Created by: suixf0415

• 版本、环境信息 1）PaddleHub和PaddlePaddle版本：paddlehub 1.8.0 paddlepaddle-gpu 1.8.2.post107 2）系统环境：请您描述系统类型，例如Linux/Windows/MacOS/，python版本 NAME="Ubuntu" VERSION="19.10 (Eoan Ermine)" Python 3.6.10 :: Anaconda, Inc.

• 问题描述

1. 通过paddleX训练图像分类模型water_recog
2. 通过hub convert命令将模型转换为hubserving模型 hub convert --model_dir ./inference_model/ --module_name water_recog --module_version 1.0 --output_dir hubserving_model
3. 通过hub install命令安装转化好的模型 hub install ./hubserving_model/water_recog.tar.gz
4. 通过hub serving命令启动服务 hub serving start -m water_recog --port 5006 ***以上4部执行过程正确没有任何错误提示。
5. 通过client代码访问服务，python serving_client.py -i test2.jpg -p 5006 -n water_recog. 结果服务端提示:Incorrect padding,

######服务端错误信息如下:

suixf@suixf-MS-7B61 ~/p/P0006-T0007_export_model> hub serving start -m water_recog --port 5006                                              (venvpy36) 
water_recog == 1.0
 * Serving Flask app "paddlehub.serving.app_single" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: off
2020-07-31 19:00:14,292-INFO:  * Running on http://0.0.0.0:5006/ (Press CTRL+C to quit)
2020-07-31 19:02:46  -  Incorrect padding
2020-07-31 19:02:46,886-INFO: 127.0.0.1 - - [31/Jul/2020 19:02:46] "POST /predict/water_recog HTTP/1.1" 200 -
2020-07-31 19:05:33  -  Incorrect padding
2020-07-31 19:05:33,303-INFO: 127.0.0.1 - - [31/Jul/2020 19:05:33] "POST /predict/water_recog HTTP/1.1" 200 -

######client代码如下:

import requests
import json
import cv2
import base64
import json
import argparse
import glob

def cv2_to_base64(image):
    data = cv2.imencode('.jpg', image)[1]
    return base64.b64encode(data.tostring()).decode('utf8')
if __name__ == '__main__':
    ap = argparse.ArgumentParser()
    ap.add_argument("-i", "--image_path", required=True,
                    help="path to input image")
    ap.add_argument("-p", "--port", required=True,
                    help="serving port")
    ap.add_argument("-n", "--module_name", required=True,
                    help="module_name")
          
    args = vars(ap.parse_args())
    data_path = args['image_path']
    serving_port = args['port']
    module_name = args['module_name']
    
    // 获取图片的base64编码格式
    img1 = cv2_to_base64(cv2.imread(data_path))
    data = {'images': img1}
    // 指定content-type
    headers = {"Content-type": "application/json"}
    // 发送HTTP请求
    url = "http://127.0.0.1:"+serving_port+"/predict/"+module_name
    r = requests.post(url=url, headers=headers, data=json.dumps(data))

    // 打印预测结果
    print(r.json()["results"])

serving_client.txt