Each of their serverless functions has lambda.js and main.js files. The main file contains the business logic of a serverless function. And the lambda.js file is in charge of wiring the adapters and invoking the main.js file.
XGBoost is a recent, most preferred and powerful gradient boosting method. Instead of making hard Yes and No Decision at the Leaf Nodes, XGBoost assigns positive and negative values to every decision made. All Trees are weak learners and provide a decisions slightly better than a random guess. But collectively averaged out, XGBoost performs really well.
Put simply, Fit a model to the given Training set. Calculate the Residual Errors which become the new training instances. A new model is trained on these and so on. An addition of all the models is selected for making predictions.