AutoTrain - Part 2: Recipes
As you can see, we are using config.header.objects in a number of places where one is expected to give inputs and store outputs. The rest of the code is largely the same, i.e., we can use the same set of files to train any number of classification tasks without worrying about creating duplicate code bases for each project. Let's go through [ingredients of Classification, Object Detection and Segmentation experiments]({% post_url 2021-10-09-auto-train-use-cases %}) to understand how easy it is to train with a good fixed base.