From b631ff0a240e7a1c317cadae75a07ccd7671dca3 Mon Sep 17 00:00:00 2001 From: ShiningZhang Date: Mon, 11 Oct 2021 17:19:17 +0800 Subject: [PATCH 1/2] fix: use_calib is not configured properly --- python/paddle_serving_server/server.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/python/paddle_serving_server/server.py b/python/paddle_serving_server/server.py index 115488b00..f21e13aaf 100755 --- a/python/paddle_serving_server/server.py +++ b/python/paddle_serving_server/server.py @@ -564,7 +564,7 @@ def run_server(self): "-num_threads {} " \ "-port {} " \ "-precision {} " \ - "-use_calib {} " \ + "-use_calib={} " \ "-reload_interval_s {} " \ "-resource_path {} " \ "-resource_file {} " \ From 072114e71856fa303077712196c46727afa6c276 Mon Sep 17 00:00:00 2001 From: ShiningZhang Date: Mon, 11 Oct 2021 17:39:21 +0800 Subject: [PATCH 2/2] update description of use_calib in README --- README.md | 2 +- README_CN.md | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 6c6d0924b..34bc15495 100644 --- a/README.md +++ b/README.md @@ -188,7 +188,7 @@ python3 -m paddle_serving_server.serve --model uci_housing_model --thread 10 --p | `use_lite` (Only for Intel x86 CPU or ARM CPU) | - | - | Run PaddleLite inference | | `use_xpu` | - | - | Run PaddleLite inference with Baidu Kunlun XPU | | `precision` | str | FP32 | Precision Mode, support FP32, FP16, INT8 | -| `use_calib` | bool | False | Only for deployment with TensorRT | +| `use_calib` | bool | False | Use TRT int8 calibration | | `gpu_multi_stream` | bool | False | EnableGpuMultiStream to get larger QPS | #### Description of asynchronous model diff --git a/README_CN.md b/README_CN.md index a1bb9f9e7..d4bcc1a20 100644 --- a/README_CN.md +++ b/README_CN.md @@ -187,7 +187,7 @@ python3 -m paddle_serving_server.serve --model uci_housing_model --thread 10 --p | `use_lite` (Only for Intel x86 CPU or ARM CPU) | - | - | Run PaddleLite inference | | `use_xpu` | - | - | Run PaddleLite inference with Baidu Kunlun XPU | | `precision` | str | FP32 | Precision Mode, support FP32, FP16, INT8 | -| `use_calib` | bool | False | Only for deployment with TensorRT | +| `use_calib` | bool | False | Use TRT int8 calibration | | `gpu_multi_stream` | bool | False | EnableGpuMultiStream to get larger QPS | #### 异步模型的说明