Description: The training covers techniques on how to best utilize Dask-ml for scaling machine learning algorithms. Dask is very effective at chunking big data and distributing it for parallel computation. One of the benefits of using Dask is that it is built with python, which allows the user to stay within the python ecosystem; allowing for utilization of any of the other scientific modules. With Dask for Machine Learning (Dask-ml), we can scale the model size and data size, to ensure computations are performed optimally without exceeding CPU and/or memory limitations. During the lecture, we will go through several model building examples; utilizing sckit-Learn, PyTorch, and Keras (tensorflow) with Dask.

Presenter(s): Gedion Teklemariam, GDIT/PET
Location: Webcast
Date & Time: October 5 & 7, 2021, 2:00p – 3:30p ET

Distribution Statement D. Distribution authorized to the Department of Defense and U.S. DoD contractors only, Administrative or Operational Use, 27 September 2021.  Other requests for this document shall be referred to the High Performance Computing Modernization Office, 3909 Halls Ferry Road, Vicksburg, MS 39180.