Merge branch 'PaddlePaddle:develop' into develop

doc/Python_Pipeline/Pipeline_Design_CN.md

@@ -700,6 +700,8 @@ Pipeline Serving支持低精度推理，CPU、GPU和TensoRT支持的精度类型
   - fp16
   - int8 
 
+使用int8时，要开启use_calib: True
+
 参考[simple_web_service](../../examples/Pipeline/simple_web_service)示例
 ***
 

doc/Serving_Configure_CN.md

@@ -489,4 +489,7 @@ Python Pipeline支持低精度推理，CPU、GPU和TensoRT支持的精度类型
 #GPU 支持: "fp32"(default), "fp16(TensorRT)", "int8"；
 #CPU 支持: "fp32"(default), "fp16", "bf16"(mkldnn); 不支持: "int8"
 precision: "fp32"
+
+#cablic, open it when using int8
+use_calib: True
 ```

doc/Serving_Configure_EN.md

@@ -495,4 +495,7 @@ Python Pipeline supports low-precision inference. The precision types supported
 #GPU support: "fp32"(default), "fp16(TensorRT)", "int8"；
 #CPU support: "fp32"(default), "fp16", "bf16"(mkldnn); not support: "int8"
 precision: "fp32"
+
+#cablic, open it when using int8
+use_calib: True
 ```