docs(distributed.functional): add return type for all_reduce_max (jira #MGE-2706)

GitOrigin-RevId: a29f1f1880674a45fa593582b34b807be1e46786

docs(distributed.functional): add return type for all_reduce_max (jira #MGE-2706)
GitOrigin-RevId: a29f1f1880674a45fa593582b34b807be1e46786
551cc701 · Megvii Engine Team · huangxinda · 72ff7aec · 551cc701
隐藏空白更改
内联并排

Showing with 1 addition and 0 deletion

imperative/python/megengine/distributed/functional.py imperative/python/megengine/distributed/functional.py +1 -0

未找到文件。
--- a/imperative/python/megengine/distributed/functional.py
+++ b/imperative/python/megengine/distributed/functional.py
@@ -271,6 +271,7 @@ def all_reduce_max(
    :param inp: input tensor.
    :param group: communication group.
    :param device: execution device.
+    :returns: reduced tensor. 
    """
    mode = CollectiveComm.Mode.ALL_REDUCE_MAX
    return collective_comm(inp, mode, group, device)