cs5830_final

Files of Interest

Because this is a larger-scale project that required resources from multiple sources, files are distributed throughout the repository.

• Pre-Processing Files
  • final_project.ipynb processes the images and stores their data in CSV files
  • img_data_edges.csv and img_data_raw.csv store processed image data for edge-detected and regular images, respectively
• Decision Tree
  • decision_tree.ipynb generates a selection of decision trees, including SVGs for each
  • (regular|edge)_(d3|)_dt.png is an image representation of the generated decision tree
• CNN
  • do_cnn_models.py does the training and testing for all the CNNs.
  • output.txt is the summary of CNN output
  • (cnn)_confusion_matrix.png is the confusion matrix for cnn
  • (cnn)_plot_accuracy.png plots the accuracy over the course of CNN training
  • (cnn)_plot_train.png plots the training loss over the course of CNN training
  • (cnn)_plot_validation.png plots the validation loss over the course of CNN training