Model Repository

Model Repository is available out-of-the-box when you subclass from h1.Model API e.g. model.persist() and model.load(). H1st knows how to persist/load Scikit-Learn and Keras models. A simple ML model example with persist/load is available in examples/HelloWorld/ml_model.py.

All you need to do is to set Please set MODEL_REPO_PATH in the project's config.py or calling h1.init(MODEL_REPO_PATH="s3://..."). Model Repository supports local disk or S3 storage.