Comparing Discriminative and Generative Machine Learning Models for Image Classification
- 0 Collaborators
Exploratory project comparing various machine learning algorithms including but not limited to support vector machines (SVMs), Decision Trees (DTs), Convolutional Neural Networks (CNNs), etc. The goal to is compare the performance of each image classifier with metrics like AUROC. ...learn more
Project status: Under Development
oneAPI, Artificial Intelligence
Intel Technologies
oneAPI
Overview / Usage
This project involves the comparison of discriminative and generative machine learning models. This involves comparison of various machine learning models and optimization techniques to compare and contrast their performance on image classification. The problem to be solved is understanding the costs and benefits of using one type of model over another for various use cases in image classification.
Methodology / Approach
Developing standard image data processing steps including data segmentation, augmentation, annotations, etc. Then building various machine learning models of varying hyperparameters and specifications to compare and contrast optimized machine learning models. This project will use OneAPI technology as well as python programming to test various machine learning model implementations. Performance of ML models will be compared using metrics such as Area Under the Receiver Operator Curve, Accuracy, etc.