From the course: Amazon Web Services Machine Learning Essential Training
Unlock the full course today
Join today to access over 22,400 courses taught by industry experts or purchase this course individually.
Work with the AWS Deep Learning AMI - Amazon Web Services (AWS) Tutorial
From the course: Amazon Web Services Machine Learning Essential Training
Work with the AWS Deep Learning AMI
- So here we are working with the familiar digit classification problem. So, let's start this off. We're loading MXNet and we're setting up the batch sizes. And we're using the same Multilayer Perceptron to solve the problem. Flattening the images. Now we're defining the deep neural network. Defining the layers. This is the same reproach that we saw in earlier movies running on SageMaker. Continuing to build. This is a drawing. It's a nice drawing, actually, of what we've built. The input layers and the hidden layers. And now we'll do the training. Notice we're training on CPU. Now we'll go to the prediction. And continue on... And we get a 96 percent, almost 97 percent accuracy. Again, we're going to use a different method. We're going to use a Convolutional Neural Network. Define the layers. Nice picture of what we're doing here. Although, this says we're training on GPU, we're actually training on CPU. So let's train on CPU first. And you can see here the batch is running. It's…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
Understanding ML virtual servers4m 7s
-
Understanding deep learning2m 36s
-
Work with Gluon for MXNet in SageMaker5m 14s
-
Work with MXNet in SageMaker9m 1s
-
Databricks on AWS7m 2s
-
Work with MXNet in Databricks9m 2s
-
Set up the AWS Deep Learning AMIs6m 38s
-
Work with the AWS Deep Learning AMI4m 16s
-
Work with EMR for machine learning8m 40s
-
-
-