Win $20,000. Help build the future of education. Answer the call. Learn more

Watson Machine Learning Accelerator videos

Build a movie recommendation system using deep learning

In this video, learn how to use the IBM Watson Machine Learning Accelerator API to accelerate the training of a movie recommendation model using a graphics processing unit (GPU). There’s also an accompanying article. Watson Machine Learning Accelerator is available as a part of the IBM Watson Studio family of AI tools on IBM Cloud Pak for Data. It is a GPU-accelerated deep learning service that lets you share GPU resources across business units to deliver faster training results.

Build a TensorFlow model using Deep Learning Experiment Builder

Learn how to build a TensorFlow model using the MNIST data set for deep learning with Watson Machine Learning Accelerator, which is available as a part of the IBM Watson Studio family of AI tools on IBM Cloud Pak for Data. It is a GPU-accelerated deep learning service that lets you share GPU resources across business units to deliver faster training results.

Overview of Watson Machine Learning Accelerator

Get an overview of IBM Watson Machine Learning Accelerator, a complete AI data science platform for enterprises. Watson Machine Learning Accelerator is available as a part of the IBM Watson Studio family of AI tools on IBM Cloud Pak for Data. It is a GPU-accelerated deep learning service that lets you share GPU resources across business units to deliver faster training results.

Automate hyperparameter optimization training with the Watson Machine Learning Accelerator API

Submit a model and data set to the Watson Machine Learning Accelerator API to run hyperparameter optimization, or HPO. You use the Pytorch MNIST HPO as the training model, inject hyperparameters for the sub-training during search, submit a tuning metric for better results, then query for the best job results.

Introduction to Watson Machine Learning Accelerator for Cloud Pak for Data

Learn about this GPU-accelerated deep learning service that lets you share GPU resources across business units to deliver faster training results.

Balance deep learning jobs with Elastic Distributed Training

See how you can use elastic distributed training to balance IBM Watson® Machine Learning Accelerator jobs in IBM Cloud Pak® for Data. Using Watson Machine Learning Accelerator, GPUs are dynamically allocated to data scientists currently running workloads.

Deliver faster deep learning results on GPUs

Watch how a deep learning workload is run on a CPU, then see how much faster the same deep learning workload runs on a GPU. Using a Cloud Pak for Data 3.5 notebook, you can see that deep learning training is 10 times faster on GPUs.