CIFAR Challenge!

Instructions

• CIFAR-10 is a small computer vision dataset that was an important benchmark for some of the seminal early work in deep-learning for computer vision.

• Download cifar_challenge.py and cifar_test.py

• If you are working on one of the lab machines, open a terminal and activate the correct Anaconda environment:

  source /cs/shr/cs445/anaconda3/bin/activate
  conda activate gpu

• The provided model has:

  • 3072 input dimensions (32x32 images with three color channels)
  • 16 hidden units
  • 10 output units (one for each class)

  How many total weights will this network have? Once you have calculated your answer by hand, run cifar_challenge.py to check your work.

• Let your the model run for the full 20 epochs then start tensorboard to examine your model:

  tensorboard --logdir <logs path>

  Examine the contents of the “Time Series” tab and the “Graphs” tab.

  Does it look like 20 epochs of training was enough? Too much? Are we continuing to see improvement in the validation loss and accuracy

• Repeat the previous steps for the following two models:

  • Larger three layer model:
    • 3072 inputs
    • 1024 hidden units
    • 10 output units.
  • Deeper model:
    • 3072 inputs
    • 512 hidden units
    • 512 hidden units
    • 512 hidden units
    • 10 output units.

  For each network:

  • Calculate the number of weights (by hand).
  • Run the model for 20 epochs
  • Look at the results in Tensorboard.

• Tune the model to get the best possible testing performance before the end of class. See the header comments in cifar_challenge.py for instructions.