在 python 中为 Watson NLU 调用多个 API

问题描述

我正在研究 Watson NLU，我需要对问卷数据进行分析。来自不同人的大约 300 个答案。 我能够在“...”格式文本上运行它，但我很想获得一些帮助，了解如何一次性运行所有 300 个。我当前的输入是带有 ID 列的 excel。 感谢您对此进行调查。

nlu_api_key = "MY API KEY"
nlu_url = "https://api.eu-gb.natural-language-understanding.watson.cloud.ibm.com/instances/MY INSTANCE"


import json
from ibm_watson import NaturalLanguageUnderstandingV1
from ibm_cloud_sdk_core.authenticators import IAMAuthenticator
from ibm_watson.natural_language_understanding_v1 import Features,EntitiesOptions,KeywordsOptions,CategoriesOptions,SentimentOptions
import pandas as pd


gtm_Q6 = pd.read_excel(r'C:\Users\...\INPUT FILE.xlsx',sheet_name='OUPUT1')
print(gtm_Q6)


authenticator = IAMAuthenticator(nlu_api_key)
natural_language_understanding = NaturalLanguageUnderstandingV1(
    version='2020-08-01',authenticator=authenticator)

natural_language_understanding.set_service_url(nlu_url)

response = natural_language_understanding.analyze(
    text='Where is the firetruck with the flaming paint the tigers on top?',features=Features(
        entities=EntitiesOptions(emotion=True,sentiment=True,limit=5),keywords=KeywordsOptions(emotion=True,categories=CategoriesOptions(limit=3),sentiment=SentimentOptions(targets=['investments']) #sentiment=SentimentOptions(targets=['stocks'])
        )).get_result()

print(json.dumps(response,indent=2))

RESP_ID	答案
Q6_109.000000	团队建设
Q6_110.000000	技术和服务之间的支持和协调
Q6_111.000000	技能培养
Q6_113.000000	快速获取合适的资源
Q6_114.000000	关于目前变更的实用性的信息

解决方法

所以已经提供了解决方案 py T.J. Python 中的 Crowder： 这是他原始答案的链接：How to get rid of "\n" and " ' ' " in my json file

json_arr = []
for text in gtm_Q6_df['ANSWER']:
    response = natural_language_understanding.analyze(
        text=text,language = 'en',features=Features(
            entities=EntitiesOptions(sentiment=True,emotion=False,limit=3),categories=CategoriesOptions(limit=3),#sentiment=SentimentOptions(EntitiesOptions)
            )).get_result()
    
    #text = ("{ Answer: " + text + "}")
    json_arr.append(text)
    json_arr.append (response) # <==== change is here
    with open(r'C:\Users\...\Documents\Python NLP\WATSON NLU\OUTPUT JSON\data.json','w') as outline: #with open('data.json','w') as outline: ## data.json is the output file created ## the file can be renamed
        json.dump(json_arr,outline,indent = 2) # indent = 2 creates the pretty json!!
    
print("break")
print(json_arr)