Welcome to the repository for my Machine Learning Certificate projects. This repository contains all the code, reports, datasets, and supplementary materials related to the projects completed as part of the certification.
The Machine Learning Certificate is designed to provide a comprehensive understanding of machine learning and its practical applications. It covers a range of topics including supervised and unsupervised learning, model evaluation, and deployment strategies.
-
Project 1: [Housing prices analysis]
- Description: In this project we explore, clean and apply feature engineering to the housing prices dataset. Once cleaned and preprocessed a discussion of the findings and insights is presented. Then a hypothesis is made and evaluated by the Pearson correlation. Finally suggestions are added and some comments on the quality of the dataset used.
- Code: Housing prices analysis code
- Report: Housing prices analysis report
-
Project 2: [Fat prediction with regression]
- Description: In this report we utilize the ‘Extended Body Fat Dataset’ to predict the body fat using different regression models. Before performing any regression model we explore the data in order to clean it and apply feature engineering techniques that allows for a better performance. Finally we compare different regression models with unseen data.
- Code: Fat prediction with regression code
- Report: Fat prediction with regression report
-
Project 3: [Diabetes classification]
- Description: In this model we utilize
- Code: Diabetes classification code
- Report: Diabetes classification report
-
Project 4: [Anime Recommendation Clustering]
- Description: In this project, I leverage the `Anime Recommendation Database 2020' from Kaggle to develop a recommendation model using unsupervised clustering algorithms. I employ three distinct models and compare their performance. Prior to modeling, the data undergoes cleaning and transformation to enhance the system's effectiveness.
- Code: Anime Recommendation Clustering
- Report: Anime Recommendation Report
-
Project 5: [Hand-Sign Recognition with CNN]
- Description: In this project, I utilize pytorch to create convolutional neural networks to classify hand-sign alphabet (excluding 'j' and 'z' which are movements). I utilize different regulation techniques such as dropout and data augmentation. Finally I compare all models and performance, mainly accuracy and loss.
- Code: Hand-Sign Recognition code
- Report: Unavailable
- /Learning Codes: Contains codes with purpose of understanding some ML models.
- /Project: Contains the datasets, reports and codes of each project.
To explore the projects and materials included in this repository, simply navigate through the folders corresponding to each project. Each project folder contains the necessary code and report files for review.
My name is Gilberto Juárez, glad to meet you. I'm an applied physics engineer with passion in Data Science, Machine Learning and Quantum Computing!!! Here is my linkedin profile if interested to get in contact: Gilberto Juárez Rangel