Upload app.py

app.py

@@ -0,0 +1,172 @@
+import gradio as gr
+from sentence_transformers import SentenceTransformer, util
+import openai
+import os
+import os
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+
+# Initialize paths and model identifiers for easy configuration and maintenance
+filename = "output_country_details.txt"  # Path to the file storing country-specific details
+retrieval_model_name = 'output/sentence-transformer-finetuned/'
+
+openai.api_key = 'sk-proj-BVO7g5ig8PKdlQwDCZSeT3BlbkFJAvilYAEcPFbA0XOjz7ce'
+
+
+
+# Attempt to load the necessary models and provide feedback on success or failure
+try:
+    retrieval_model = SentenceTransformer(retrieval_model_name)
+    print("Models loaded successfully.")
+except Exception as e:
+    print(f"Failed to load models: {e}")
+
+def load_and_preprocess_text(filename):
+    """
+    Load and preprocess text from a file, removing empty lines and stripping whitespace.
+    """
+    try:
+        with open(filename, 'r', encoding='utf-8') as file:
+            segments = [line.strip() for line in file if line.strip()]
+        print("Text loaded and preprocessed successfully.")
+        return segments
+    except Exception as e:
+        print(f"Failed to load or preprocess text: {e}")
+        return []
+
+segments = load_and_preprocess_text(filename)
+
+def find_relevant_segment(user_query, segments):
+    """
+    Find the most relevant text segment for a user's query using cosine similarity among sentence embeddings.
+    This version tries to match country names in the query with those in the segments.
+    """
+    try:
+        # Lowercase the query for better matching
+        lower_query = user_query.lower()
+        # Filter segments to include only those containing country names mentioned in the query
+        country_segments = [seg for seg in segments if any(country.lower() in seg.lower() for country in ['Guatemala', 'Mexico', 'U.S.', 'United States'])]
+        
+        # If no specific country segments found, default to general matching
+        if not country_segments:
+            country_segments = segments
+        
+        query_embedding = retrieval_model.encode(lower_query)
+        segment_embeddings = retrieval_model.encode(country_segments)
+        similarities = util.pytorch_cos_sim(query_embedding, segment_embeddings)[0]
+        best_idx = similarities.argmax()
+        return country_segments[best_idx]
+    except Exception as e:
+        print(f"Error in finding relevant segment: {e}")
+        return ""
+
+
+def generate_response(user_query, relevant_segment):
+    """
+    Generate a response using the latest GPT-3 model available via OpenAI's API.
+    """
+    try:
+        prompt = f"Thank you for your question! Here's additional information: {relevant_segment}"
+        response = openai.Completion.create(
+            engine="gpt-3.5-turbo-instruct",  # Updated to a currently supported engine
+            prompt=prompt,
+            max_tokens=150,
+            temperature=0.7,
+            top_p=1,
+            frequency_penalty=0,
+            presence_penalty=0
+        )
+        return response.choices[0].text.strip()
+    except Exception as e:
+        print(f"Error in generating response: {e}")
+        return f"Error in generating response: {e}"
+
+
+
+# Define and configure the Gradio application interface to interact with users.
+# Define and configure the Gradio application interface to interact with users.
+def query_model(question):
+    """
+    Process a question, find relevant information, and generate a response, specifically for U.S. visa questions.
+    """
+    if question == "":
+        return "Welcome to VisaBot! Ask me anything about U.S. visa processes."
+    relevant_segment = find_relevant_segment(question, segments)
+    if not relevant_segment:
+        return "Could not find U.S.-specific information. Please refine your question."
+    response = generate_response(question, relevant_segment)
+    return response
+
+
+# Define the welcome message and specific topics and countries the chatbot can provide information about.
+welcome_message = """
+# Welcome to VISABOT!
+
+## Your AI-driven visa assistant for all travel-related queries.
+"""
+
+topics = """
+### Feel Free to ask me anything from the topics below!
+- Visa issuance
+- Documents needed
+- Application process
+- Processing time
+- Recommended Vaccines
+- Health Risks
+- Healthcare Facilities
+- Currency Information
+- Embassy Information 
+- Allowed stay
+"""
+
+countries = """
+### Our chatbot can currently answer questions for these countries!
+- 🇨🇳 China
+- 🇫🇷 France
+- 🇬🇹 Guatemala
+- 🇱🇧 Lebanon
+- 🇲🇽 Mexico
+- 🇵🇭 Philippines
+- 🇷🇸 Serbia
+- 🇸🇱 Sierra Leone
+- 🇿🇦 South Africa
+- 🇻🇳 Vietnam
+"""
+
+# Define and configure the Gradio application interface to interact with users.
+def query_model(question):
+    """
+    Process a question, find relevant information, and generate a response.
+    
+    Args:
+        question (str): User's input question.
+
+    Returns:
+        str: Generated response or a default welcome message if no question is provided.
+    """
+    if question == "":
+        return welcome_message
+    relevant_segment = find_relevant_segment(question, segments)
+    response = generate_response(question, relevant_segment)
+    return response
+
+# Setup the Gradio Blocks interface with custom layout components
+with gr.Blocks() as demo:
+    gr.Markdown(welcome_message)  # Display the formatted welcome message
+    with gr.Row():
+        with gr.Column():
+            gr.Markdown(topics)  # Show the topics on the left side
+        with gr.Column():
+            gr.Markdown(countries)  # Display the list of countries on the right side
+    with gr.Row():
+        img = gr.Image(os.path.join(os.getcwd(), "poster.png"), width=500)  # Include an image for visual appeal
+    with gr.Row():
+        with gr.Column():
+            question = gr.Textbox(label="Your question", placeholder="What do you want to ask about?")
+            answer = gr.Textbox(label="VisaBot Response", placeholder="VisaBot will respond here...", interactive=False, lines=10)
+            submit_button = gr.Button("Submit")
+            submit_button.click(fn=query_model, inputs=question, outputs=answer)
+
+# Launch the Gradio app to allow user interaction
+demo.launch(share= True)
+