Solving Deep Learning Data Operations Challenges

Contributed Talk | Day 2 | 10:25 am | 40 Minute Duration | Grand Gallery C
  • Jim Scott
    MapR Technologies Global Architect & Engineering Leader

Solving Deep Learning Data Operations Challenges

Contributed Talk | Day 2 | 10:25 am | 40 Minute Duration | Grand Gallery C

By working with a variety of my customers across many industries — chemical sciences, health care, oil and gas, online media — I have documented a number of problems which are major impediments to successful completion of projects in the AI/ML space. All of my examples, explanations, and solutions presented are a direct result of these customer interactions. The core demo/example shows 2 input files that quickly grew to over 75 with 9 different data formats, including CSV, HDF5, PKL, and Image on GPU/CPU.