Watson Machine Learning Accelerator is an enterprise AI infrastructure to make deep learning and machine learning more accessible–bringing the benefits of AI to your business. This includes the complete life-cycle management from installation and configuration; data ingest and preparation; building, optimizing, and distributing the training model; to moving the model into production. Watson Machine Learning Accelerator is available as an on-premises offering or as a service for IBM Cloud Pak for Data.
In this article, we use Watson Machine Learning Accelerator 1.2.1 and a Jupyter Notebook to provide an overview of the data science experience in Watson Machine Learning Accelerator on-premises. We’ll also show you how increased productivity is possible with a set of robust tooling starting with data ingestion, hyperparameter tuning, model training and inference.
The detailed steps for this article can be found in the associated Jupyter Notebook. Within this notebook, you’ll:
- Upload this notebook to your Watson ML Accelerator environment.
- Download the dataset and model.
- Import the dataset.
- Build the model.
- Tune the hyperparameter.
- Run the training.
- Inspect the training run.
- Create an inference model.
- Test it out.
This article provided an overview of data science experience in Watson Machine Learning Accelerator and how it helps data scientist accelerating time to results and accuracy. To learn more about Watson Machine Learning Accelerator, see Watson Machine Learning Accelerator on IBM Developer.