Z (10 months ago)
- Last Synced: 2025-01-26T11:08:15.800Z (4 months ago)
- Topics: data-science, data-visualization, machine-learning, python3, sql, streamlit
- Language: Jupyter Notebook
- Homepage:
- Size: 9.37 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 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 : [email protected]