@@ -86,6 +86,10 @@ class PythonPredictor:
8686 Useful for tasks that the client doesn't need to wait on before
8787 receiving a response such as recording metrics or storing results.
8888
89+ Note: post_predict() and predict() run in the same thread pool. The
90+ size of the thread pool can be increased by updating
91+ `threads_per_process` in the api configuration yaml.
92+
8993 Args:
9094 response (optional): The response as returned by the predict method.
9195 payload (optional): The request payload (see below for the possible
@@ -245,6 +249,10 @@ class TensorFlowPredictor:
245249 Useful for tasks that the client doesn't need to wait on before
246250 receiving a response such as recording metrics or storing results.
247251
252+ Note: post_predict() and predict() run in the same thread pool. The
253+ size of the thread pool can be increased by updating
254+ `threads_per_process` in the api configuration yaml.
255+
248256 Args:
249257 response (optional): The response as returned by the predict method.
250258 payload (optional): The request payload (see below for the possible
@@ -353,6 +361,10 @@ class ONNXPredictor:
353361 Useful for tasks that the client doesn't need to wait on before
354362 receiving a response such as recording metrics or storing results.
355363
364+ Note: post_predict() and predict() run in the same thread pool. The
365+ size of the thread pool can be increased by updating
366+ `threads_per_process` in the api configuration yaml.
367+
356368 Args:
357369 response (optional): The response as returned by the predict method.
358370 payload (optional): The request payload (see below for the possible
0 commit comments