Creating a model using Amazon Lookout for Vision
In this section, we will learn how to choose the type of model you need for your use case, how to create training and test datasets, start model training, improve your model, and analyze images using the trained model.
Choosing the model type based on your business goals
Similar to Rekognition Custom Labels, Lookout for Vision provides a couple of choices in choosing the model type: image classification and image segmentation:
- The image classification model: This type of model predicts whether the image contains an anomaly or not. It will not provide other information, such as the location of the anomaly or the type of anomaly. An example of this is if you want to know whether a capsule shell is mashed or cracked on an assembly line. An image classification model would be a good choice for this use case. As capsule shells are relatively cheap, you may not want to investigate the location of mashing or cracks.
- The image...