MNK/portfolio
Machine learning

Cat Dog Classification

CNN regularization comparison

Implemented CNN models with and without pooling and dropout layers, demonstrating regularization techniques.

Problem

Show how pooling and dropout change a CNN's behaviour.

Approach

  1. CNN models with and without pooling and dropout layers
  2. Side-by-side comparison of the regularization effect
Cat Dog Classification workflow: 2 stages from input to result
Workflow

Results

  • Clear comparison of regularization techniques

Stack

  • Python
  • TensorFlow
  • Keras
  • CNN

By Muzammil Nawaz Khan, AI Engineer.