Update app.py

app.py

@@ -7,28 +7,42 @@ import nltk
 import openai
 import time
 
-import pip
-import subprocess
-import sys 
-
 # install required libraries
 subprocess.check_call([sys.executable, "-m", "pip", "install", "-r", "requirements.txt"])
 
 # download required NLTK data packages
 nltk.download('punkt')
-nltk.download('all')  # or any other packages your project depends on
 
 # Put your OpenAI API key here
 openai.api_key = os.getenv('OpenAPI') 
 
-def call_openai_api(text, user_prompt):
+def create_persona(text):
+    max_retries = 5
+    for attempt in range(max_retries):
+        try:
+            response = openai.ChatCompletion.create(
+                model="gpt-3.5-turbo",
+                messages=[
+                    {"role": "system", "content": "You are a knowledgeable assistant."},
+                    {"role": "user", "content": text},
+                ]
+            )
+            return response['choices'][0]['message']['content']
+        except Exception as e:
+            if attempt < max_retries - 1:  # if it's not the last attempt
+                time.sleep(1)  # wait for 1 seconds before retrying
+                continue
+            else:
+                return str(e)  # return the exception message after the last attempt
+
+def call_openai_api(persona, user_prompt):
     max_retries = 5
     for attempt in range(max_retries):
         try:
             response = openai.ChatCompletion.create(
                 model="gpt-3.5-turbo",
                 messages=[
-                    {"role": "system", "content": f"You are a knowledgeable assistant with a persona based on this: \n{text}"},
+                    {"role": "system", "content": f"You are a knowledgeable assistant with a persona based on this: \n{persona}"},
                     {"role": "user", "content": user_prompt},
                 ]
             )
@@ -51,20 +65,11 @@ def pdf_to_text(file, user_prompt):
             text = ''
             for page in pdf.pages:
                 text += page.extract_text()
-            # Tokenize text
-            tokens = nltk.word_tokenize(text)
-            # If tokens are more than 2000, split into chunks
-            if len(tokens) > 2000:
-                for i in range(0, len(tokens), 2000):
-                    chunk = tokens[i:i + 2000]
-                    chunk_str = ' '.join(chunk)
-                    # Using OpenAI API
-                    response = call_openai_api(chunk_str, user_prompt)
-                    texts.append(response)
-            else:
-                # Using OpenAI API
-                response = call_openai_api(text, user_prompt)
-                texts.append(response)
+            # Create persona from text
+            persona = create_persona(text)
+            # Using OpenAI API
+            response = call_openai_api(persona, user_prompt)
+            texts.append(response)
     return '\n'.join(texts)
 
 iface = gr.Interface(