This repository contains a set of useful functions for Machine Learning implemented using Julia, and libraries like Flux and Scikit-learn.
Moreover, these functions are used to predict the type of tumor samples into one of five categories based on RNA-seq data; providing a straightforward example of its usage, and additionally serving to introduce core Machine Learning concepts.
- Data normalization
- One-hot encoding
- Artificial Neural Networks with early stopping during training
- Ensemble models, with the possibility to include:
- Support Vector Machines
- Decision Tree Classifiers
- k-Nearest Neighbours Classifiers
- Cross-validation
- Confusion matrices
To use with Jupyter Notebooks be sure to install a Julia kernel. To access the functions include("utils/ML1functions.jl").
The "gene expression cancer RNA-Seq Data Set" is available at https://archive.ics.uci.edu/ml/datasets/gene+expression+cancer+RNA-Seq. Save the data.csv and labels.csv under a folder named dataset.