Setup conda environment

(if necessary) module load anaconda

conda env create -f environment.yaml

Alternative setup

If the conda installation is taking too much time/memory, use manual setup:

conda create -n qmnist python=3.9
conda activate qmnist
conda install pip
pip install -r requirements.txt

Run code

conda activate qmnist

python mnist.py

mnist.py command line arguments:

• --epochs: (int) num. epochs to train for, default is 2
• --noise: (float) std. deviation of noise added to images, default 0 which means no noise
• --model_name: (string) the name of the model to run, default is "QNN":
  • "ClassicalNN" is also supported

Generate visuals

The utils folder contains files to visualize the results. Here is an example generated plot based on our results:

Here is a noise chart generated by util/overlay.py: