From c0a085a12953c70fdd67699b9d1f82c2acf6885d Mon Sep 17 00:00:00 2001
From: Tianlei Wu <tlwu@microsoft.com>
Date: Fri, 26 Jan 2024 10:34:59 -0800
Subject: [PATCH] [CUDA] update python doc for user_compute_stream (#19245)

### Description
Update python doc about user_compute_stream in CUDA python API for
https://github.com/microsoft/onnxruntime/pull/19229.

### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->
---
 docs/execution-providers/CUDA-ExecutionProvider.md | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/docs/execution-providers/CUDA-ExecutionProvider.md b/docs/execution-providers/CUDA-ExecutionProvider.md
index f73df1ebf0909..33861a896b712 100644
--- a/docs/execution-providers/CUDA-ExecutionProvider.md
+++ b/docs/execution-providers/CUDA-ExecutionProvider.md
@@ -61,7 +61,15 @@ Default value: 0
 Defines the compute stream for the inference to run on.
 It implicitly sets the `has_user_compute_stream` option. It cannot be set through `UpdateCUDAProviderOptions`, but rather `UpdateCUDAProviderOptionsWithValue`.
 This cannot be used in combination with an external allocator.
-This can not be set using the python API.
+
+Example python usage:
+```python
+providers = [("CUDAExecutionProvider", {"device_id":torch.cuda.current_device(), "user_compute_stream": str(torch.cuda.current_stream().cuda_stream)})]
+sess_options = ort.SessionOptions()
+sess = ort.InferenceSession("my_model.onnx", sess_options=sess_options, providers=providers)
+```
+
+To take advantage of user compute stream, it is recommended to use [I/O Binding](../api/python/api_summary.html#data-on-device) to bind inputs and outputs to tensors in device.
 
 ### do_copy_in_default_stream
 Whether to do copies in the default stream or use separate streams. The recommended setting is true. If false, there are race conditions and possibly better performance.