未验证 提交 01cb5057 编写于 作者: T TeslaZhao 提交者: GitHub

Update Serving_Configure_EN.md

上级 26e0796e
......@@ -397,6 +397,9 @@ op:
#ir_optim, When running on TensorRT,must set ir_optim=True
ir_optim: True
#CPU 计算线程数,在CPU场景开启会降低单次请求响应时长
#thread_num: 10
#precsion, Decrease accuracy can increase speed
#GPU 支持: "fp32"(default), "fp16", "int8";
#CPU 支持: "fp32"(default), "fp16", "bf16"(mkldnn); 不支持: "int8"
......@@ -435,6 +438,9 @@ op:
#ir_optim, When running on TensorRT,must set ir_optim=True
ir_optim: True
#CPU 计算线程数,在CPU场景开启会降低单次请求响应时长
#thread_num: 10
#precsion, Decrease accuracy can increase speed
#GPU 支持: "fp32"(default), "fp16", "int8";
#CPU 支持: "fp32"(default), "fp16", "bf16"(mkldnn); 不支持: "int8"
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册