My take on Data Science Interview Questions [ Part 1 ]


My current understanding of feature vectors is very limited. Lets say we are building a classifier using standard Convolutional Neural Network (CNN) and Fully-connected Neural Network (FNN). Where CNN is responsible of extracting high level features from an image (starting from edges and corners to faces etc). In the transition from the CNN to FNN we would most likely vectorize the images, this is where I would call that vector a feature vector. I was surprised by the fact that even raw pixel values can be thought of as a feature vector.

Duplicate Docs Excel Report

None found

Similar Docs  Excel Report  more

None found