ChatGPT 介紹

• platform.openai.com/examples
• platform.openai.com/docs/api-re…

ChatGPT可以實(shí)現(xiàn)chat,生成圖片,識別關(guān)鍵,改錯等等功能，本文簡單介紹如何使用python調(diào)用ChatGPT API 接口。

1. 生成API Key

從openai官網(wǎng)網(wǎng)址：platform.openai.com/account/api…，生成我們的API key：

獲得key后我們就可以調(diào)用API接口了。

2. 安裝openai

使用pip安裝openai庫,命令： pip install openai。

安裝openai庫: pip install openai 

3. python代碼調(diào)用API

3.1 主要步驟

# 調(diào)用openai api的步驟 # 1. 安裝openai庫 pip install openai # 2. 設(shè)置openai的api_key  # 3. 調(diào)用openai的api # 4. 參考文檔 # https://platform.openai.com/docs/api-reference/completions/create # https://platform.openai.com/docs/api-reference/authentication # https://platform.openai.com/docs/api-reference/completions/create # https://platform.openai.com/docs/libraries/community-libraries   

3.2 代碼示例

API可以實(shí)現(xiàn)chat,生成圖片,識別關(guān)鍵,改錯等等功能。 下面是主要示例代碼。 注意：openai.api_key = "sk-xxxFQ" #要更換成自已的API KEY

 #!/usr/local/bin/python3.8 # -*- coding: utf8 -*-   # 調(diào)用openai api的步驟 # 1. 安裝openai庫 pip install openai # 2. 設(shè)置openai的api_key  # 3. 調(diào)用openai的api # 4. 參考文檔 # https://platform.openai.com/docs/api-reference/completions/create # https://platform.openai.com/docs/api-reference/authentication # https://platform.openai.com/docs/api-reference/completions/create # https://platform.openai.com/docs/libraries/community-libraries    import os import openai import json # 1. 準(zhǔn)備好請求的url #openai.organization = "YOUR_ORG_ID" #  #openai.api_key = os.getenv("OPENAI_API_KEY") openai.api_key = "sk-xxxFQ" #要更換成自已的API KEY   # 查看可以使用的模型列表 def get_model_list():     models= openai.Model.list()     print(models)  # 生成文本示例 def generate_text(prompt):     completions = openai.Completion.create(         engine="text-davinci-002",         prompt=prompt,         max_tokens=1024,         n=1,         stop=None,         temperature=0.5,     )     message = completions.choices[0].text     return message.strip()  # 調(diào)用openai 畫圖示例 def generate_image(prompt):     response = openai.Image.create(         prompt = prompt,         n=1,         size="512x512"     )     image_url = response["data"
][0]["url"]     return image_url  # 調(diào)用openai 問答示例 def chat(prompt):     response = openai.ChatCompletion.create(     model="gpt-3.5-turbo",     messages=[         {"role": "user", "content":prompt}     ] )     answer = response.choices[0].message.content     return answer  # 調(diào)用openai 改正錯詞輸出正確句子 def correct():     prompt="改正錯詞輸出正確句子:\n\n我在京東電商平臺買了蘋果耳幾和華為體脂稱"  #建議prompt: 改正錯詞輸出正確句子:\n\n input_sentence     response = openai.ChatCompletion.create(     model="gpt-3.5-turbo",     messages=[         {"role": "user", "content":prompt}     ] )     answer = response.choices[0].message.content     return answer  # 調(diào)用openai 識別關(guān)鍵詞 def keyword():     prompt="對下面內(nèi)容識別2個關(guān)鍵詞，每個詞字?jǐn)?shù)不超過3個字:\n\n齊選汽車掛件車內(nèi)掛飾車載后視鏡吊墜高檔實(shí)心黃銅玉石出入平安保男女 紅流蘇-玉髓平安扣"  #建議prompt: 對下面內(nèi)容識別n個關(guān)鍵詞，每個詞字?jǐn)?shù)不超過m個字:\n\n input data     response = openai.ChatCompletion.create(     model="gpt-3.5-turbo",     messages=[         {"role": "user", "content":prompt}     ] )     answer = response.choices[0].message.content     return answer  # 抽取文本向量 (Embedding) def embedding():     content = "蘋果手機(jī)"     response = openai.Embedding.create(     model="text-embedding-ada-002",     input=content )     answer = response.data[0].embedding     return answer   def api_test():     # 測試chat      # prompt = "人口最多的國家?"     # response = chat(prompt)     # print(response) #      # 測試generate_text     # prompt = "Hello, how are you today?"     # response = generate_text(prompt)     # print(response)       # 測試generate_image     #prompt = "a delicious dessert"     #response = generate_image(prompt)     #print(response)      # 測試correct     # response = correct()     # print(response) #輸出結(jié)果: 我在京東電商平臺買了蘋果耳機(jī)和華為體脂秤。      # 測試keyword     #response = keyword()     #print(response) #輸出結(jié)果: 掛件、平安扣      # 測試embedding     result = embedding()     print(len(result))     print(result)  if __name__ == "__main__":     api_test()  

4. flask實(shí)現(xiàn)chat效果的示例

github.com/openai/open…

下載git代碼：

git clone https://github.com/openai/openai-quickstart-python.git cd openai-quickstart-python cp .env.example .env  python -m venv venv . venv/bin/activate pip install -r requirements.txt flask run 

運(yùn)行效果：

<

本文鏈接：http://www.51huadong.com/cloundnews/11008716.html