Welcome to the FCTL creative inquiry site! This website is a repository of resources for students who want to get involved with research in our lab. In particular, our lab does research in computer architecture, high-performance computing, and machine learning.
We welcome all students who are interested in any of the topics listed above! Here's how to get started:
- To register for the CI course, you must fill out this form and bring it to Dr. Melissa Smith to sign.
- The course number is ECE 1990/2990/3990/4990 (depending on your year), and the section is 018.
- If you're a new student, you may only take the CI for one credit hour. After the first semester you may take the CI for two or three credit hours per semester. If you would like to appeal for more credit hours in your first semester, please contact one of the graduate students.
- Although not absolutely required, it is recommended that you have some experience with Linux and Python before joining the CI. Please consult the skills pages below for an introduction to these topics.
Our creative inquiry meets twice a week (meeting times vary by semester). Refer to the Course Schedule for relevant course information.
For more information, please contact one of the CI mentors:
- Dr. Melissa Smith ([email protected])
Doing research in our lab requires a wide variety of skills. These days you can learn just about anything from the Internet, but here we provide a condensed set of pages that will introduce you to the skills you will need.
- Getting Started
- Datasets
- Git
- High-performance computing
- LaTeX
- Linux
- Machine Learning
- Palmetto Cluster
- Python
- Research
Our primary mode of teaching is through Jupyter notebooks. Here are all of the notebooks we have so far:
- Introduction
- Working with Data
- Supervised Learning
- Unsupervised Learning
- Neural Networks: Dense Layers
- Neural Networks: Convolutional Layers
If you have created an Anaconda environment on Palmetto, you can run these notebooks through Palmetto OnDemand. Go to the Getting Started page for instructions.
You can also run these notebooks very easily using Google Colaboratory! Simply select the "Github" tab, search for the cufctl/mlbd repository, and then you'll be able to select from a list of our notebooks. Once you open a notebook you can save a copy of it to Google Drive. You may also need to go to "Edit" > "Notebook settings" to change the runtime to Python 3 and to use a GPU if you need it. Otherwise, it just works!
This section lists projects from former students. These projects are good learning resources, but they are also available for you to pick up and improve!
- EEG Seizure Prediction
- Effect of Normalization on Tumor Classification
- Emotion Recognition
- Fantasy Football Draft Prediction
- Lung Tissue Classification (PyTorch)
- Object Detection with Mask R-CNN
- Respiratory Disease Classification
- Sentiment Analysis
- Star Classification
- Stock Price Forecasting
- Student Major Prediction
- Tumor Classification
- Uncertainty Quantification
- Word Prediction
Most of these notebooks can be run out of the box with the mlbd Anaconda environment, however some projects may require you to install a few additional packages. Additionally, some of these projects may require data that is not publicly available. If you're having a hard time finding the data, check our Box folder.