Preparing the test dataset for batch transform inference jobs
In this recipe, we will prepare the test dataset that will be used in the recipe Using batch transform for inference, which makes use of the Batch Transform capability of SageMaker. With Batch Transform, we can perform inference on multiple records all at the same time without having a persistent endpoint running.
Note that when using Batch Transform with a BlazingText model, it is important that the input test dataset is in jsonlines
format. As we have in Figure 8.9, each line in the file is a valid JSON value.
Getting ready
Here are the prerequisites for this recipe:
- This recipe continues from Generating a synthetic dataset for text classification problems.
- A SageMaker Studio notebook running the Python 3 (Data Science) kernel.
How to do it…
The steps in this recipe focus on converting...