snapml

Recognising Handwritten digits with 98.89% accuracy, precision and f1-score

Recognizing handwritten digits with classical machine learning with a 97% accuracy and f1-score

Dectecting chronic heart disease using machine learning 100%

SnapML library's Decision Tree classifier and SVM was used to train a model on a real dataset to identify fraudulent credit card transactions. The Decision Tree model resulted in ROC-AUC score = 0.92 and the SVM yielded ROC-AUC score = 0.93 and hinge loss = 0.15. Multi-threaded CPU was implemented to reduce model training time.

Python scripts exploring machine learning algorithms with libraries like scikit-learn and snapml. Great for beginners!

Alphabet recognition

Detecting the fraudulent credit card transactions by training Decision Tree model using Scikit-learn and SnapML

This repository contains a machine learning project focused on detecting credit card fraud using Decision Tree and Support Vector Machine (SVM) classifiers.

Prediction bankruptcy of Taiwanese bank in the years 1999 to 2009.

A decision tree regressor from SnapML by IBM was used to predict taxi tips on a dataset from the NYC TL Commission. The model was trained on over 3 million data samples in 0.636 seconds using multithreaded CPU/GPU acceleration. Achieved mean squared error = 1.62 on test data.

Scam honeypot + real-time payment guard on IBM LinuxONE. IBM Z Datathon, team Forest.