diff --git a/deploy/README.md b/deploy/README.md index 09654788f6b1037d708e7843ec14e426573ec4a6..137afa02af8b853528f8e62c646add348c766f41 100644 --- a/deploy/README.md +++ b/deploy/README.md @@ -1,6 +1,9 @@ # PaddleDetection 预测部署 -`PaddleDetection`目前支持使用`Python`和`C++`部署在`Windows` 和`Linux` 上运行。 +`PaddleDetection`目前支持： +- 使用`Python`和`C++`部署在`Windows` 和`Linux` 上运行 +- [在线服务化部署](./serving/README.md) +- [移动端部署](https://github.com/PaddlePaddle/Paddle-Lite-Demo) ## 模型导出 训练得到一个满足要求的模型后，如果想要将该模型接入到C++服务器端预测库或移动端预测库，需要通过`tools/export_model.py`导出该模型。 @@ -20,4 +23,5 @@ yolov3_darknet # 模型目录 ## 预测部署 - [1. Python预测(支持 Linux 和 Windows)](https://github.com/PaddlePaddle/PaddleDetection/blob/master/deploy/python) - [2. C++预测(支持 Linux 和 Windows)](https://github.com/PaddlePaddle/PaddleDetection/blob/master/deploy/cpp) -- [3. 移动端部署参考Paddle-Lite文档](https://paddle-lite.readthedocs.io/zh/latest/) +- [3. 在线服务化部署](./serving/README.md) +- [4. 移动端部署](https://github.com/PaddlePaddle/Paddle-Lite-Demo) diff --git a/deploy/serving/README.md b/deploy/serving/README.md new file mode 100644 index 0000000000000000000000000000000000000000..48fc2f03d687328aa1f8d26116ca8165ab202e98 --- /dev/null +++ b/deploy/serving/README.md @@ -0,0 +1,106 @@ +# 服务端预测部署 + +`PaddleDetection`训练出来的模型可以使用[Serving](https://github.com/PaddlePaddle/Serving) 部署在服务端。 +本教程以在路标数据集[roadsign_voc](https://paddlemodels.bj.bcebos.com/object_detection/roadsign_voc.tar) 使用`configs/yolov3_mobilenet_v1_roadsign.yml`算法训练的模型进行部署。 +预训练模型权重文件为[yolov3_mobilenet_v1_roadsign.pdparams](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1_roadsign.pdparams) 。 + +## 1. 首先验证模型 +``` +python tools/infer.py -c configs/yolov3_mobilenet_v1_roadsign.yml -o use_gpu=true weights=https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1_roadsign.pdparams --infer_img=demo/road554.png +``` + +## 2. 安装 paddle serving +``` +# 安装 paddle-serving-client +pip install paddle-serving-client -i https://mirror.baidu.com/pypi/simple + +# 安装 paddle-serving-server +pip install paddle-serving-server -i https://mirror.baidu.com/pypi/simple + +# 安装 paddle-serving-server-gpu +pip install paddle-serving-server-gpu -i https://mirror.baidu.com/pypi/simple +``` + +## 3. 导出模型 +PaddleDetection在训练过程包括网络的前向和优化器相关参数，而在部署过程中，我们只需要前向参数，具体参考:[导出模型](https://github.com/PaddlePaddle/PaddleDetection/blob/master/docs/advanced_tutorials/deploy/EXPORT_MODEL.md) + +``` +python tools/export_serving_model.py -c configs/yolov3_mobilenet_v1_roadsign.yml -o use_gpu=true weights=https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1_roadsign.pdparams --output_dir=./inference_model +``` + +以上命令会在./inference_model文件夹下生成一个`yolov3_mobilenet_v1_roadsign`文件夹： +``` +inference_model +│ ├── yolov3_mobilenet_v1_roadsign +│ │ ├── infer_cfg.yml +│ │ ├── serving_client +│ │ │ ├── serving_client_conf.prototxt +│ │ │ ├── serving_client_conf.stream.prototxt +│ │ ├── serving_server +│ │ │ ├── conv1_bn_mean +│ │ │ ├── conv1_bn_offset +│ │ │ ├── conv1_bn_scale +│ │ │ ├── ... +``` + +`serving_client`文件夹下`serving_client_conf.prototxt`详细说明了模型输入输出信息 +`serving_client_conf.prototxt`文件内容为： +``` +feed_var { + name: "image" + alias_name: "image" + is_lod_tensor: false + feed_type: 1 + shape: 3 + shape: 608 + shape: 608 +} +feed_var { + name: "im_size" + alias_name: "im_size" + is_lod_tensor: false + feed_type: 2 + shape: 2 +} +fetch_var { + name: "multiclass_nms_0.tmp_0" + alias_name: "multiclass_nms_0.tmp_0" + is_lod_tensor: true + fetch_type: 1 + shape: -1 +} +``` + +## 4. 启动PaddleServing服务 + +``` +cd inference_model/yolov3_mobilenet_v1_roadsign/ + +# GPU +python -m paddle_serving_server_gpu.serve --model serving_server --port 9393 --gpu_ids 0 + +# CPU +python -m paddle_serving_server.serve --model serving_server --port 9393 +``` + +## 5. 测试部署的服务 +准备`label_list.txt`文件 +``` +# 进入到导出模型文件夹 +cd inference_model/yolov3_mobilenet_v1_roadsign/ + +# 将数据集对应的label_list.txt文件拷贝到当前文件夹下 +cp ../../dataset/roadsign_voc/label_list.txt . +``` + +设置`prototxt`文件路径为`serving_client/serving_client_conf.prototxt` 。 +设置`fetch`为`fetch=["multiclass_nms_0.tmp_0"])` + +测试 +``` +# 进入目录 +cd inference_model/yolov3_mobilenet_v1_roadsign/ + +# 测试代码 test_client.py 会自动创建output文件夹，并在output下生成`bbox.json`和`road554.png`两个文件 +python ../../deploy/serving/test_client.py ../../demo/road554.png +``` diff --git a/deploy/serving/test_client.py b/deploy/serving/test_client.py new file mode 100644 index 0000000000000000000000000000000000000000..7c2a6395f1ef34c9ca5a7bea4d23c47e9b7a63b9 --- /dev/null +++ b/deploy/serving/test_client.py @@ -0,0 +1,40 @@ +# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +import sys +import numpy as np +from paddle_serving_client import Client +from paddle_serving_app.reader import * +import cv2 +preprocess = Sequential([ + File2Image(), BGR2RGB(), Resize( + (608, 608), interpolation=cv2.INTER_LINEAR), Div(255.0), Transpose( + (2, 0, 1)) +]) + +postprocess = RCNNPostprocess("label_list.txt", "output", [608, 608]) +client = Client() + +client.load_client_config("serving_client/serving_client_conf.prototxt") +client.connect(['127.0.0.1:9393']) + +im = preprocess(sys.argv[1]) +fetch_map = client.predict( + feed={ + "image": im, + "im_size": np.array(list(im.shape[1:])), + }, + fetch=["multiclass_nms_0.tmp_0"]) +fetch_map["image"] = sys.argv[1] +postprocess(fetch_map)