Intel® DAAL 2017 Bug Fixes List

Published: 08/22/2016  

Last Updated: 08/22/2016

Intel® DAAL 2017  Update 2 (Feb 2017)

DPD200412254, 

DPD200412255

Fixed performance bugs in DAAL integration with Spark MLlib that affected k-means, PCA, and linear regression.

Intel® DAAL 2017  (Sep 2016)

DPD200585378,

DPD200581004,

DPD200577773,

DPD200379770

Numerous bug fixes and usability improvements for daalvars.sh and daalvars.csh scripts.
DPD200584192 Added K-means examples to show how to assign new vector to clusters.
DPD200584011 Fixed problem of CategoricalFeatureDictionary class being not serializable. 
DPD200582124 Fixed EM-GMM algorithm crashes with ill-conditioned covariance matrix.
DPD200581239 Fixed K-means performance problem on MPI clusters.
DPD200580862 Removed OpenMP dependency on Windows*.
DPD200580604 Fixed unhandled exceptions from linear regression when the number of observations is smaller than the number of features.
DPD200580426 Fixed problems of some data management functions being not well documented.
DPD200578344 Fixed performance problems of K-means and correlation matrix on Knights Landing.
DPD200577777 Fixed performance problems of correlation and variance-covariance matrix in the distributed processing mode caused by sequential execution of the primary process.
DPD200577109 Fixed problem of ALS result being unnecessarily transposed. 

 

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.