app.py

from huggingface_hub import InferenceClient
# Step 1 from the Semantic Search
from sentence_transformers import SentenceTransformer
import torch
import gradio as gr
import random 
 #beginning of copilot




#end of copilot
# Making requests to the model to generate responses:
client = InferenceClient('Qwen/Qwen2.5-72B-Instruct') 

# ============================================

# Step 2 from the semantic search
# Open the water_cycle.txt file in read mode with UTF-8 encoding
with open("Joy_Scout_info.txt", "r", encoding="utf-8") as file:
  # Read the entire contents of the file and store it in a variable
  joyscout_info_text = file.read()

# Print the text below
print(joyscout_info_text)


# =============================================
# Step 3: 
def preprocess_text(text):
  # Strip extra whitespace from the beginning and the end of the text
  cleaned_text = text.strip()

  # Split the cleaned_text by every newline character (\n)
  chunks = cleaned_text.split("\n")

  # Create an empty list to store cleaned chunks
  cleaned_chunks = []

  # Write your for-in loop below to clean each chunk and add it to the cleaned_chunks list
  for chunk in chunks:
      cleaned_chunks.append(chunk)

  # Print cleaned_chunks
  print(cleaned_chunks)

  # Print the length of cleaned_chunks
  print(len(cleaned_chunks))

  # Return the cleaned_chunks
  return cleaned_chunks

# Call the preprocess_text function and store the result in a cleaned_chunks variable
cleaned_chunks = preprocess_text(joyscout_info_text) # Complete this line

# Load the pre-trained embedding model that converts text to vectors
model = SentenceTransformer('all-MiniLM-L6-v2')

# ============================================

# Step 4: 
def create_embeddings(text_chunks):
  # Convert each text chunk into a vector embedding and store as a tensor
  chunk_embeddings = model.encode(text_chunks, convert_to_tensor=True) # Replace ... with the text_chunks list

  # Print the chunk embeddings
  print(chunk_embeddings)

  # Print the shape of chunk_embeddings
  print(len(chunk_embeddings))

  # Return the chunk_embeddings
  return chunk_embeddings

# Call the create_embeddings function and store the result in a new chunk_embeddings variable
chunk_embeddings = create_embeddings(cleaned_chunks) # Complete this line


# =====================================

# Step 5: 
# Define a function to find the most relevant text chunks for a given query, chunk_embeddings, and text_chunks
def get_top_chunks(query, chunk_embeddings, text_chunks):
  # Convert the query text into a vector embedding
  query_embedding = model.encode(query, convert_to_tensor=True) # Complete this line

  # Normalize the query embedding to unit length for accurate similarity comparison
  query_embedding_normalized = query_embedding / query_embedding.norm()

  # Normalize all chunk embeddings to unit length for consistent comparison
  chunk_embeddings_normalized = chunk_embeddings / chunk_embeddings.norm(dim=1, keepdim=True)

  # Calculate cosine similarity between query and all chunks using matrix multiplication
  similarities = torch.matmul(chunk_embeddings_normalized, query_embedding_normalized) # Complete this line

  # Print the similarities
  print(similarities)

  # Find the indices of the 3 chunks with highest similarity scores
  top_indices = torch.topk(similarities, k=3).indices

  # Print the top indices
  print(top_indices)

  # Create an empty list to store the most relevant chunks
  top_chunks = []
  # Creating an empty list to now store our most SIMILAR indices


  # Loop through the top indices and retrieve the corresponding text chunks
  for index in top_indices: # Looping through where our chunks are currently stored and now appending the most similar to be in our new list
    top_chunks.append(text_chunks[index])
    # List of the actual chunks needs to be created based on the index values that the top indices list consists of

  # Return the list of most relevant chunks
  return top_chunks
# =====================================

# Step 7: Putting data into the dictionary: 


# ======================================
def respond(message, history, hobby_type, age):
    best_chunks = get_top_chunks(message, chunk_embeddings, cleaned_chunks) 
    print(best_chunks)

    str_chunks = "/n".join(best_chunks)
    messages = [{'role':'system', 'content': f'You are a very kind chatbot giving people hobby suggestions to help them spend less time on their electronic devices. Make sure to be specific to {hobby_type} hobbies and make sure they are appropriate for someone who is {age} years old. You answer their questions based on ' + str_chunks + '.'}]
      
    if history:
        messages.extend(history)


    messages.append({'role':'user', 'content': message})
      
    response = client.chat_completion(messages, max_tokens=250, temperature=1.7, top_p=.7)
    # Temp and top_p control randomness
    
    return response['choices'][0]['message']['content'].strip()

with gr.Blocks(theme='earneleh/paris') as chatbot:
    with gr.Row(scale=1):
        with gr.Column(scale=0.7):
            hobby_type = gr.CheckboxGroup(['Crafty + DIY', 'Outdoor','Physical', 'Performance + Stage', 'Animal-related'], label = "What category of hobbies are you interested in?")
            age = gr.Textbox(label = "How old are you", info = "Enter your age", placeholder = "Type a sentence here...")
        gr.Image("ai.png")
    with gr.Row(scale=1):
        gr.ChatInterface(fn=respond, additional_inputs = [hobby_type, age], type= "messages")

chatbot.launch()