Sonar: Rocks vs. Mines Classification

Dataset from the UCI Berkeley Data Science Repository, available at this link:,+mines+vs.+rocks) .

We train 8 different classification algorithms and compare their performance. Planning to add Stacking and Ensembling of Classification Models as an enhancement. Code available on the link below:

The Gradient Boosting Classifier algorithm did the best so far, at an average of 85% accuracy.  Accuracy will increase on ensembling or stacking the classification models, a possible enhancement.

MNIST in Julia, Python, and Keras

MNIST dataset

MNIST digit recognition and image classification is one of the most common ML sample applications in the world today. Continuing this tradition, this project implements image classification using the MNIST dataset in Python, Julia, and Keras. The source code and datasets are available at the links below:




Note that Python and Julia were fast but Keras was the most accurate at 99.56% accuracy, but complete execution took > 2 hours! Julia had 95% accuracy in 2 minutes. Vectorized Python took 1 minute but had just 88% accuracy.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.