Skip to content

MartFrancisco/ML-Training

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Machine Learning - Training Projects

This repository contains two short machine learning projects I completed as part of my training.


Glass Classification

This project addresses a classical problem of classifying unbalanced classes using the Glass Identification dataset from the UCI repository.

  • Exploratory Data Analysis (EDA): I began by performing EDA to examine the data distribution and key statistics, as well as to check for duplicates or missing values. I visualized the data using pair plots, violin plots, and correlation matrices for better insight. You can view the EDA here.

  • Modeling: I built and tuned five classification models, comparing their performance before and after tuning using four key metrics: F1-score, recall, accuracy, and precision. You can explore the models and their results here.


QSAR Androgen Receptor

In this project, I built machine learning models to predict whether molecules are active in an androgen receptor using the QSAR Androgen Receptor dataset from the UCI repository. The dataset includes 1,024 molecular fingerprint attributes for 1,687 molecules.

  • Objective: The project focuses on understanding the impact of unbalanced data on model performance. I applied Support Vector Classifier (SVC) and Random Forest models on the original dataset, followed by re-training on a balanced dataset using SMOTE (Synthetic Minority Oversampling Technique).

  • Findings: Balancing the dataset resulted in significant performance improvements across both models. The details of the models and their evaluation can be found here.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published