Update app.py

app.py

@@ -1,50 +1,39 @@
-from gpt_index import SimpleDirectoryReader, GPTSimpleVectorIndex, LLMPredictor, PromptHelper
-from langchain.chat_models import ChatOpenAI
 import gradio as gr
+import openai
 import os
 
-# Set your OpenAI API key here
+# Ensure the OPENAI_API_KEY environment variable is set correctly
 os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY")
 
-def construct_index(directory_path):
-    max_input_size = 4096
-    num_outputs = 512
-    max_chunk_overlap = 20
-    chunk_size_limit = 600
-
-    # Initialize the prompt helper
-    prompt_helper = PromptHelper(max_input_size, num_outputs, max_chunk_overlap, chunk_size_limit=chunk_size_limit)
-    
-    # Initialize the LLM predictor with an updated model name if available
-   #llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0.7, model_name="gpt-4", max_tokens=num_outputs))  # Updated to GPT-4 or latest available model
-    #llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0.7, model_name="gpt-3.5-turbo", max_tokens=num_outputs))
-    llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0.7, model_name="gpt-4", max_tokens=num_outputs))
-
-    # Load documents from the directory
-    documents = SimpleDirectoryReader(directory_path).load_data()
+def generate_text(prompt):
+    # Add custom text to the prompt
+    input_text = (prompt + " You are an expert with many years of experience in the insurance loss adjusting business. "
+                           "Can you give your expert opinion on the question or statement made at the start of this prompt? "
+                           "Answer it in terms of the laws, regulations, and terms of the insurance industry. "
+                           "Make it accurate, but written in a simple and clear style. "
+                           "Please provide references or insurance law precedents where possible.")
     
-    # Create and populate the index
-    index = GPTSimpleVectorIndex(documents, llm_predictor=llm_predictor, prompt_helper=prompt_helper)
-    index.save_to_disk('index.json')
-    
-    return index
-
-def chatbot(input_text):
-    # Load the index from disk
-    index = GPTSimpleVectorIndex.load_from_disk('index.json')
-    
-    # Query the index and return the response
-    response = index.query(input_text, response_mode="compact")
-    return response.response
-
-# Set up the Gradio interface
-iface = gr.Interface(fn=chatbot,
-                     inputs=gr.components.Textbox(lines=7, label="Enter your text"),
-                     outputs="text",
-                     title="Custom-trained AI Chatbot")
-
-# Construct the index with your documents directory
-index = construct_index("docs")
-
-# Launch the Gradio interface
-iface.launch(share=False)
+    # Generate text using OpenAI API. Adjust "model" to the latest available or desired model.
+    response = openai.Completion.create(
+        model="gpt-3.5-turbo",  # Update this to the latest model you wish to use
+        prompt=input_text,
+        max_tokens=1024,
+        n=1,
+        stop=None,
+        temperature=0.7
+    )
+
+    # Return the generated text
+    return response.choices[0].text.strip()
+
+# Define the Gradio interface using the updated syntax
+iface = gr.Interface(
+    fn=generate_text,
+    inputs=gr.Textbox(lines=5, label="What do you wish to know about insurance?"),
+    outputs=gr.Textbox(label="Generated text:"),
+    title="AI Adjuster",
+    description="AI tool to help you understand and answer any insurance-related queries."
+)
+
+# Run the Gradio app
+iface.launch()