Update app.py

app.py

@@ -1,66 +1,70 @@
 import gradio as gr
 import spaces
-from huggingface_hub import InferenceClient
+from transformers import RagTokenizer, RagRetriever, RagSequenceForGeneration
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# ——— Configuration ———
+MODEL_NAME   = "facebook/rag-sequence-nq"
+DATASET_NAME = "username/mealplan-chunks"
 
+# ——— Initialize RAG ———
+tokenizer = RagTokenizer.from_pretrained(MODEL_NAME)
+retriever = RagRetriever.from_pretrained(
+    MODEL_NAME,
+    index_name="exact",
+    use_dummy_dataset=False,
+    dataset_name=DATASET_NAME
+)
+model = RagSequenceForGeneration.from_pretrained(MODEL_NAME, retriever=retriever)
+
+# ——— Core chat callback ———
 @spaces.GPU
 def respond(
-    message,
+    message: str,
     history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
+    goal: str,
+    diet: list[str],
+    meals: int,
+    avoid: str,
+    weeks: str,
 ):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
+    # parse avoidances
+    avoid_list = [a.strip() for a in avoid.split(",") if a.strip()]
+    # build prefs string
+    prefs = (
+        f"Goal={goal}; "
+        f"Diet={','.join(diet)}; "
+        f"Meals={meals}/day; "
+        f"Avoid={','.join(avoid_list)}; "
+        f"Duration={weeks}"
+    )
+    # system guardrail + prefs + question
+    prompt = (
+        "SYSTEM: Only answer using the provided CONTEXT. "
+        "If it’s not there, say “I’m sorry, I don’t know.”\n"
+        f"PREFS: {prefs}\n"
+        f"Q: {message}\n"
+    )
+    # generate
+    inputs  = tokenizer([prompt], return_tensors="pt")
+    outputs = model.generate(**inputs, num_beams=2, max_new_tokens=200)
+    answer  = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]
 
-    response = ""
+    history = history or []
+    history.append((message, answer))
+    return history
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# ——— Build the Chat Interface ———
+# preference controls
+goal  = gr.Dropdown(["Lose weight","Bulk","Maintain"], value="Lose weight", label="Goal")
+diet  = gr.CheckboxGroup(["Omnivore","Vegetarian","Vegan","Keto","Paleo","Low-Carb"], label="Diet Style")
+meals = gr.Slider(1, 6, value=3, step=1, label="Meals per day")
+avoid = gr.Textbox(placeholder="e.g. Gluten, Dairy, Nuts, Eggs, Soy…", label="Avoidances (comma-separated)")
+weeks = gr.Dropdown(["1 week","2 weeks","3 weeks","4 weeks"], value="1 week", label="Plan Length")
 
 demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+    fn=respond,
+    additional_inputs=[goal, diet, meals, avoid, weeks],
 )
 
-
 if __name__ == "__main__":
     demo.launch()