Karan Batta from Microsoft
In this video, you’ll learn how to start submitting deep neural network (DNN) training jobs in Azure by using Azure Batch to schedule the jobs to your GPU compute clusters.
“Previously, few people had access to the computing power for these scenarios. With Azure Batch, that power is available to you when you need it. Azure Batch AI Training, a new addition to Azure Batch that will target data scientists and AI researchers, as well as those who train and test deep learning and other AI models, will soon be available. Users will specify their job requirements and the path to model and data, and Batch will take care of provisioning a cluster and running the jobs. Batch AI Training will support any learning framework on Linux or Windows including CNTK, TensorFlow, Chainer, Caffe, and others. The service will make it easy to run hyper-parameter sweeps to optimize models, scaling up and down a GPU cluster as needed. Batch AI Training will provide a Python API to use with scripts and Notebooks, and an SDK for automation.”
