Machine learning models are commonly used today in many applications to drive insights and to make predictions. However, many high performing models such as random forests and neural networks are black boxes and lose explainability where users are not able to fully understand the reasons behind a model's predictions! This …

Smile detection applications are countless from activating a camera shutter automatically, enhancing human-robot interaction, to being incorporated into assistive communication devices for people with disabilities. Using deep learning convolutional neural network algorithms, a smile detection algorithm with an accuracy of 89% was achieved!

As a project I created Dad Jokes …

A Convolutional Neural Network (CNN) is an algorithm which can take in an image, assign importance (learnable weights and biases) to various aspects/features/filters in the image and be able to differentiate one from the other.

The architecture of a CNN can be thought to be similar to that …

In machine learning, working with unbalanced classes can be challenging. Unbalanced classes occurs when there's an unequal representation of classes, where the minority class is often of higher interest to us. This is prominent in areas where data collection of the minority classes is difficult due to time and cost …

It's been three weeks into my Data Science immersive bootcamp and I'm having a blast coding again! I'm starting to get used to Python and new libraries (pandas, numpy, altair, etc.). If you're like me and you've previously programmed in other languages like C/Java, when learning any new language …