app.py

import gradio as gr

# Define the theme with custom colors and styles, including larger text sizes
theme = gr.themes.Default(
    primary_hue=gr.themes.Color(
       c100="ed83b5", c200="#ec297b", c300="#fff1f4", c400="#f2e3e6",
        c50="#fff48b", c500="#ffffff", c600="#ec297b", c700="#ec297b",
        c800="##ec297b", c900="#ec297b", c950="#ec297b"

    
    ),
    secondary_hue="red",
    neutral_hue="pink",
    font=[gr.themes.GoogleFont('jack armstrong'), 'ui-sans-serif', 'system-ui', 'sans-serif'],
    font_mono=[gr.themes.GoogleFont('xkcd'), 'ui-monospace', 'Consolas', 'monospace'],
).set(
    body_text_color='*primary_950',
    body_text_color_dark='*secondary_50',
    body_text_size='26px',  # Increase body text size
    body_text_color_subdued='*primary_400',
    body_text_weight='500',
    background_fill_primary='*primary_300',
    background_fill_primary_dark='*primary_800',
    background_fill_secondary='*primary_50',
    background_fill_secondary_dark='*primary_600',
    border_color_accent='*secondary_300',
    border_color_accent_dark='*body_text_color',
    border_color_accent_subdued='*border_color_accent',
    link_text_color='*secondary_800',
    code_background_fill='*neutral_200',
    code_background_fill_dark='*neutral_100',
    block_shadow='none',
    block_shadow_dark='none',
    form_gap_width='0px',
    checkbox_label_background_fill='*button_secondary_background_fill',
    checkbox_label_background_fill_dark='*button_secondary_background_fill',
    checkbox_label_background_fill_hover='*button_secondary_background_fill_hover',
    checkbox_label_background_fill_hover_dark='*button_secondary_background_fill_hover',
    checkbox_label_shadow='none',
    error_background_fill_dark='*background_fill_primary',
    input_background_fill='*neutral_100',
    input_background_fill_dark='*neutral_700',
    input_border_width='0px',
    input_border_width_dark='0px',
    input_shadow='none',
    input_shadow_dark='none',
    input_shadow_focus='*input_shadow',
    input_shadow_focus_dark='*input_shadow',
    stat_background_fill='*primary_300',
    stat_background_fill_dark='*primary_500',
    button_shadow='none',
    button_shadow_active='none',
    button_shadow_hover='none',
    button_transition='background-color 0.2s ease',
    button_primary_background_fill='*primary_200',
    button_primary_background_fill_dark='*primary_700',
    button_primary_background_fill_hover='*button_primary_background_fill',
    button_primary_background_fill_hover_dark='*button_primary_background_fill',
    button_primary_border_color_dark='*primary_600',
    button_secondary_background_fill='*neutral_200',
    button_secondary_background_fill_dark='*neutral_600',
    button_secondary_background_fill_hover='*button_secondary_background_fill',
    button_secondary_background_fill_hover_dark='*button_secondary_background_fill',
    button_cancel_background_fill='*button_secondary_background_fill',
    button_cancel_background_fill_dark='*button_secondary_background_fill',
    button_cancel_background_fill_hover='*button_cancel_background_fill',
    button_cancel_background_fill_hover_dark='*button_cancel_background_fill',
    button_cancel_border_color='*button_secondary_border_color',
    button_cancel_border_color_dark='*button_secondary_border_color',
    button_cancel_text_color='*button_secondary_text_color',
    button_cancel_text_color_dark='*button_secondary_text_color'
)

from sentence_transformers import SentenceTransformer, util
import openai
import os
import random

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 health-specific details
retrieval_model_name = 'output/sentence-transformer-finetuned/'

openai.api_key = os.environ["HealthHub"]

system_message = "You are a health chatbot specialized in providing accurate health information on Menstrual health, COVID 19, and Mental Health."
# Initial system message to set the behavior of the assistant
messages = [{"role": "system", "content": system_message}]

# 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 finds the best match based on the content of the query.
    """
    try:
        # Lowercase the query for better matching
        lower_query = user_query.lower()
        
        # Encode the query and the segments
        query_embedding = retrieval_model.encode(lower_query)
        segment_embeddings = retrieval_model.encode(segments)
        
        # Compute cosine similarities between the query and the segments
        similarities = util.pytorch_cos_sim(query_embedding, segment_embeddings)[0]
        
        # Find the index of the most similar segment
        best_idx = similarities.argmax()
        
        # Return the most relevant segment
        return 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 emphasizing the bot's capability in providing health information.
    """
    try:
        user_message = f"Here's the information on health: {relevant_segment}"

        # Append user's message to messages list
        messages.append({"role": "user", "content": user_message})
        
        response = openai.ChatCompletion.create(
            model="gpt-3.5-turbo",
            messages=messages,
            max_tokens=450,
            temperature=0.2,
            top_p=1,
            frequency_penalty=0,
            presence_penalty=0
        )
        
        # Extract the response text
        output_text = response['choices'][0]['message']['content'].strip()
        
        # 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):
    """
    Process a question, find relevant information, and generate a response.
    """
    if question == "":
        return "Welcome to HealthHub! Ask me anything about menstrual health, mental health, and COVID-19."    
    relevant_segment = find_relevant_segment(question, segments)
    if not relevant_segment:
        return "Could not find specific information. Please refine your question."
    response = generate_response(question, relevant_segment)
    return response

# Define the welcome message and specific topics the chatbot can provide information about
welcome_message = """
# 💊 Welcome to HealthHub!

## Your AI-driven assistant for all GenZ health queries based on Menstrual Health, COVID 19, and Mental Health. Created by Diya, Jayna, and Kibali of the 2024 Kode With Klossy New York City Camp. 



### Feel Free to ask me anything from the topics below!  
"""

Covid19 = """
### Covid 19
- Vaccines
- Affected Groups
- Symptoms
"""

MenstrualHealth = """
### Menstrual Health
- Period Poverty
- Stigmas
- Cycle
"""

MentalHealth = """
### Mental Health
- Therapist
- Anxiety
- Depression
"""
stats_facts = ['25% women struggle to afford or access period products',
'69% of women have to use unsuitable alternate products',
'42% girls miss their schools or work due to periods', '25% women encounter difficulties accessing hygiene facilities',
'Gen Z experiences mental health issues at twice the rate of adults, with anxiety and depression being most common.',    
'Over 50% of Gen Z women report experiencing heavy menstrual bleeding, impacting daily life and mental health.',
'COVID-19 pandemic exacerbated mental health challenges for Gen Z, with anxiety and depression rates increasing by 25%.',
'Menstrual health education is lacking for Gen Z, with only 33% of schools providing comprehensive information.',
'Period poverty affects millions of Gen Z females, hindering education and overall well-being.',
'Social media use correlates with increased anxiety and depression in Gen Z, with 70% spending 3+ hours daily online.',
'Gen Z with mental health disorders are less likely to graduate high school or attend college compared to peers.',
'Access to mental health services is limited for Gen Z, with long wait times and affordability barriers.',
'Stigma surrounding mental health prevents many Gen Z individuals from seeking help, worsening outcomes.',
'Early intervention for mental health issues in Gen Z leads to better long-term outcomes, reducing disability and costs.']
stat = random.choice(stats_facts)

fun_fact = """
# Fun Fact
"""

# Setup the Gradio Blocks interface with custom layout components
with gr.Blocks(theme=theme) as demo:
    gr.Image("Beige Feminine Personal LinkedIn Banner.jpg", show_label = False, show_share_button = False, show_download_button = False, width = 1500, height = 300)
    gr.Markdown(welcome_message)  # Display the formatted welcome message
    with gr.Row():
        with gr.Column():
            gr.Markdown(Covid19)  # Show the topics on the left side
        with gr.Column():
            gr.Markdown(MenstrualHealth)
        with gr.Column():
            gr.Markdown(MentalHealth)
        with gr.Column():

            question = gr.Textbox(label="Your question", placeholder="What do you want to ask about?")
            answer = gr.Textbox(label="HealthHub Response", placeholder="HealthHub will respond here...", interactive=False, lines=10)
            submit_button = gr.Button("Submit")
            submit_button.click(fn=query_model, inputs=question, outputs=answer)
    with gr.Row():
        with gr.Column():
            gr.Markdown(fun_fact)
            gr.Markdown(stat)
        
    with gr.Row():
            big_block = gr.HTML("<button><a href='https://www.cdc.gov/mentalhealth/tools-resources/index.htm'>Mental Health Resources</a></button>")
            big_block = gr.HTML("<button><a href='https://www.humanrightscareers.com/issues/menstrual-justice-organizations-working-for-health-equity/'>Menstrual Health Resources</a></button>")
            big_block = gr.HTML("<button><a href='https://www.covid.gov/'>COVID-19 Resources</a></button>")
# Launch the Gradio app to allow user interaction
demo.launch(share=True)