Create app.py

app.py

@@ -1,70 +1,91 @@
+import json
+import difflib
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    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(token=hf_token.token, model="openai/gpt-oss-20b")
-
-    messages = [{"role": "system", "content": system_message}]
-
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = 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
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    type="messages",
-    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)",
-        ),
-    ],
-)
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
 
+# -------------------------------------------------
+# Load Knowledge Base
+# -------------------------------------------------
+with open("destinations.json", "r", encoding="utf-8") as f:
+    DESTS = json.load(f)
+
+DEST_NAMES = [d["name"] for d in DESTS]
+
+def retrieve_destinations(query, n=3):
+    query = query.lower()
+    exact = [
+        d for d in DESTS
+        if d["name"].lower() in query or any(tag.lower() in query for tag in d.get("tags", []))
+    ]
+
+    if exact:
+        return exact[:n]
+
+    matches = difflib.get_close_matches(query, DEST_NAMES, n=n, cutoff=0.4)
+    return [d for d in DESTS if d["name"] in matches]
+
+
+def build_prompt(user_message: str, retrieved):
+    kb_text = ""
+    if retrieved:
+        details = []
+        for d in retrieved:
+            details.append(
+                f"{d['name']} — {d['summary']}\n"
+                f"Top attractions: {', '.join(d['top_attractions'])}\n"
+                f"Best months: {d['best_months']}"
+            )
+        kb_text = "Destination Knowledge:\n" + "\n\n".join(details) + "\n\n"
+
+    return (
+        f"{kb_text}"
+        f"User Query: \"{user_message}\"\n"
+        "You are a helpful travel guide. Provide:\n"
+        "- Best suited destination(s)\n"
+        "- Why it matches the user's need\n"
+        "- Best time to visit\n"
+        "- 2–3 activities to do\n"
+        "- Short travel/safety tips"
+    )
+
+
+# -------------------------------------------------
+# Load CPU-friendly Model
+# -------------------------------------------------
+MODEL = "facebook/blenderbot-400M-distill"
+
+tokenizer = AutoTokenizer.from_pretrained(MODEL)
+model = AutoModelForSeq2SeqLM.from_pretrained(MODEL)   # CPU ONLY → no .to('cuda')
+
+
+def chatbot_reply(message, history):
+    retrieved = retrieve_destinations(message)
+    prompt = build_prompt(message, retrieved)
+
+    inputs = tokenizer(prompt, return_tensors="pt", truncation=True)
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=200,
+        do_sample=True,
+        top_p=0.9,
+        temperature=0.6
+    )
+    reply = tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+    history.append((message, reply))
+    return history, history
+
+
+# -------------------------------------------------
+# Gradio UI
+# -------------------------------------------------
 with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
+    gr.Markdown("# 🌍 Travel Recommendation Chatbot (CPU Only)")
+
+    chatbox = gr.Chatbot()
+    txt = gr.Textbox(placeholder="Ask: 'Where should I travel in winter for beaches?'", label="Your Message")
+    state = gr.State([])
 
+    txt.submit(chatbot_reply, [txt, state], [chatbox, state])
 
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()