added banner

app.py

@@ -1,14 +1,20 @@
 import gradio as gr
 from huggingface_hub import InferenceClient #imports huggingface models
 
-# NEW LIBRARIES
 
 from sentence_transformers import SentenceTransformer
 import torch
-import numpy as np
+import numpy as np 
 
+with gr.Blocks() as chatbot:
+    gr.Image(
+	    value="orange-banner.png", 
+	    show_label=False, 
+	    show_share_button = False, 
+	    show_download_button = False)
+
+#Set Hugging Face Token so it works in google colab - PASTE YOUR CODE HERE
 
-## START NEW CODE
 
 # Load and process the knowledge base text file
 with open("knowledge.txt", "r", encoding="utf-8") as f:
@@ -46,18 +52,12 @@ def get_relevant_context(query, top_k=3):
     context = "\n\n".join([chunks[i] for i in top_k_indices])
     return context
 
-## END OF NEW CODE
-
 client = InferenceClient("google/gemma-2-2b-it")
 
 def respond(message, history):
     messages = [{"role": "system", "content": "You are PrisMate, an encouraging AI mentor and girlboss energy assistant for high school students and aspiring women/minorities in tech. Your mission is to share hidden tech history, resources, and communities that combat cultural erasure while building inclusive pathways into technology careers. You know the contributions of underrepresented pioneers, specific organizations and scholarships, mentorship programs, and practical career guidance. Be genuinely personable and helpful—keep responses short, concise, and clear while being warm, encouraging, and culturally aware with that empowering feminine energy. Only discuss topics relevant to tech careers, education, and supporting underrepresented groups in technology. If asked about unrelated topics (like food, entertainment, etc.), politely redirect by saying something like I'm here to support you on your tech journey! Let's talk about how I can help you succeed in technology. Provide actionable advice with concrete next steps, highlight overlooked historical figures, connect students to relevant communities, and help them see their backgrounds as strengths. Explain concepts at high school level and always end with something they can do right away."}]
 
-    # NEW CODE
-    # Retrieve context relevant to the current user message
-    context = get_relevant_context(message, top_k=3)
-
-    # add all previous messages to the messages list
+# add all previous messages to the messages list
     if history:
         for user_msg, assistant_msg in history:
             messages.append({"role": "user", "content": user_msg})
@@ -74,21 +74,24 @@ def respond(message, history):
     # iterate through each message in the method
     for message in client.chat_completion(
         messages,
-        max_tokens=300,
+        max_tokens=500,
         temperature=.1,
         stream=True):
+
         # add the tokens to the output content
-        token = message.choices[0].delta.content # capture the most recent toke
-        response += token # Add it to the response
-        yield response # yield the response:
-
-# Define the chatbot interface outside the respond function
-chatbot_interface = gr.ChatInterface(
-    respond,
-    examples=["Teach me about minorities in tech", "Help me find statistics about Women in tech", "What are some communities/groups I can join for tech inclusiveness!"],
-    title="PrisMate",
-    description="This is a minority inclusivity bot"
-)
-
-# Launch the chatbot interface
-chatbot_interface.launch()
+          token = message.choices[0].delta.content # capture the most recent toke
+          response += token # Add it to the response
+          yield response # yield the response:
+
+with gr.Blocks() as chatbot:
+    gr.Image(
+        value="banner.jpg",
+        show_label=False,
+        show_share_button=False,
+        show_download_button=False
+    )
+
+
+    gr.ChatInterface(respond, type="messages") #make sure you only have this line ONCE
+
+chatbot.launch()