app.py

import gradio as gr
from sentence_transformers import SentenceTransformer, util
import openai
import os
import random
#import transformers
#from transformers import pipeline

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

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

system_message = "You are a friendly AI chatbot specialized in providing information on AI usage, helpful tools, and teaching users about AI."
# 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 AI information.
    """
    try:
        user_message = f"Here's the information on AI: {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=350,
            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})
        fun_int=random.randint(0,11)
        fun_facts=["Young Einstein didn't talk until much later in his childhood.","Einstein had larger-than-average perietal lobes.","Einstein was a talented violinist","Einstein's brain was preserved after his death!","Einstein started as a teacher, but couldn't find a job.","Einstein's famous equation E=mc² was announced in 1905.","Einstein won The Nobel Prize in Physics in 1921","Einstien did not wear socks!","Einstein loved sailing.",'Einstein once said -"If you can not explain it simply, you don not understand it well enough."','Einstein once said- "Logic will get you from A to B. Imagination will get you anywhere."']
        output_text=output_text+"\n\n Here is a fun fact about Albert Einstein!: " + fun_facts[fun_int-1]
        ai_int=random.randint(0,10)
        ai_helpers=["https://chatgpt.com/ - An AI chatbot","https://www.grammarly.com/ - Help with grammar and writing!","https://www.any.do/ - Creates a to do list to help you get your tasks completed!","https://scheduler.ai/- AI optimizes your schedule and works around pre-scheduled deadlines","ChatGPT Data Analyst - Helps you visualize and analize your data","ChatGPT Logo creator - Helps to create professional logos for companies or brands","ScholarGPT - Enhances your reaserch capabilities","ChatGPT's Math solver","Tutor Me by Khan Academy","Travel Guide by capchair - helps find destinations, plan trips, and manage budgets"]
        output_text=output_text+"\n\n Here is a helpful chatbot tool for you!: "+ ai_helpers[ai_int-1]
        return output_text

        # Create pipeline for text generation with confidence scores
        #generator = pipeline("text-davinci-003", device=0)  # Adjust device if needed

        # Generate response and get confidence score
        #response = generator(query=f"Here's the information on AI: {relevant_segment} {user_query}", max_length=150, temperature=0.2, top_p=1)
        #generated_text = response[0]['generated_text'].strip()
        #confidence_score = response[0]['score']

        #return generated_text, confidence_score, 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 AI-nstein! Ask me anything about AI ML, and helpful tools you may want to use!"
    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

link = "https://chatgpt.com/"
links=["https://chatgpt.com/","https://www.grammarly.com/","https://www.any.do/","https://scheduler.ai/","https://huggingface.co/spaces/MukBot/MukBot","https://huggingface.co/spaces/CalmConnect/calm-connect","https://chatgpt.com/g/g-3DGi2iLag-scholargpt","https://chatgpt.com/g/g-E7eSRUHy6-travel-guide"]       
link = random.choice(links)

link_gen =f'''
    {link}
'''

#def get_random_link():
    #link = random.choice(links)
    #return link

#def create_random_link():
    #random_link = get_random_link()
    #return f'<a href="{random_link}" target="_blank"><img src = "einy.png" style="width:500px;"></a>'
    #return f'<a href="{random_link}" target="_blank">I Feel Tech-y!</a>'
    
# Define the welcome message and specific topics the chatbot can provide information about
welcome_message = """
## Albot AI-nstein is your AI-driven assistant, for all things AI.
"""
topicList = """
### _Ask me anything from the topics below!_
### I give you a fun chatbot and an Einstein fact with every answer.
"""

topics1 = """
    \n- AI Usage
    \n- AI Safety
    \n- How AI Works

    """
topics2 = """
    \n- Basics of AI
    \n- Fun Facts about AI
    \n- Examples of AI

    """
headline="""
# **_Welcome to AI-nstein!_**
"""

summary="""
### AI-nstein strives to endow the young (and the less young) with basic knowledge on artificial intelligence! \nWe want youngsters to be comfortable with and knowledgable about AI, because it is an essential part of our future, and not to fear the unknown. 
### Einstein's Theory of Relativity states that what is observed depends on the observer's position. \nAI-nstein will provide users with various viewpoints to further their understanding of AI, and bring more light to this new technology.
      
"""

#def display_image():
    #return "https://i.giphy.com/media/v1.Y2lkPTc5MGI3NjExZzdqMnkzcWpjbGhmM3hzcXp0MGpuaTF5djR4bjBxM3Biam5zbzNnMCZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9cw/GxMnTi3hV3qaIgbgQL/giphy.gif"
    #return "https://cdn-uploads.huggingface.co/production/uploads/6668622b72b61ba78fe7d4bb/PkWjNxvGm9MOqGkZdiT4e.png"
theme = gr.themes.Monochrome(
).set(
    body_text_color='#054A91', #TEXT COLOR
    body_text_color_dark='#000000',
    background_fill_primary='#FBCF95',  # BACKGROUND
    background_fill_primary_dark='#81A4CD',
    background_fill_secondary='#884e4c',  # BUTTON HOVER
    background_fill_secondary_dark='#EDDEC0', #LOADING BAR
    border_color_accent='#EDDEC0',
    border_color_accent_dark='#EDDEC0',
    border_color_accent_subdued='#EDDEC0',
    border_color_primary='#F17300',
    block_border_color='#F17300',
    button_primary_background_fill='#054A91',
    button_primary_background_fill_dark='#884e4c'
    
)

# Setup the Gradio Blocks interface with custom layout components
with gr.Blocks(theme=theme) as demo:
     with gr.Row(equal_height=True):
        with gr.Column():
            gr.Image("ally.png", container = False, show_share_button = False, show_download_button = False, label="output", show_label=True, elem_id="output_image", scale=0, width=500)
            gr.Markdown(welcome_message)  # Display the formatted welcome message
            with gr.Row():
                with gr.Column():
                    gr.Markdown(topicList)
                    with gr.Row(equal_height=True):
                        gr.Markdown(topics1)  # Show the topics on the left side
                        gr.Markdown(topics2)
            gr.Markdown(summary)
        with gr.Row():
            with gr.Column():
                gr.Markdown("      ")
                gr.Markdown("      ")                
                gr.Markdown(headline)
                question = gr.Textbox(label="Your question:", placeholder="What do you want to ask about?")
                submit_button = gr.Button("Submit!")
                answer = gr.Textbox(label="AI-nswer:", placeholder="Hello, World! \nAsk me anything about AI ML, and helpful tools you may want to use!", interactive=False, lines=10)
                submit_button.click(fn=query_model, inputs=question, outputs=answer)
                gr.Image("einy.png", container = False, show_share_button = False, show_download_button = False, label="output", show_label=True, elem_id="output_image", scale=0, width=500)
                gr.Markdown("### Look what Albot found:")
                gr.Markdown(link_gen)
                #portal_button = gr.Button("What examples of AI are there?")
                #portal_button.click(fn=get_random_link, inputs=[], outputs=link)
                #interface = gr.Interface(fn=create_random_link, inputs=[], outputs="html")
                #interface.launch()

                #portal_button = gr.Image("einy.png", container = False, show_share_button = False, show_download_button = False, label="output", show_label=True, elem_id="output_image", scale=0, width=500)
                #portal_button.click(fn=create_random_link, inputs=[], outputs="html")

            #def display_response(question):
                #response, confidence_score = query_model(question)
                #answer.value = f"Response: {response}\nConfidence Score: {confidence_score:.2f}"  

            #submit_button.click(fn=display_response, inputs=question, outputs=None)
            

            
            

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