comments adjusted for teaching

app.py

@@ -3,7 +3,7 @@ from huggingface_hub import InferenceClient #InferenceClient class
 
 client = InferenceClient("deepseek-ai/DeepSeek-R1-Distill-Qwen-32B") #Create an instance of InferenceClient connected to the Qwen/Qwen2.5-7B-Instruct text-generation model
 #this client will handle making requests to the model to generate responses
-
+#allows us to access the LLM we chose (og: "Qwen/Qwen2.5-7B-Instruct")
 
 def respond(message, history): #function for Gradio to call
 #Gradio passes arguments as parameters: the user's most recent input which is a string ("message"), and "history" which is the list of past messages
@@ -12,25 +12,28 @@ def respond(message, history): #function for Gradio to call
     messages = [
         {"role": "system",
          "content": "You are a friendly chatbot."}
-    ] #dict in list to store messages
+    ] #dictionary in list to store messages
 
     #Add convo history to the messages if there's convo history
+    #adds everytime there's a new message
     if history: 
         messages.extend(history) # adds history to the end of messages list via .extend() method
 
+    #user's side:
     messages.append({"role": "user", "content": message}) #add the current user’s message to the messages list
 
-    # chat completion API call forwarding the messages & other params to model
+    # chat completion API call forwarding the messages (& other params) to model
+    # connect the client to chatbot
     response = client.chat_completion(messages, max_tokens=100, temperature = 2, top_p=0.95) #deepseek R1 recomended temp range: 0.5-0.7
-    return response.choices[0].message.content.strip()
         # max_tokens, to limit the number of tokens that can be  generated
         # temperature, which controls randomness (higher = more random)
         # top_p, an alternative to sampling with temperature
-
+    return response.choices[0].message.content.strip()    
 
 # defining chatbot
 chatbot = gr.ChatInterface(respond, title = "", description = "") #using gradio to quickly build a chatbot UI (w/ convo history & user input)
 # passing fxn into a fxn, passing echo for gradio to call each time the user sends a message
 # Adding parentheses would call the function and pass its return value instead, I didn't include () because I want Gradio to call it later, not right now
 
-chatbot.launch() #launch chatbot
+chatbot.launch(debug=True) #launch chatbot
+#(? not 100% sure how)--debug=True will give us detailed messages if something is wrong so we can debug