PulseFocusPlatform/static/deploy/serving/README.md

3.6 KiB
Raw Blame History

服务端预测部署

PaddleDetection训练出来的模型可以使用Serving 部署在服务端。
本教程以在路标数据集roadsign_voc 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

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

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fetchfetch=["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