[Serving] Add a simple Python serving (#962)

* init simple serving

* simple serving is working

* ppyoloe demo

* Update README_CN.md

* update readme

* complete vision result to json

examples/vision/detection/paddledetection/python/serving/README.md

@@ -0,0 +1 @@
+README_CN.md

examples/vision/detection/paddledetection/python/serving/README_CN.md

@@ -0,0 +1,43 @@
+简体中文 | [English](README_EN.md)
+
+# PaddleDetection Python轻量服务化部署示例
+
+在部署前，需确认以下两个步骤
+
+- 1. 软硬件环境满足要求，参考[FastDeploy环境要求](../../../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)  
+- 2. FastDeploy Python whl包安装，参考[FastDeploy Python安装](../../../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)
+
+服务端：
+```bash
+# 下载部署示例代码
+git clone https://github.com/PaddlePaddle/FastDeploy.git
+cd FastDeploy/examples/vision/detection/paddledetection/python/serving
+
+# 下载PPYOLOE模型文件（如果不下载，代码里会自动从hub下载）
+wget https://bj.bcebos.com/paddlehub/fastdeploy/ppyoloe_crn_l_300e_coco.tgz
+tar xvf ppyoloe_crn_l_300e_coco.tgz
+
+# 安装uvicorn
+pip install uvicorn
+
+# 启动服务，可选择是否使用GPU和TensorRT，可根据uvicorn --help配置IP、端口号等
+# CPU
+MODEL_DIR=ppyoloe_crn_l_300e_coco DEVICE=cpu uvicorn server:app
+# GPU
+MODEL_DIR=ppyoloe_crn_l_300e_coco DEVICE=gpu uvicorn server:app
+# GPU上使用TensorRT （注意：TensorRT推理第一次运行，有序列化模型的操作，有一定耗时，需要耐心等待）
+MODEL_DIR=ppyoloe_crn_l_300e_coco DEVICE=gpu USE_TRT=true uvicorn server:app
+```
+
+客户端：
+```bash
+# 下载部署示例代码
+git clone https://github.com/PaddlePaddle/FastDeploy.git
+cd FastDeploy/examples/vision/detection/paddledetection/python/serving
+
+# 下载测试图片
+wget https://gitee.com/paddlepaddle/PaddleDetection/raw/release/2.4/demo/000000014439.jpg
+
+# 请求服务，获取推理结果（如有必要，请修改脚本中的IP和端口号）
+python client.py
+```

examples/vision/detection/paddledetection/python/serving/README_EN.md

@@ -0,0 +1,44 @@
+English | [简体中文](README_CN.md)
+
+# PaddleDetection Python Simple Serving Demo
+
+
+## Environment
+
+- 1. Prepare environment and install FastDeploy Python whl, refer to [download_prebuilt_libraries](../../../../../../docs/en/build_and_install/download_prebuilt_libraries.md)
+
+Server:
+```bash
+# Download demo code
+git clone https://github.com/PaddlePaddle/FastDeploy.git
+cd FastDeploy/examples/vision/detection/paddledetection/python/serving
+
+# Download PPYOLOE model
+wget https://bj.bcebos.com/paddlehub/fastdeploy/ppyoloe_crn_l_300e_coco.tgz
+tar xvf ppyoloe_crn_l_300e_coco.tgz
+
+# Install uvicorn
+pip install uvicorn
+
+# Launch server, it's configurable to use GPU and TensorRT,
+# and run 'uvicorn --help' to check how to specify IP and port, etc.
+# CPU
+MODEL_DIR=ppyoloe_crn_l_300e_coco DEVICE=cpu uvicorn server:app
+# GPU
+MODEL_DIR=ppyoloe_crn_l_300e_coco DEVICE=gpu uvicorn server:app
+# GPU and TensorRT
+MODEL_DIR=ppyoloe_crn_l_300e_coco DEVICE=gpu USE_TRT=true uvicorn server:app
+```
+
+Client:
+```bash
+# Download demo code
+git clone https://github.com/PaddlePaddle/FastDeploy.git
+cd FastDeploy/examples/vision/detection/paddledetection/python/serving
+
+# Download test image
+wget https://gitee.com/paddlepaddle/PaddleDetection/raw/release/2.4/demo/000000014439.jpg
+
+# Send request and get inference result (Please adapt the IP and port if necessary)
+python client.py
+```

examples/vision/detection/paddledetection/python/serving/client.py

@@ -0,0 +1,28 @@
+import requests
+import json
+import cv2
+import base64
+import fastdeploy as fd
+
+if __name__ == '__main__':
+    url = "http://127.0.0.1:8000/fd/ppyoloe"
+    headers = {"Content-Type": "application/json"}
+
+    im = cv2.imread("000000014439.jpg")
+    data = {
+        "data": {
+            "image": fd.serving.utils.cv2_to_base64(im)
+        },
+        "parameters": {}
+    }
+
+    resp = requests.post(url=url, headers=headers, data=json.dumps(data))
+    if resp.status_code == 200:
+        r_json = json.loads(resp.json()["result"])
+        det_result = fd.vision.utils.json_to_detection(r_json)
+        vis_im = fd.vision.vis_detection(im, det_result, score_threshold=0.5)
+        cv2.imwrite("visualized_result.jpg", vis_im)
+        print("Visualized result save in ./visualized_result.jpg")
+    else:
+        print("Error code:", resp.status_code)
+        print(resp.text)

examples/vision/detection/paddledetection/python/serving/server.py

@@ -0,0 +1,40 @@
+import fastdeploy as fd
+import os
+import logging
+
+logging.getLogger().setLevel(logging.INFO)
+
+# Get arguments from envrionment variables
+model_dir = os.environ.get('MODEL_DIR')
+device = os.environ.get('DEVICE', 'cpu')
+use_trt = os.environ.get('USE_TRT', False)
+
+# Prepare model, download from hub or use local dir
+if model_dir is None:
+    model_dir = fd.download_model(name='ppyoloe_crn_l_300e_coco')
+
+model_file = os.path.join(model_dir, "model.pdmodel")
+params_file = os.path.join(model_dir, "model.pdiparams")
+config_file = os.path.join(model_dir, "infer_cfg.yml")
+
+# Setup runtime option to select hardware, backend, etc.
+option = fd.RuntimeOption()
+if device.lower() == 'gpu':
+    option.use_gpu()
+if use_trt:
+    option.use_trt_backend()
+    option.set_trt_cache_file('ppyoloe.trt')
+
+# Create model instance
+model_instance = fd.vision.detection.PPYOLOE(
+    model_file=model_file,
+    params_file=params_file,
+    config_file=config_file,
+    runtime_option=option)
+
+# Create server, setup REST API
+app = fd.serving.SimpleServer()
+app.register(
+    task_name="fd/ppyoloe",
+    model_handler=fd.serving.handler.VisionModelHandler,
+    predictor=model_instance)