This SageMaker example showcases how you can create a dataset, dataset group and predictor with Amazon Forecast and SageMaker Pipelines.

sm_pipeline_with_amazon_forecast.ipynb: Notebook explaining the pipeline step-by-step.

preprocess.py: Script used in the ForecastPreProcess step in pipeline for data preparation used for training and evaluation.

train.py: Script used in ForecastTrainAndEvaluate step in pipeline to train and evaluate the Amazon Forecast model.

conditional_delete.py: Script used in ForecastCondtionalDelete step in pipeline to delete all Forecast resources if the score achieved on a particular metric is not satisfactory.

data: data folder containing the train.csv.