Memory and Speed in Black and White Images

Github for project

NOTE to Dr. Zhang and Dat:,


Many problems with current networks are with memory consumption and speed. We will pre-process the images into black and white in order to decrease memory for new networks, by needing fewer scalars, and increase the networks speed, by needing fewer hidden layers to get a prediction faster. More specifically, we will be testing a common architecture, with normal pictures, against a simpler one, with pre-processing done, to see if the pre-processing of the pictures can yield similar accuracy. Then, we will test the memory usage and time for prediction per architecture.




  • Have the selected dataset ready and being used.
  • Have the common CNN obtained, and have the simplified CNN working.

  • Collect and report the data
  • Compare the 2 CNNs.
  • Submit.


Fashion mnist dataset

CIFAR-10 dataset




We will first create a simple python program to turn the CIFAR-10 dataset black and white. Then we will create 2 other python programs to create the normal and simplified CNN architectures that will be used.


  1. 3D Convolutional Neural Network for Brain Tumor Segmentation
  2. Dimensionality reduction, and function approximation of poly(lactic-co-glycolic acid) micro- and nanoparticle dissolution rate


Progress Report

Sample pictures

frog 50 dog 50 car 50 bird 50 frog 100 dog 100 car 100 bird 100 frog 150 dog 150 car 150 bird 150 frog 200 dog 200 car 200 bird 200