You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Super cool. I have an existing pytorch project that has over 100 .to(device) calls. is there an easy way to transform our codebase to incorporate runhouse? or should i manually change all my .to calls to accomodate for runhouse?
cc: @carolineechen@dongreenberg
The text was updated successfully, but these errors were encountered:
Hi Qasim! No, you don't need to modify those calls, because PyTorch modules and tensors should still be sent to devices within your program once the program is sent to remote hardware. The easiest way to get started with Runhouse is to take an existing Python function you'd like to run on remote hardware and sent it over through the rh.function(my_fn).to(my_hw) API. One example might be an inference function, like in this tutorial: https://runhouse-docs.readthedocs-hosted.com/en/latest/tutorials/examples/inference.html
If you'd like to chat in more detail about your code and how to get started I'm happy to chat further.
Super cool. I have an existing pytorch project that has over 100 .to(device) calls. is there an easy way to transform our codebase to incorporate runhouse? or should i manually change all my .to calls to accomodate for runhouse?
cc: @carolineechen @dongreenberg
The text was updated successfully, but these errors were encountered: