Convolutional Neural Networks (CNNs) are designed to map image data (or 2D multi-dimensional data) to an output variable (1 dimensional data). They have proven so effective that they are the ready to use method for any type of prediction problem involving image data as an input.
Convolutional Neural Network is great for data with a spatial relationship
Read this article to understand the intuition behind using Convolutional Neural Network
Have a doubt or thought? Join the discussion now
This is a companion discussion topic for the original entry at http://iq.opengenus.org/when-to-use-convolutional-neural-network-cnn/