Refactor the project

README.md

@@ -4,11 +4,11 @@ emoji: 🏢
 colorFrom: indigo
 colorTo: green
 sdk: gradio
-sdk_version: 4.36.1
+sdk_version: 3.20.1
 app_file: app.py
 pinned: false
 license: afl-3.0
 duplicated_from: bhaskartripathi/pdfChatter
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -1,13 +1,11 @@
-import os
-os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'  # Suppress TF logging
-os.environ["CUDA_VISIBLE_DEVICES"] = "-1"  # Disable GPU
 import urllib.request 
-import fitz 
+import fitz
 import re
 import numpy as np
 import tensorflow_hub as hub
 import openai
 import gradio as gr
+import os
 from sklearn.neighbors import NearestNeighbors
 
 def download_pdf(url, output_path):
@@ -102,11 +100,11 @@ def load_recommender(path, start_page=1):
 
 def generate_text(openAI_key, prompt, model="gpt-3.5-turbo"):
     openai.api_key = openAI_key
-    temperature = 0.1
-    max_tokens = 256
-    top_p = 1
-    frequency_penalty = 0
-    presence_penalty = 0
+    temperature=0.7
+    max_tokens=256
+    top_p=1
+    frequency_penalty=0
+    presence_penalty=0
 
     if model == "text-davinci-003":
         completions = openai.Completion.create(
@@ -119,19 +117,19 @@ def generate_text(openAI_key, prompt, model="gpt-3.5-turbo"):
         )
         message = completions.choices[0].text
     else:
-        response = openai.ChatCompletion.create(
+        message = openai.ChatCompletion.create(
             model=model,
             messages=[
                 {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "assistant", "content": "Here is some initial assistant message."},
                 {"role": "user", "content": prompt}
             ],
-            temperature=temperature,
+            temperature=.3,
             max_tokens=max_tokens,
             top_p=top_p,
             frequency_penalty=frequency_penalty,
             presence_penalty=presence_penalty,
-        )
-        message = response['choices'][0]['message']['content']
+        ).choices[0].message['content']
     return message
 
   
@@ -240,93 +238,50 @@ title = 'PDF GPT Turbo'
 description = """ PDF GPT Turbo allows you to chat with your PDF files. It uses Google's Universal Sentence Encoder with Deep averaging network (DAN) to give hallucination free response by improving the embedding quality of OpenAI. It cites the page number in square brackets([Page No.]) and shows where the information is located, adding credibility to the responses."""
 
 with gr.Blocks(css="""#chatbot { font-size: 14px; min-height: 1200; }""") as demo:
+
     gr.Markdown(f'<center><h3>{title}</h3></center>')
     gr.Markdown(description)
 
     with gr.Row():
-        with gr.Column():
-            # API Key and File Inputs
-            with gr.Accordion("API Key and PDF"):
-                openAI_key = gr.Textbox(label='Enter your OpenAI API key here', type='password')
-                url = gr.Textbox(label='Enter PDF URL here (Example: https://arxiv.org/pdf/1706.03762.pdf ; https://link.springer.com/content/pdf/10.1007/s10614-022-10325-8.pdf)')
+        
+        with gr.Group():
+            gr.Markdown(f'<p style="text-align:center">Get your Open AI API key <a href="https://platform.openai.com/account/api-keys">here</a></p>')
+            with gr.Accordion("API Key"):
+                openAI_key = gr.Textbox(label='Enter your OpenAI API key here', password=True)
+                url = gr.Textbox(label='Enter PDF URL here   (Example: https://arxiv.org/pdf/1706.03762.pdf )')
                 gr.Markdown("<center><h4>OR<h4></center>")
-                file = gr.File(label='Upload your PDF/Research Paper/Book here', file_types=['.pdf'])
-            
-            # Model Selection
-            model = gr.Radio(
-                choices=[
-                    'gpt-4o-mini',
-                    'gpt-4o',
-                    'gpt-4',
-                ], 
-                label='Select Model', 
-                value='gpt-4o-mini'
-            )
-
-            # Chat Interface
-            chatbot = gr.Chatbot(label="Chat History", type="messages")
-            msg = gr.Textbox(label="Enter your question here", lines=2)
-            submit_btn = gr.Button("Submit")
-            clear = gr.ClearButton([msg, chatbot])
-
-            # Example Questions
+                file = gr.File(label='Upload your PDF/ Research Paper / Book here', file_types=['.pdf'])
+            question = gr.Textbox(label='Enter your question here')
             gr.Examples(
                 [[q] for q in questions],
-                inputs=[msg],
-                label="PRE-DEFINED QUESTIONS: Click on a question to auto-fill the input box",
+                inputs=[question],
+                label="PRE-DEFINED QUESTIONS: Click on a question to auto-fill the input box, then press Enter!",
             )
-
-    def respond(message, chat_history, url_value, file_value, key_value, model_value):
-        if message.strip() == "":
-            return "", chat_history  # Return empty message if no input
-
-        try:
-            # Ensure chat_history is initialized properly
-            if chat_history is None:
-                chat_history = []
-
-            if key_value.strip() == '':
-                chat_history.append({"role": "user", "content": message})
-                chat_history.append({"role": "assistant", "content": '[ERROR]: Please enter your OpenAI API key'})
-                return "", chat_history
-
-            if url_value.strip() == '' and file_value is None:
-                chat_history.append({"role": "user", "content": message})
-                chat_history.append({"role": "assistant", "content": '[ERROR]: Both URL and PDF are empty. Provide at least one'})
-                return "", chat_history
-
-            # Process PDF and generate answer
-            if url_value.strip() != '':
-                download_pdf(url_value, 'corpus.pdf')
-                load_recommender('corpus.pdf')
-            else:
-                old_file_name = file_value.name
-                file_name = old_file_name[:-12] + old_file_name[-4:]
-                os.rename(old_file_name, file_name)
-                load_recommender(file_name)
-
-            answer = generate_answer(message, key_value, model_value)
-
-            chat_history.append({"role": "user", "content": message})
-            chat_history.append({"role": "assistant", "content": answer})
-
-            return "", chat_history
-
-        except Exception as e:
-            chat_history.append({"role": "user", "content": message})
-            chat_history.append({"role": "assistant", "content": f'[ERROR]: {str(e)}'})
-            return "", chat_history
-
-    submit_btn.click(
-        respond, 
-        [msg, chatbot, url, file, openAI_key, model], 
-        [msg, chatbot]
+            model = gr.Radio([
+                'gpt-3.5-turbo', 
+                'gpt-3.5-turbo-16k', 
+                'gpt-3.5-turbo-0613', 
+                'gpt-3.5-turbo-16k-0613', 
+                'text-davinci-003',
+                'gpt-4',
+                'gpt-4-32k'
+            ], label='Select Model', default='gpt-3.5-turbo')
+            btn = gr.Button(value='Submit')
+
+            btn.style(full_width=True)
+
+        with gr.Group():
+            chatbot = gr.Chatbot(placeholder="Chat History", label="Chat History", lines=50, elem_id="chatbot")
+
+
+#
+    # Bind the click event of the button to the question_answer function
+    btn.click(
+        question_answer,
+        inputs=[chatbot, url, file, question, openAI_key, model],
+        outputs=[chatbot],
     )
 
-    msg.submit(
-        respond, 
-        [msg, chatbot, url, file, openAI_key, model], 
-        [msg, chatbot]
-    )
+demo.launch()
+
 
-demo.launch()

requirements.txt

@@ -1,9 +1,7 @@
-urllib3
-PyMuPDF==1.18.19
+gradio
+PyMuPDF
 numpy
-tensorflow-hub
-openai
-gradio>=4.44.1
 scikit-learn
-openai==0.28
-anthropic>=0.7.0
+tensorflow
+tensorflow-hub
+openai