Home » FAQs

FAQs

The best place to get started with Habana is our Developer Website.  Specifically, the Getting Started page provides information on how to migrate models to the Habana Architecture, watch videos and get linked to our detailed documentation.

From the Getting Started page on our Developer Website, refer to the TensorFlow Migration Guide or PyTorch Migration guide. The migration guide will walk you through the steps to ensure the model is functional on the Gaudi HPU.

For general issues and questions, you should post your questions to the Habana Forum. For Specific questions related to using Habana’s Reference TensorFlow or PyTorch models on our GitHub page, you can post specific questions in the Issues section of the Model References GitHub page.

You can start at Setup and Install page for more information on how to setup your environment. This will cover the following: how to download and run Habana specific Docker Images, how to create custom Docker images based on the dockerfiles provided, and bare metal driver installation. Additionally, you can refer to the Habana Developer Website for the latest Drivers to download.

Users can start at our Setup and Install GitHub page.  There are sections to show how do download and run Habana specific Docker Images and create custom Docker images based on the dockerfiles provided

End users can refer to the model roadmap on our GitHub page – https://github.com/HabanaAI/synapseai-roadmap

If you still have questions, post your questions on the Habana Developer Forum. For Specific questions related to using Habana’s Reference TensorFlow or PyTorch models on our GitHub page, you can post specific questions in the Issues section of the Model References GitHub page.

Sign up for the latest Habana developer news, events, training, and updates.