app.py

import gradio as gr
from sentence_transformers import SentenceTransformer, util
from transformers import GPT2LMHeadModel, GPT2Tokenizer
import os
import torch

os.environ["TOKENIZERS_PARALLELISM"] = "false"

# Initialize paths and model identifiers for easy configuration and maintenance
filename = "output_topic_details.txt"  # Path to the file storing destress-specific details
retrieval_model_name = 'output/sentence-transformer-finetuned/'

# Load GPT-2 model and tokenizer
tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
model = GPT2LMHeadModel.from_pretrained("gpt2")

system_message = "You are a comfort chatbot specialized in providing information on therapy, destressing activities, and student opportunities."
messages = [{"role": "system", "content": system_message}]
messages.append({
    "role": "system",
    "content": "Do not use Markdown Format. Do not include hashtags or asterisks"
})

# Load the retrieval model
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):
    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):
    try:
        lower_query = user_query.lower()
        query_embedding = retrieval_model.encode(lower_query)
        segment_embeddings = retrieval_model.encode(segments)
        similarities = util.pytorch_cos_sim(query_embedding, segment_embeddings)[0]
        best_idx = similarities.argmax()
        return segments[best_idx]
    except Exception as e:
        print(f"Error in finding relevant segment: {e}")
        return ""

def generate_response_with_context(user_query, relevant_segment):
    """
    Generate a response based on a user query and a relevant segment.
    
    Parameters:
    - user_query (str): The user's query.
    - relevant_segment (str): A relevant fact or detail.
    
    Returns:
    - str: Formatted response incorporating the relevant segment.
    """
    try:
        # Prepare the prompt incorporating the relevant segment
        prompt = f"User: {user_query}\n\nAssistant: Here is some helpful information based on your topic: {relevant_segment}"
        
        # Calculate the maximum tokens allowed for the response
        max_tokens = len(tokenizer(prompt)['input_ids']) + 100
        
        # Generate the response using the model
        response = gpt_model(prompt, max_length=max_tokens, temperature=0.7)[0]['generated_text']
        
        # Clean up the response for better formatting and clarity
        return clean_up_response(response, relevant_segment)
    
    except Exception as e:
        print(f"Error generating response: {e}")
        return "I'm sorry, but there was an error generating your response. Please try again."


def generate_response(user_query, relevant_segment):
    try:
        user_message = f"Here's the information on your request: {relevant_segment}"
        messages.append({"role": "user", "content": user_message})

        # Encode the input and generate a response
        input_ids = tokenizer.encode(user_message, return_tensors='pt')

        # Create the attention mask (1 for real tokens, 0 for padding tokens)
        attention_mask = torch.ones(input_ids.shape, dtype=torch.long)  # Create a tensor of ones

        # Generate the response using the model
        output = model.generate(
            input_ids,
            attention_mask=attention_mask,
            max_length=150,
            num_return_sequences=1,
            pad_token_id=tokenizer.eos_token_id  # Set pad_token_id to eos_token_id
        )

        output_text = tokenizer.decode(output[0], skip_special_tokens=True)

        # Append assistant's message to messages list for context
        messages.append({"role": "assistant", "content": output_text})
        return output_text

    except Exception as e:
        print(f"Error in generating response: {e}")
        return f"Error in generating response: {e}"

def query_model(question):
    if question == "":
        return "Welcome to CalmConnect! Ask me anything about destressing strategies or student opportunities. Feel free to talk to our online therapist!"
    relevant_segment = find_relevant_segment(question, segments)
    if not relevant_segment:
        return "Could not find specific information. Please refine your question or head to our resources page."
    response = generate_response(question, relevant_segment)
    return response

# Define the HTML iframe content
iframe = '''
<iframe style="border-radius:12px" src="https://docs.google.com/spreadsheets/d/e/2PACX-1vRroWVBXq1Fa0x7SvRTzSBMHFIp59VtVEWCxeg8kWJU4ll1_o4yzBnt4ArT88s7g4TQrMKEXZUQAeHF/pubhtml?widget=true&amp;headers=false" width="100%" height="352" frameBorder="0" allowfullscreen="true" allow="autoplay; clipboard-write; encrypted-media; fullscreen; picture-in-picture" loading="lazy"></iframe>
'''

iframe2 = '''
<iframe style="border-radius:12px" src="https://open.spotify.com/embed/playlist/6wwxTePuIKYMqt6RCytB7X?utm_source=generator" width="100%" height="300" frameBorder="0" allowfullscreen="" allow="autoplay; clipboard-write; encrypted-media; fullscreen; picture-in-picture" loading="lazy"></iframe>
'''


# Define the welcome message and specific topics the chatbot can provide information about
welcome_message = """
<span style="color:#718355; font-size:24px; font-weight:bold;"> 🪷 Welcome to CalmConnect! 🪷</span>
"""

"""
## Your AI-driven assistant for destressing and extracurricular opportunity queries. Created by Olivia W, Alice T, and Cindy W of the 2024 Kode With Klossy CITY Camp. 
"""

topics = """
### If you are interested in the following below, click on our Student Opportunities Database!
- Engineering
- Technology / Computer Science
- Research : STEM
- Finance
- Law / Political Science / Debate
- The Arts
- Business / Leadership
- Pyschology
- Medicine / Biology
- Literature / Writing
- College Prep
- Advocacy: Non-Profit, Environment or Identity 
- Volunteering 
- Study Abroad
"""

topics2= """
### Feel Free to ask CalmBot (Our Therapist Bot) anything from the topics below!
- Arts and Crafts (When asking for arts and crafts ideas, state whether you have 15 min, 30 min, 45 min, 1 hour, 1 hour and a half, 2 hours, 2 hours and a half, 3 hours or greater)
- Destressing strategies (Breathing Exercises, stretches, etc.)
- Mental Health
- Identity (Sexual, Gender, etc.)
- Bullying
- Racism
- Relationships (Family, Friends, etc.)
- Abuse (Emotional, Physical, Sexual, Mental, etc.)
- Support Resources
"""

# Create a Gradio HTML component
def display_iframe():
    return iframe

def display_iframe2():
    return iframe2


theme = gr.themes.Default(
    primary_hue="neutral",
    secondary_hue="neutral",
).set(
    background_fill_primary='#e3e9da',
    background_fill_primary_dark='#e3e9da',
    background_fill_secondary="#f8f1ea",
    background_fill_secondary_dark="#f8f1ea",
    border_color_accent="#f8f1ea",
    border_color_accent_dark="#e3e9da",
    border_color_accent_subdued="#f8f1ea",
    border_color_primary="#f8f1ea",
    block_border_color="#f8f1ea",
    button_primary_background_fill="#f8f1ea",
    button_primary_background_fill_dark="#f8f1ea"
)
    
# Setup the Gradio Blocks interface with custom layout components
with gr.Blocks(theme=theme) as demo:
    gr.Image("CalmConnect.jpg", show_label=False, show_share_button=False, show_download_button=False)
    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
            gr.HTML(iframe)  # Embed the iframe on the left side
            gr.HTML(iframe2)  # Embed the iframe on the right side
        
        with gr.Column():
            gr.Markdown(topics2)  # Show the topics on the left side
            with gr.Row():
                with gr.Column():
                    question = gr.Textbox(label="You", placeholder="What do you want to talk to CalmBot about?")
                    answer = gr.Textbox(label="CalmBot's Response :D", placeholder="CalmBot will respond here..", interactive=False, lines=20)
                    submit_button = gr.Button("Submit")
                    submit_button.click(fn=query_model, inputs=question, outputs=answer)
    
    with gr.Row():
        big_block = gr.HTML("### <button><a href='https://www.headspace.com/teens'>FREE: HEADSPACE FOR TEENS </a></button>")
        big_block2 = gr.HTML("<button><a href='https://calmconnect-flower.replit.app/'>PLAY FLOWER GAME</a></button>")
        big_block3 = gr.HTML("<button><a href='https://www.nyc.gov/site/doh/health/health-topics/teenspace.page'>NYC: TEENSPACE (free services)</a></button>")
        big_block4 =gr.HTML("<button><a href='https://www.teenlife.com/blog/mental-health-resources-for-teens/'>TEEN MENTAL HEALTH RESOURCES (free services)</a></button>")


demo.launch()





# Launch the Gradio app to allow user interaction
demo.launch(share=True)