Python 通过logic app上的函数部署AZURE自定义标签表单模型
我使用标签工具来训练自定义模型。该工具提供了一个用于使用它的python脚本。它在终点站运行良好。现在我想把它放在逻辑应用程序中的Azure函数中。pdf电子邮件附件将触发表单分析器模型,然后我将解析JSON响应。到目前为止,我还无法:Python 通过logic app上的函数部署AZURE自定义标签表单模型,python,azure,Python,Azure,我使用标签工具来训练自定义模型。该工具提供了一个用于使用它的python脚本。它在终点站运行良好。现在我想把它放在逻辑应用程序中的Azure函数中。pdf电子邮件附件将触发表单分析器模型,然后我将解析JSON响应。到目前为止,我还无法: 已在Azure函数中成功部署提供的脚本 在一个逻辑应用程序中使用它来识别我将存入blob或通过电子邮件发送的每个pdf 这是Azure表单识别器工具提供的脚本: ########### Python Form Recognizer Async Analyze #
########### Python Form Recognizer Async Analyze #############
import json
import time
import getopt
import sys
import os
from requests import get, post
def main(argv):
input_file, output_file, file_type = getArguments(argv)
runAnalysis(input_file, output_file, file_type)
def runAnalysis(input_file, output_file, file_type):
# Endpoint URL
endpoint = r"##################################/"
# Subscription Key
apim_key = "####################################"
# Model ID
model_id = "######################################"
# API version
API_version = "v2.1-preview.3"
post_url = endpoint + "/formrecognizer/%s/custom/models/%s/analyze" % (API_version, model_id)
params = {
"includeTextDetails": True
}
headers = {
# Request headers
'Content-Type': file_type,
'Ocp-Apim-Subscription-Key': apim_key,
}
try:
with open(input_file, "rb") as f:
data_bytes = f.read()
except IOError:
print("Inputfile not accessible.")
sys.exit(2)
try:
print('Initiating analysis...')
resp = post(url = post_url, data = data_bytes, headers = headers, params = params)
if resp.status_code != 202:
print("POST analyze failed:\n%s" % json.dumps(resp.json()))
quit()
print("POST analyze succeeded:\n%s" % resp.headers)
print
get_url = resp.headers["operation-location"]
except Exception as e:
print("POST analyze failed:\n%s" % str(e))
quit()
n_tries = 15
n_try = 0
wait_sec = 5
max_wait_sec = 60
print()
print('Getting analysis results...')
while n_try < n_tries:
try:
resp = get(url = get_url, headers = {"Ocp-Apim-Subscription-Key": apim_key})
resp_json = resp.json()
if resp.status_code != 200:
print("GET analyze results failed:\n%s" % json.dumps(resp_json))
quit()
status = resp_json["status"]
if status == "succeeded":
if output_file:
with open(output_file, 'w') as outfile:
json.dump(resp_json, outfile, indent=2, sort_keys=True)
print("Analysis succeeded:\n%s" % json.dumps(resp_json, indent=2, sort_keys=True))
quit()
if status == "failed":
print("Analysis failed:\n%s" % json.dumps(resp_json))
quit()
# Analysis still running. Wait and retry.
time.sleep(wait_sec)
n_try += 1
wait_sec = min(2*wait_sec, max_wait_sec)
except Exception as e:
msg = "GET analyze results failed:\n%s" % str(e)
print(msg)
quit()
print("Analyze operation did not complete within the allocated time.")
def getArguments(argv):
input_file = ''
file_type = ''
output_file = ''
try:
opts, args = getopt.gnu_getopt(argv, "ht:o:", [])
except getopt.GetoptError:
printCommandDescription(2)
for opt, arg in opts:
if opt == '-h':
printCommandDescription()
if len(args) != 1:
printCommandDescription()
else:
input_file = args[0]
for opt, arg in opts:
if opt == '-t':
if arg not in ('application/pdf', 'image/jpeg', 'image/png', 'image/tiff', 'image/bmp'):
print('Type ' + file_type + ' not supported')
sys.exit()
else:
file_type = arg
if opt == '-o':
output_file = arg
try:
open(output_file, 'a')
except IOError:
print("Output file not creatable")
sys.exit(2)
if not file_type:
file_type = inferrType(input_file)
return (input_file, output_file, file_type)
def inferrType(input_file):
filename, file_extension = os.path.splitext(input_file)
if file_extension == '':
print('File extension could not be inferred from inputfile. Provide type as an argument.')
sys.exit()
elif file_extension == '.pdf':
return 'application/pdf'
elif file_extension == '.jpeg':
return 'image/jpeg'
elif file_extension == '.bmp':
return 'image/bmp'
elif file_extension == '.png':
return 'image/png'
elif file_extension == '.tiff':
return 'image/tiff'
else:
print('File extension ' + file_extension + ' not supported')
sys.exit()
def printCommandDescription(exit_status=0):
print('analyze.py <inputfile> [-t <type>] [-o <outputfile>]')
print
print('If type option is not provided, type will be inferred from file extension.')
sys.exit(exit_status)
if __name__ == '__main__':
main(sys.argv[1:])
Python表单识别器异步分析#############
导入json
导入时间
导入getopt
导入系统
导入操作系统
从请求导入获取、发布
def总管(argv):
输入文件、输出文件、文件类型=getArguments(argv)
运行分析(输入文件、输出文件、文件类型)
def运行分析(输入文件、输出文件、文件类型):
#端点URL
终点=r
#订阅密钥
apim#u key=“#########################
#模型ID
3月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月月日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日日#”
#API版本
API_version=“v2.1-preview.3”
post_url=endpoint+“/formrecognizer/%s/custom/models/%s/analyze”%(API_版本,model_id)
参数={
“includeTextDetails”:True
}
标题={
#请求头
“内容类型”:文件类型,
“Ocp Apim订阅密钥”:Apim_密钥,
}
尝试:
打开(输入_文件“rb”)作为f:
数据字节=f.read()
除IOError外:
打印(“Inputfile不可访问”)
系统出口(2)
尝试:
打印('启动分析…')
resp=post(url=post\u url,data=data\u字节,headers=headers,params=params)
如果响应状态\ U代码!=202:
打印(“后期分析失败:\n%s”%json.dumps(resp.json()))
退出
打印(“后期分析成功:\n%s”%resp.headers)
打印
get_url=resp.headers[“操作位置”]
例外情况除外,如e:
打印(“后期分析失败:\n%s”%str(e))
退出
n_=15
n_try=0
等待秒=5
最大等待时间=60秒
打印()
打印('正在获取分析结果…')
当n_trydef main(req: func.HttpRequest) -> func.HttpResponse:
logging.info('Python HTTP trigger function processed a request.')
...............
然后需要将上面提供的python代码编写到函数中,并进行一些修改。将代码融合到函数HttpRequest代码体中
之后,该函数将从本地迁移到azure
然后你可以在你的逻辑应用程序中使用该功能