issue #80: openai migrate

ailab/models/openai.py

@@ -1,59 +1,62 @@
 """"Fetch embeddings from the Microsoft Azure OpenAI API"""
+
 import os
-import openai
+
 import tiktoken
+from dotenv import load_dotenv
+from openai import AzureOpenAI
 
-import dotenv
-dotenv.load_dotenv()
+load_dotenv()
 
 # https://learn.microsoft.com/en-us/azure/cognitive-services/openai/how-to/embeddings?tabs=python
 
+
 def safe_get(key):
     value = os.environ.get(key)
     if not value:
         raise Exception(f"Environment variable {key} not defined")
     return value
 
+
 OPENAI_API_KEY = safe_get("OPENAI_API_KEY")
 OPENAI_ENDPOINT = safe_get("OPENAI_ENDPOINT")
 
-openai.api_type = "azure"
-openai.api_key = OPENAI_API_KEY
-openai.api_base = OPENAI_ENDPOINT
-openai.api_version = "2023-05-15" # be sure it's the good one
-
+client = AzureOpenAI(
+    api_key=OPENAI_API_KEY, azure_endpoint=OPENAI_ENDPOINT, api_version="2023-05-15"
+)
 enc = tiktoken.get_encoding("cl100k_base")
 
+
 def fetch_embedding(tokens):
     """Fetch embedding for a list of tokens from the Microsoft Azure OpenAI API"""
     OPENAI_API_ENGINE = safe_get("OPENAI_API_ENGINE")
-
-    response = openai.Embedding.create(
-        input=tokens,
-        engine=OPENAI_API_ENGINE
-    )
-    embeddings = response['data'][0]['embedding']
+
+    response = client.embeddings.create(input=tokens, engine=OPENAI_API_ENGINE)
+    embeddings = response["data"][0]["embedding"]
     return embeddings
 
+
 # def fetch_tokens_embeddings(text):
 #     tokens = get_tokens_from_text(text)
 #     embeddings = fetch_embedding(tokens)
 #     return (tokens, embeddings)
 
+
 def get_tokens_from_text(text):
     tokens = enc.encode(text)
     return tokens
 
+
 def get_chat_answer(system_prompt, user_prompt, max_token):
     OPENAI_API_ENGINE = safe_get("OPENAI_API_ENGINE")
 
-    response = openai.ChatCompletion.create(
+    response = client.chat.completions.create(
         engine=OPENAI_API_ENGINE,
         temperature=0,
         max_tokens=max_token,
-        messages = [
+        messages=[
             {"role": "system", "content": system_prompt},
-            {"role": "user", "content": (user_prompt)}
-        ]
+            {"role": "user", "content": (user_prompt)},
+        ],
     )
     return response

requirements.txt

@@ -10,7 +10,7 @@ idna>=3.4
 iniconfig>=2.0.0
 multidict>=6.0.4
 numpy>=1.25.1
-openai>=0.27.8
+openai>=1.14.3
 openpyxl>=3.1.2
 packaging>=23.1
 pandas>=2.0.3

tests/fetch_embedding.py

@@ -1,27 +1,29 @@
-
-import openai
-import os 
+import os
 import sys
-import dotenv
-dotenv.load_dotenv()
 
-openai.api_type = "azure"
-openai.api_key = os.environ["OPENAI_API_KEY"]
-openai.api_base = f"https://{os.environ['AZURE_OPENAI_SERVICE']}.openai.azure.com"
-openai.api_version = "2023-05-15"
+from dotenv import load_dotenv
+from openai import AzureOpenAI
+
+load_dotenv()
+
+
+client = AzureOpenAI(
+    api_key=os.environ["OPENAI_API_KEY"],
+    azure_endpoint=f"https://{os.environ['AZURE_OPENAI_SERVICE']}.openai.azure.com",
+    api_version="2023-05-15",
+)
+
 
 def fetch_embedding(text):
     """Fetch embedding for a list of tokens from the Microsoft Azure OpenAI API"""
-    response = openai.Embedding.create(
-        input=text,
-        engine="ada"
-    )
-    embeddings = response['data'][0]['embedding']
+    response = client.embeddings.create(input=text, model="ada")
+    embeddings = response.data[0].embedding
     return embeddings
 
-if __name__ == '__main__':
+
+if __name__ == "__main__":
     text = " ".join(sys.argv[1:])
     if len(text) == 0:
-        print('Please provide a text to embed')
+        print("Please provide a text to embed")
         raise SystemExit
-    print(fetch_embedding(text))
+    print(fetch_embedding(text))