Upload 4 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+trip_index.faiss filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,14 +1,14 @@
 ---
-title: TravelAdviser
-emoji: 😻
-colorFrom: purple
-colorTo: purple
+title: TripPlanner
+emoji: ✈️
+colorFrom: blue
+colorTo: green
 sdk: gradio
-sdk_version: 5.42.0
+sdk_version: 4.8.0
 app_file: app.py
 pinned: false
 license: mit
-short_description: help in travel planning
+short_description: 'Your personal AI trip planner! ✈️'
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+An example trip recommendation chatbot using [Gradio](https://gradio.app), the `datasets` and `sentence-transformers` libraries, and FAISS for similarity search.

app.py

@@ -0,0 +1,144 @@
+import gradio as gr
+from datasets import load_dataset
+from sentence_transformers import SentenceTransformer, util
+import faiss
+import numpy as np
+from transformers import pipeline
+import time
+
+# --- 1. DATA LOADING AND INITIALIZATION ---
+print("===== Application Startup =====")
+start_time = time.time()
+
+# Load the travel dataset and limit to the first 20,000 rows (same as index)
+print("Loading TravelPlanner dataset...")
+dataset = load_dataset("osunlp/TravelPlanner", "test")
+print("Dataset ready.")
+
+# --- 2. EMBEDDING AND RECOMMENDATION ENGINE ---
+print("Loading embedding model...")
+model_name = "all-mpnet-base-v2"
+embedding_model = SentenceTransformer(f"sentence-transformers/{model_name}")
+
+index_file = "trip_index.faiss"
+
+print(f"Loading FAISS index from {index_file}...")
+
+try:
+    index = faiss.read_index(index_file)
+    print(f"Index is ready. Total vectors in index: {index.ntotal}")
+except RuntimeError:
+    print(f"Error: FAISS index file '{index_file}' not found.")
+    print("Please run the `build_index.py` script first to create the index.")
+    exit()
+
+
+# --- 3. SYNTHETIC GENERATION ---
+print("Loading generative model...")
+generator = pipeline('text-generation', model='gpt2')
+
+def get_recommendations_and_generate(query_text, k=3):
+    # 1. Get Recommendations from existing data
+    query_vector = embedding_model.encode([query_text])
+    query_vector = np.array(query_vector, dtype=np.float32)
+    distances, indices = index.search(query_vector, k)
+    
+    results = []
+    for idx_numpy in indices[0]:
+        idx = int(idx_numpy)
+        trip_plan = {
+            "title": dataset[idx]['title'],
+            "plan": dataset[idx]['plan']
+        }
+        results.append(trip_plan)
+        
+    while len(results) < 3:
+        results.append({"title": "No trip plan found", "plan": ""})
+        
+    # 2. Create a prompt for the generative model
+    prompt = f"Write a complete travel plan that includes a title and a day-by-day itinerary. The trip must be about: {query_text}."
+    
+    # 3. Generate 10 new, creative trip ideas
+    print("Generating 10 synthetic trip ideas...")
+    generated_outputs = generator(
+        prompt, 
+        max_new_tokens=250,  # Increased tokens for more detailed plans
+        num_return_sequences=10, 
+        pad_token_id=50256
+    )
+    
+    # 4. Find the best trip out of the 10 generated
+    print("Finding the most relevant generated trip...")
+    generated_texts = [output['generated_text'].replace(prompt, "").strip() for output in generated_outputs]
+    
+    # Embed all 10 generated texts
+    generated_embeddings = embedding_model.encode(generated_texts)
+    
+    # Calculate cosine similarity between the user's query and each generated text
+    similarities = util.cos_sim(query_vector, generated_embeddings)
+    
+    # Find the index of the most similar generated trip
+    best_recipe_index = np.argmax(similarities)
+    best_generated_trip = generated_texts[best_recipe_index]
+
+    return results[0], results[1], results[2], best_generated_trip
+
+# --- 4. GRADIO USER INTERFACE ---
+def format_trip_plan(trip):
+    # Formats the recommended trips with markdown
+    if not trip:
+        return "### No similar trip plan found."
+    return f"### {trip['day']}-days trip to {trip['dest'].upper()}\n**Suggested Plan:**\n{trip['reference_information']}"
+
+def format_generated_trip(trip_text):
+    return trip_text
+
+def trip_planner_wizard(destination, days):
+    # Combine user inputs into a single query for processing
+    days = int(days) # Ensure days is an integer for the f-string
+    query_text = f"a {days}-day trip to {destination}"
+    rec1, rec2, rec3, gen_rec_text = get_recommendations_and_generate(query_text)
+    return format_trip_plan(rec1), format_trip_plan(rec2), format_trip_plan(rec3), format_generated_trip(gen_rec_text)
+
+end_time = time.time()
+print(f"Models and data loaded in {end_time - start_time:.2f} seconds.")
+
+# Gradio Interface
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    gr.Markdown("# ✈️ TripPlanner AI")
+    gr.Markdown("Enter your destination and desired trip length, and get plan recommendations plus a new AI-generated idea!")
+    
+    with gr.Row():
+        destination_input = gr.Textbox(label="Destination", placeholder="e.g., Paris")
+        days_input = gr.Number(label="Number of Days", value=3)
+        
+    with gr.Row():
+        submit_btn = gr.Button("Get Trip Plans", variant="primary")
+
+    with gr.Row():
+        with gr.Column(scale=2):
+            gr.Markdown("### Recommended Trip Plans from Dataset")
+            output_rec1 = gr.Markdown()
+            output_rec2 = gr.Markdown()
+            output_rec3 = gr.Markdown()
+        with gr.Column(scale=1):
+            gr.Markdown("### ✨ New AI-Generated Idea")
+            output_gen = gr.Textbox(label="AI Generated Trip Plan", lines=20, interactive=False)
+
+    submit_btn.click(
+        fn=trip_planner_wizard,
+        inputs=[destination_input, days_input],
+        outputs=[output_rec1, output_rec2, output_rec3, output_gen]
+    )
+    
+    gr.Examples(
+        examples=[
+            ["Paris", 3],
+            ["Orlando", 7],
+            ["Tokyo", 5],
+            ["the Greek Islands", 10]
+        ],
+        inputs=[destination_input, days_input]
+    )
+
+demo.launch()

requirements.txt

@@ -0,0 +1,8 @@
+gradio==4.31.5
+datasets==2.19.1
+sentence-transformers==2.7.0
+faiss-cpu==1.8.0
+transformers==4.41.2
+torch==2.3.1
+pyarrow==16.1.0
+huggingface-hub==0.23.3

trip_index.faiss

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c8d2a2eb720f78a9be02358e0d87e287766a257179553b784831ba7b207c875
+size 3072045