This project focused on developing a high-accuracy machine learning model for lithology classification using a publicly available well log dataset. The workflow included exploratory data analysis (EDA) to assess data quality and understand the underlying characteristics of the variables, followed by extensive data visualization that integrated my petroleum engineering expertise with modern data science tools such as Matplotlib and Seaborn. Multiple multiclass classification models were developed, trained, and evaluated using data from a single well. The best-performing model was then selected and validated on unseen data to assess its generalization capability and predictive performance.

Workflow

Code - Jupyter Notebook

Github: https://github.com/alfonsovecino/well-logs-for-lithology-classification

Overview

Release date: June, 2026

Contact

My social media.

© 2025. All rights reserved. - Developed by Alfonso Vecino. alfonso_uis@hotmail.com