Pre-Notebook with GPU

GPU Workspaces

Part 8 here shows you how to accelerate network computations using a GPU. So, the next workspace is GPU-enabled, which means you can select to train on a GPU instance. Since you are limited on the GPU hours you have, the recommendation is this:

Work in CPU mode while developing your models and such.
Make sure the network is learning (the training loss is dropping) using just the CPU.
When you're ready to train for real and optimize the hyperparameters, enable the GPU

All models and data they see as input will have to be moved to the GPU device, so take note of the relevant movement code in the model creation and training process.

Next Concept