Engine

Engine is the main force of the Runner. It defines the logic of hardware communication and different deep learning techniques usage like distributed or half-precision training:

Thanks to the Engine design, it’s straightforward to adapt your pipeline for different hardware accelerators. For example, you could easily support:

• PyTorch distribute setup.

• Nvidia-Apex setup.

• AMP distributed setup.

Moreover, we are working on other hardware accelerators support like DeepSpeed, Horovod, or TPU. You could watch the progress of engine development under the Engine API section.

If you haven’t found the answer for your question, feel free to join our slack for the discussion.