Added second dataset/checkboxes

app.py

@@ -12,8 +12,14 @@ with open("quentins_knowledge.txt", "r", encoding="utf-8") as file:
   # Read the entire contents of the file and store it in a variable
   quentins_knowledge = file.read()
 
+#SECOND FEATURE
+with open("quentins_alt_knowledge.txt", "r", encoding="utf-8") as file:
+  # Read the entire contents of the file and store it in a variable
+  quentins_alt_knowledge = file.read()
+
 # Print the text below
 print(quentins_knowledge)
+print(quentins_alt_knowledge)
 
 #STEP 3 FROM SEMANTIC SEARCH
 def preprocess_text(text):
@@ -44,6 +50,9 @@ def preprocess_text(text):
 # Call the preprocess_text function and store the result in a cleaned_chunks variable
 cleaned_chunks = preprocess_text(quentins_knowledge)
 
+#SECOND FEATURE
+cleaned_alt_chunks = preprocess_text(quentins_alt_knowledge)
+
 #STEP 4 FROM SEMANTIC SEARCH
 # Load the pre-trained embedding model that converts text to vectors
 model = SentenceTransformer('all-MiniLM-L6-v2')
@@ -64,6 +73,9 @@ def create_embeddings(text_chunks):
 # Call the create_embeddings function and store the result in a new chunk_embeddings variable
 chunk_embeddings = create_embeddings(cleaned_chunks)
 
+#SECOND FEATURE
+alt_chunk_embeddings = create_embeddings(cleaned_alt_chunks)
+
 #STEP 5 FROM SEMANTIC SEARCH
 # 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):
@@ -99,18 +111,15 @@ def get_top_chunks(query, chunk_embeddings, text_chunks):
   # Return the list of most relevant chunks
   return top_chunks
 
-#STEP 6 FROM SEMANTIC SEARCH
-# Call the get_top_chunks function with the original query
-top_results = get_top_chunks("How does water get into the sky?", chunk_embeddings, cleaned_chunks)
-
-# Print the top results
-print(top_results)
-
 client = InferenceClient("google/gemma-3-27b-it")
 
 def respond(message, history, name, mood):
-    duck_chunks = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
-    print(duck_chunks)
+    if quentin_topic == "Self Help":
+        duck_chunks = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
+        print(duck_chunks)
+    elif quentin_topic == "Duck Facts":
+        duck_chunks = get_top_chunks(message, chunk_embeddings, cleaned_alt_chunks)
+        print(duck_chunks)
     duck_info = "\n".join(duck_chunks)
     messages = [{"role": "system", "content": f"You are an extremely {mood} chatbot named Quentin. You are a rubber duck, with strong human emotions who helps the user with their problem. You talk to the user, whose name is {name}, in a way that reflects your {mood} mood. Make sure to use duck-themed references in your responses. Refer to the user by name as much as possible. Base your response on the provided context: {duck_info}. Always end your response with a brief, punchy tagline."}]
 
@@ -161,6 +170,7 @@ with gr.Blocks(theme=gr.themes.Citrus(
 )) as chatbot:
         with gr.Row(scale=1):
             gr.Image("ask_quentin_banner.jpg", show_label = False, show_share_button = False, show_download_button = False)
+            quentin_topic = gr.CheckboxGroup(["Self Help", "Duck Facts"], label="What do you want help with?")
         with gr.Row(scale=3):
             with gr.Column(scale=1):
                 gr.Image("Quentin.png", show_label = False, show_share_button = False, show_download_button = False)