Copyright (C) 2024, Advanced Micro Devices. All rights reserved.

AOCL DA - Release Notes - AOCL version 5.0.0
----------------------------------------------

AOCL DA Contents
--------------------
AOCL-DA is a new data analytics library providing optimized building blocks for data analysis and classical machine learning. 

The following key functionality is available in AOCL-DA: 

  - Loading data from memory by reading CSV files 
  - Preprocessing data by removing missing values, standardizing and selecting subsets of data 
  - Processing data (using, e.g. principal component analysis, linear models etc.) 

AOCL-DA can be called in the following ways: 

  - Using the C-compatible interface to enable seamless integration with multiple programming languages 
  - Using C++ overloads for increased data abstraction 
  - Using the Python API 
  - Using the scikit-learn patch to enable existing scikit-learn users to benefit from AOCL-DAs performance with minimal code changes 

The following data processing algorithms are available at 5.0: 
  - Linear models, including logistic regression, lasso and ridge regression 
  - k-means clustering 
  - principal component analysis and singular value decomposition 
  - k-nearest neighbors 
  - decision trees and random forests 
  - nonlinear least squares fitting 
  - basic statistics
