AOCL-Data Analytics

The AOCL-Data Analytics Library (AOCL-DA) provides optimized building blocks for data analysis and classical machine learning. APIs are available in C, C++, and Python, with a patch for scikit-learn that helps you leverage AOCL-DA performance with minimal code changes. The key features include:

• Loading data from CSV files
• Preprocessing data by standardizing, removing missing values, and extracting subsets
• Processing data using a variety of algorithms including linear models, clustering methods, decompositions, decision forest, and nearest neighbor algorithms

Highlights of Release 5.3

• Python wheels are now also available on PyPI for easy installation via pip
• Trained models can now be saved in binary and loaded later for inference
• New APIs for approximate nearest neighbors, radius neighbors classification and radius neighbors regression
• Delivered performance improvements to support vector machines, DBSCAN, k-means clustering, and train_test_split

Documentation

• Source code: GitHub
  
• Prior versions: AOCL-Data Analytics Archive
• AOCL-DA API Guide

Downloads