Skip to content

Commit e446860

Browse files
committed
Add documentation for data parallel mode
1 parent f8a022d commit e446860

File tree

1 file changed

+3
-1
lines changed

1 file changed

+3
-1
lines changed

inferentia/README.md

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -111,7 +111,9 @@ INPUT__1 refers to the second input, etc.
111111

112112
Additionally, `--neuron-core-range` specifies the neuron cores to
113113
be used while serving this models. Currently, only
114-
`torch.neuron.DataParallel()` mode is supported.
114+
`torch.neuron.DataParallel()` mode is supported. See
115+
[Data Parallel Inference](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/neuron-guide/appnotes/perf/torch-neuron-dataparallel-app-note.html)
116+
for more information.
115117

116118
The invocation should create a triton model directory with following
117119
structutre:

0 commit comments

Comments
 (0)