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
I made a clone of this issue here since it seems like he doesn't have access to this repo. The following is what the tt-xla issue states:
Runtime stitching update has been merged here that includes an update to the runtime submit API. Currently the old API is still supported for backwards compatibility but will be deprecated soon, so it would be desirable to update to the new API. To maintain current functionality, the only code change needed would be to change from
runtime.submit(..., output_tensor_containers)
to
output_tensors = runtime.submit(...)
for i, tensor in enumerate(output_tensors):
runtime.memcpy(output_tensor_containers[i], tensor)
An example can also be found under ttrt/common/run.py
This is a clone of this tt-xla issue: tenstorrent/tt-xla#102
I made a clone of this issue here since it seems like he doesn't have access to this repo. The following is what the tt-xla issue states:
FYI: @AleksKnezevic
The text was updated successfully, but these errors were encountered: