56
edits
Changes
→Parallelization Methods
return t
== Parallelization Methods Data Parallelism ==
This section details the ways to parallelize your NN. As image recognition is graphical in nature, multiple GPUs are the best way to parallelize dataset training. === Data Parallelism ===
As image recognition is graphical in nature, multiple GPUs are the best way to parallelize dataset training. <code>DataParallel</code> is a single-machine parallel model, that uses multiple GPUs. It is more convenient than a multi-machine, distributed training model.
You can easily put your model on a GPU by writing:
