CS4981 Deep Learning In Signal Processing

CS4981 Deep Learning In Signal Processing

Week 1 Lab

  1. If you didn’t already have a Rosie account, one has been created for you. You will be prompted to change the default password (see Canvas message or professor) on first login. Follow the instructions at https://msoe.dev/#/access under “Shell Access.” If you don’t have an ssh (secure shell) client on your laptop from a previous class, it is recommended that you install PuTTY as described under “SSH Terminal Program.”
  2. Start MATLAB with the Deep Learning Toolbox on Rosie as described on the course homepage. https://durant.io/courses/cs4981/
  3. Run 3 (or more) “Get Started” Examples from the MATLAB Deep Learning Toolbox. Some will not work (webcam) because of packages that are not available on Rosie.
    • First, try “Classify Image Using Pretrained Network.”
    • Next, try “Create Simple Image Classification Network Using Deep Network Designer”. Take a screen capture of your Training Progress plot to include in your report.
    • Complete at least 1 more example.
  4. Summarize what was done (in your own words, state the goal of each example)
  5. Summarize your key questions or open issues.
    • Where parts of the examples hard to understand?
    • Did you run into errors and were you able to solve them?
    • Are there topics you’d like to learn more about next?

Submit your informal report covering the above items via Canvas as a Word file, PDF, etc.