Skip to content

This project focuses on modelling industrial copper data using Python and various libraries such as pandas, numpy, scikit-learn

Notifications You must be signed in to change notification settings

SRIDHAR3131/Industry-copper-modelling

Repository files navigation

Industrial Copper Modelling

This project focuses on modelling industrial copper data using Python and various libraries such as pandas, numpy, scikit-learn. The objective of the project is to preprocess the data, handle missing values, detect outliers, and handle skewness. Additionally, regression and classification models will be built to predict the selling price and determine if a sale was won or lost. The trained models will be saved as a pickle file for later use in a Streamlit application.

HOMEPAGE

Screenshot 2023-06-20 012035

PRICE PREDICTION

Screenshot 2023-06-20 010417

PREDICT STATUS

Screenshot 2023-06-20 013726

Project Structure

The project consists of the following files and directories: data/: Directory containing the raw data file(s) for the industrial copper dataset. modeling.ipynb: Jupyter Notebook file containing the code for building regression and classification models. pickle_file_generator.ipynb: Jupyter Notebook file for generating the pickle file of the trained model. streamlit_app.py: Python file containing the Streamlit application code for using the trained model to predict selling price and status.

Prerequisites

Before running the code, ensure that you have the following dependencies installed:

  • Python (version 3.7 or above)
  • Pandas
  • Numpy
  • Scikit-learn
  • Streamlit

In this project, we have successfully developed a pipeline to preprocess industrial copper data, handle missing values, detect outliers, and address skewness. We have also built regression and classification models to predict the selling price and determine if a sale was won or lost. The trained models have been saved as a pickle file for easy retrieval. The Streamlit application provides a user-friendly interface to utilize the trained model for making predictions.

About

This project focuses on modelling industrial copper data using Python and various libraries such as pandas, numpy, scikit-learn

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published