Machine Learning Classification

Machine Learning classification model implementing Random Forest algorithm, built in Python from scratch. Tested using the IRIS flower dataset. I developed this project to enhance my understanding of ML/AI technologies, and am working on implementing it into tools used for real-time game engine filmmaking.

Model generates decision trees, and saves them into a Random Forest Object. The object is tested for accuracy and scrapped if does not meet the minimum accuracy requirement parameter. Model is trained off of 60% of the datset, and tested using the other 40% it has never been exposed to.

Feel free to clone this project and use it to generate an ML classification model for a problem you care about! Follow the instructions within the Python files for how datasets should be formatted.

Example 1: ML Model with 95.08% Accuracy

The first plot represents the actual species of certain flowers in the dataset, while the second is the Machine Learning predicted species. The model predicted species correctly 58 times and incorrectly 3 times in this example.

Example 2: ML Model with 96.72% Accuracy

The model predicted species correctly 59 times and incorrectly 2 times in this example.