create a chatbot

app.py

@@ -1,11 +1,14 @@
 import gradio as gr
-
 from huggingface_hub import InferenceClient
 import os
-client = InferenceClient(model="Qwen/Qwen2.5-7B-Instruct", token=os.environ.get("HF"))
 from sentence_transformers import SentenceTransformer
 import torch
 
+client = InferenceClient(
+    model="Qwen/Qwen2.5-7B-Instruct",
+    token=os.environ.get("HF")
+)
+
 with open("knowledge.txt", "r", encoding="utf-8") as file:
     knowledge_text = file.read()
 
@@ -13,90 +16,107 @@ def preprocess_text(text):
     cleaned_text = text.strip()
     chunks = cleaned_text.split("\n")
     cleaned_chunks = []
+
     for chunk in chunks:
         stripped_chunk = chunk.strip()
         if len(stripped_chunk) > 0:
             cleaned_chunks.append(stripped_chunk)
+
     return cleaned_chunks
+
 cleaned_chunks = preprocess_text(knowledge_text)
 
-model = SentenceTransformer('all-MiniLM-L6-v2')
+model = SentenceTransformer("all-MiniLM-L6-v2")
 
 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 cleaned_chunks list
-  # 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
+    chunk_embeddings = model.encode(text_chunks, convert_to_tensor=True)
+    return chunk_embeddings
 
+chunk_embeddings = create_embeddings(cleaned_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()
+    query_embedding = model.encode(query, convert_to_tensor=True)
 
-  # Normalize all chunk embeddings to unit length for consistent comparison
-  chunk_embeddings_normalized = chunk_embeddings / chunk_embeddings.norm(dim=1, keepdim=True)
+    query_embedding_normalized = query_embedding / query_embedding.norm()
+    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
+    similarities = torch.matmul(chunk_embeddings_normalized, query_embedding_normalized)
 
-  # Find the indices of the 3 chunks with highest similarity scores
-  top_indices = torch.topk(similarities, k=3).indices
+    top_indices = torch.topk(similarities, k=3).indices
 
+    top_chunks = []
 
-  # Create an empty list to store the most relevant chunks
-  top_chunks = []
+    for i in top_indices:
+        chunk = text_chunks[i.item()]
+        top_chunks.append(chunk)
 
-  # Loop through the top indices and retrieve the corresponding text chunks
-  # This is only one way scholars may write this, but there are other ways!
-  for i in top_indices:
-    chunk = text_chunks[i]
-    top_chunks.append(chunk)
+    return top_chunks
 
+def respond(message, history):
+    messages = [
+        {
+            "role": "system",
+            "content": "You are an emotional support chatbot. Do not talk about anything else other than mental health and helping the user. Make sure the user feels comfortable."
+        }
+    ]
 
-  # Return the list of most relevant chunks
-  return top_chunks
+    if history:
+        messages.extend(history)
 
+    messages.append({"role": "user", "content": message})
 
+    response = ""
 
-def respond(message, history):
-    messages = [{"role": "system",
-                 "content":"You are an emotional support chatbot. You would not take about anything else other than mental health and helping the users. You need to make sure the user is comfortable."
-                }] 
-    if history:
-        messages.extend(history)
-    messages.append({"role":"user",
-                 "content":message
-                })
-    response = " "
-    for msg in client.chat_completion(messages, max_tokens = 1000, temperature = 1, top_p = 0.5, stream = True):
+    for msg in client.chat_completion(
+        messages,
+        max_tokens=1000,
+        temperature=1,
+        top_p=0.5,
+        stream=True
+    ):
         token = msg.choices[0].delta.content
-        response += token 
-        yield response 
-
+        if token:
+            response += token
+            yield response
 
-#EMMA'S PRACTICE EDITS#
 about_text = """
 ## About this bot  
-    Welcome to Mind Matters, an online resource that reminds *your that your mind matters*
-    Disclaimer; Mind Matters should not be used as an alternative to seeking professional help. I am simply a support tool.
+
+Welcome to Mind Matters, an online resource that reminds you that your mind matters.
+
+Disclaimer: Mind Matters should not be used as an alternative to seeking professional help. It is simply a support tool.
 """
 
-with gr.Blocks() as demo:
+custom_theme = gr.themes.Soft(
+    primary_hue="pink",
+    secondary_hue="fuchsia",
+    neutral_hue="gray",
+    spacing_size="lg",
+    radius_size="lg",
+    text_size="lg"
+)
+
+with gr.Blocks(theme=custom_theme) as demo:
     with gr.Row():
         with gr.Column(scale=1):
             gr.Markdown(about_text)
 
         with gr.Column(scale=2):
-            gr.ChatInterface(fn=respond, title = "Mind Matters", description = "Always here to help", editable = True)
+            gr.ChatInterface(
+                fn=respond,
+                title="Mind Matters",
+                description="Always here to help",
+                editable=True
+            )
 
 demo.launch()
 
 
-chatbot = gr.ChatInterface(fn=respond, title = "Mind Matters", description = "Always here to help", editable = True)
-chatbot.launch()
+custom_theme = gr.themes.Soft(
+    primary_hue="pink",
+    secondary_hue="fuchsia",
+    neutral_hue="gray"
+)
+
+with gr.Blocks(theme=custom_theme) as demo:
+