Spaces:
Sleeping
Sleeping
File size: 23,387 Bytes
96f51f8 87d17b6 96f51f8 9304981 96f51f8 9304981 96f51f8 9304981 96f51f8 9304981 96f51f8 9304981 96f51f8 9304981 96f51f8 9304981 96f51f8 9304981 96f51f8 12c8cdf 07913c3 9304981 07913c3 96f51f8 b121792 96f51f8 b417f80 961d2c0 96f51f8 b121792 96f51f8 b121792 96f51f8 b121792 96f51f8 b121792 96f51f8 b121792 96f51f8 b121792 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 ce5a0ef 96f51f8 87d17b6 ce5a0ef 87d17b6 ce5a0ef 87d17b6 ce5a0ef 87d17b6 ef0c77d ce5a0ef 87d17b6 ce5a0ef 87d17b6 96f51f8 87d17b6 96f51f8 87d17b6 ce5a0ef 3e4d2df 87d17b6 3e4d2df 96f51f8 3e4d2df 96f51f8 87d17b6 96f51f8 23cb762 96f51f8 23cb762 96f51f8 23cb762 96f51f8 1f9349d 9304981 96f51f8 9304981 96f51f8 3e4d2df ef0c77d 3e4d2df 96f51f8 3e4d2df 96f51f8 3e4d2df 96f51f8 3e4d2df ef0c77d 3e4d2df 96f51f8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 |
import os
os.environ['KMP_DUPLICATE_LIB_OK']='TRUE'
import streamlit as st
import pandas as pd
import torch
import random
from transformers import (
T5ForConditionalGeneration,
T5Tokenizer,
Trainer,
TrainingArguments,
DataCollatorForSeq2Seq
)
from torch.utils.data import Dataset
from datetime import datetime
import numpy as np
from random import choice
import re
class TravelDataset(Dataset):
def __init__(self, data, tokenizer, max_length=512):
"""
data: DataFrame with columns ['destination', 'days', 'budget', 'interests', 'travel_plan']
"""
self.tokenizer = tokenizer
self.data = data
self.max_length = max_length
def __len__(self):
return len(self.data)
def __getitem__(self, idx):
row = self.data.iloc[idx]
input_text = self.format_input_text(row)
target_text = row['travel_plan']
# Tokenize inputs
input_encodings = self.tokenizer(
input_text,
max_length=self.max_length,
padding='max_length',
truncation=True,
return_tensors='pt'
)
# Tokenize targets
target_encodings = self.tokenizer(
target_text,
max_length=self.max_length,
padding='max_length',
truncation=True,
return_tensors='pt'
)
return {
'input_ids': input_encodings['input_ids'].squeeze(),
'attention_mask': input_encodings['attention_mask'].squeeze(),
'labels': target_encodings['input_ids'].squeeze()
}
@staticmethod
def format_input_text(row):
return f"Plan a trip to {row['destination']} for {row['days']} days with a {row['budget']} budget. Include activities related to: {row['interests']}"
def create_sample_data():
"""Create sample training data for travel plans ranging from 1 to 14 days"""
destinations = ['Paris', 'Tokyo', 'New York', 'London', 'Rome']
budgets = ['Budget', 'Moderate', 'Luxury']
interests_list = [
'Culture, History',
'Food, Shopping',
'Art, Museums',
'Nature, Adventure',
'Relaxation, Food'
]
# Activity templates for different interests
activities = {
'Culture': ['Visit historical sites', 'Explore local traditions', 'Attend cultural events',
'Visit ancient monuments', 'Experience local festivals'],
'History': ['Tour ancient ruins', 'Visit museums', 'Explore historic districts',
'Join guided history walks', 'Visit heritage sites'],
'Food': ['Try local cuisine', 'Join cooking classes', 'Visit food markets',
'Dine at famous restaurants', 'Food tasting tours'],
'Shopping': ['Browse local markets', 'Visit shopping districts', 'Shop at boutiques',
'Explore artisan shops', 'Visit shopping centers'],
'Art': ['Visit art galleries', 'Attend art exhibitions', 'Join art workshops',
'Visit artist studios', 'Explore street art'],
'Museums': ['Tour famous museums', 'Visit specialty museums', 'Join museum tours',
'Explore art collections', 'Visit cultural institutes'],
'Nature': ['Visit parks', 'Nature walks', 'Explore gardens', 'Visit natural landmarks',
'Outdoor activities'],
'Adventure': ['Join adventure tours', 'Try outdoor sports', 'Explore hidden spots',
'Take scenic hikes', 'Adventure activities'],
'Relaxation': ['Spa treatments', 'Visit peaceful gardens', 'Leisure activities',
'Relaxing sightseeing', 'Peaceful excursions']
}
def generate_daily_plan(day, total_days, interests, budget_level, destination):
"""Generate a single day's plan based on interests and duration"""
interest1, interest2 = [i.strip() for i in interests.split(',')]
# Select activities based on interests
activity1 = choice(activities[interest1])
activity2 = choice(activities[interest2])
if total_days <= 3:
# For short trips, pack more activities per day
return f"Day {day}: {activity1} in the morning. {activity2} in the afternoon/evening. Experience {destination}'s {budget_level.lower()} offerings."
elif total_days <= 7:
# Medium trips have a moderate pace
return f"Day {day}: Focus on {activity1}. Later, enjoy {activity2}."
else:
# Longer trips have a more relaxed pace
return f"Day {day}: {'Start with' if day == 1 else 'Continue with'} {activity1}. Optional: {activity2}."
data = []
for dest in destinations:
for days in range(1, 15): # 1 to 14 days
for budget in budgets:
for interests in interests_list:
# Generate multi-day plan
daily_plans = []
for day in range(1, days + 1):
daily_plan = generate_daily_plan(day, days, interests, budget, dest)
daily_plans.append(daily_plan)
# Combine all days into one plan
full_plan = "\n".join(daily_plans)
data.append({
'destination': dest,
'days': days,
'budget': budget,
'interests': interests,
'travel_plan': full_plan
})
return pd.DataFrame(data)
@st.cache_resource
def load_or_train_model():
"""Load trained model or train new one"""
model_path = "trained_travel_planner"
if os.path.exists(model_path):
try:
model = T5ForConditionalGeneration.from_pretrained(model_path)
tokenizer = T5Tokenizer.from_pretrained(model_path)
if torch.cuda.is_available():
model = model.cuda()
st.success("β Loaded existing model")
return model, tokenizer
except Exception as e:
st.warning("Could not load existing model, will train new one")
st.error(f"Error loading trained model: {str(e)}")
# If no trained model exists or loading fails, train new model
return train_model()
def train_model():
"""Train the T5 model on travel planning data"""
try:
# Initialize model and tokenizer
tokenizer = T5Tokenizer.from_pretrained('t5-base', legacy=False)
model = T5ForConditionalGeneration.from_pretrained('t5-base')
# Create or load training data
if os.path.exists('travel_data.csv'):
data = pd.read_csv('travel_data.csv')
else:
data = create_sample_data()
data.to_csv('travel_data.csv', index=False)
# Split data into train and validation
train_size = int(0.8 * len(data))
train_data = data[:train_size]
val_data = data[train_size:]
# Create datasets
train_dataset = TravelDataset(train_data, tokenizer)
val_dataset = TravelDataset(val_data, tokenizer)
# Training arguments
training_args = TrainingArguments(
output_dir=f"./travel_planner_model_{datetime.now().strftime('%Y%m%d_%H%M%S')}",
num_train_epochs=3,
per_device_train_batch_size=4,
per_device_eval_batch_size=4,
warmup_steps=500,
weight_decay=0.01,
logging_dir="./logs",
logging_steps=10,
evaluation_strategy="steps",
eval_steps=50,
save_steps=100,
load_best_model_at_end=True,
)
# Data collator
data_collator = DataCollatorForSeq2Seq(
tokenizer=tokenizer,
model=model,
padding=True
)
# Initialize trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_dataset,
eval_dataset=val_dataset,
data_collator=data_collator,
)
# Train the model
trainer.train()
# Save the model and tokenizer
model_path = "./trained_travel_planner"
model.save_pretrained(model_path)
tokenizer.save_pretrained(model_path)
return model, tokenizer
except Exception as e:
st.error(f"Error during model training: {str(e)}")
return None, None
def generate_travel_plan(destination, days, interests, budget, model, tokenizer):
"""Generate a travel plan using the trained model with enhanced features"""
try:
# Format interests into a string, limit to top 3 if more are provided
interests = interests[:3] # Limit to top 3 interests for better results
interests_str = ', '.join(interests)
# Format input prompt to match training data format
prompt = f"Plan a trip to {destination} for {days} days with a {budget} budget. Include activities related to: {interests_str}"
# Tokenize input with padding
inputs = tokenizer(
prompt,
return_tensors="pt",
max_length=512,
padding="max_length",
truncation=True
)
# Move inputs to GPU if available
if torch.cuda.is_available():
inputs = {k: v.cuda() for k, v in inputs.items()}
model = model.cuda()
# Generate output with carefully tuned parameters
outputs = model.generate(
**inputs,
max_length=512,
min_length=100, # Ensure reasonable length output
num_beams=4, # Beam search for better quality
no_repeat_ngram_size=3, # Avoid repetition
length_penalty=1.2, # Favor longer sequences
early_stopping=True,
temperature=0.8, # Slightly random but still focused
top_k=50,
top_p=0.9,
do_sample=True,
repetition_penalty=1.2 # Additional repetition avoidance
)
# Decode output
travel_plan = tokenizer.decode(outputs[0], skip_special_tokens=True)
# Handle empty output
if not travel_plan.strip():
raise ValueError("Generated plan is empty")
# Format the plan using the new formatting function
formatted_plan = format_travel_plan(travel_plan, days)
return formatted_plan
except Exception as e:
error_msg = f"Error generating travel plan: {str(e)}"
print(error_msg) # Log the error
# Generate a basic fallback plan
fallback_plan = generate_fallback_plan(destination, days, interests, budget)
return fallback_plan
def generate_fallback_plan(destination, days, interests, budget):
"""Generate a basic fallback plan if the model fails"""
# Start with the overview section
fallback_plan = f"# Emergency Travel Plan for {destination}\n\n"
# Basic activity templates
basic_activities = {
'Culture': ['Visit museums', 'Explore historical sites', 'Attend local events'],
'History': ['Tour historic landmarks', 'Visit ancient sites', 'Join history walks'],
'Food': ['Try local cuisine', 'Visit food markets', 'Take cooking classes'],
'Nature': ['Visit parks', 'Go hiking', 'Explore gardens'],
'Shopping': ['Visit markets', 'Shop at local stores', 'Explore shopping districts'],
'Adventure': ['Join tours', 'Try outdoor activities', 'Explore surroundings'],
'Relaxation': ['Visit spa', 'Relax in parks', 'Enjoy scenic views'],
'Art': ['Visit galleries', 'See street art', 'Attend exhibitions'],
'Museums': ['Visit main museums', 'Join guided tours', 'See special exhibits']
}
# Generate exactly the requested number of days
for day in range(1, days + 1):
fallback_plan += f"\nDay {day}:\n"
# Select activities based on interests
day_activities = []
available_interests = interests[:2] # Use up to 2 interests per day
for interest in available_interests:
if interest in basic_activities:
activity = random.choice(basic_activities[interest])
day_activities.append(activity)
# Add budget-appropriate text
budget_text = {
'Budget': 'Focus on free and affordable activities',
'Moderate': 'Mix of affordable and premium experiences',
'Luxury': 'Premium experiences and exclusive access'
}.get(budget, '')
# Format the day's activities
fallback_plan += f"Morning: {day_activities[0] if day_activities else 'Explore the area'}\n"
if len(day_activities) > 1:
fallback_plan += f"Afternoon/Evening: {day_activities[1]}\n"
fallback_plan += f"Note: {budget_text}\n"
# Format the fallback plan using the same formatter
return format_travel_plan(fallback_plan, days)
def format_travel_plan(plan, days):
"""Format travel plan for 1-14 days with flexible activity distribution"""
# Validate days input
days = max(1, min(days, 14))
# Initialize day activities dictionary
day_activities = {day: [] for day in range(1, days + 1)}
# Parse input plan
current_day = None
for line in plan.split('\n'):
line = line.strip()
if not line:
continue
# Detect day headers
if line.lower().startswith('day'):
try:
day_num = int(''.join(filter(str.isdigit, line.split()[0])))
if 1 <= day_num <= days:
current_day = day_num
except ValueError:
current_day = None
continue
# Collect activities
elif current_day and current_day <= days:
# Split by multiple delimiters, filter meaningful activities
activities = [
act.strip()
for act in re.split(r'[.;,]', line)
if act.strip() and len(act.strip()) > 5
]
# Add activities for current day
for activity in activities:
if len(day_activities[current_day]) < 4:
day_activities[current_day].append(activity)
# Ensure each day has activities
for day in range(1, days + 1):
if not day_activities[day]:
if day == 1:
day_activities[day].append("Explore city highlights")
else:
day_activities[day].append("Continue exploring local attractions")
# Generate formatted plan
formatted_plan = []
for day in range(1, days + 1):
formatted_plan.append(f"### Day {day}\n")
for activity in day_activities[day]:
formatted_plan.append(f"- {activity}")
formatted_plan.append("\n")
return "\n".join(formatted_plan)
def main():
st.set_page_config(
page_title="AI Travel Planner",
page_icon="βοΈ",
layout="wide"
)
st.title("βοΈ AI Travel Planner")
st.markdown("### Plan your perfect trip with AI assistance!")
# Add training button in sidebar only
with st.sidebar:
st.header("Model Management")
if st.button("Retrain Model"):
with st.spinner("Training new model... This will take a while..."):
model, tokenizer = train_model()
if model is not None:
st.session_state['model'] = model
st.session_state['tokenizer'] = tokenizer
st.success("Model training completed!")
# Add model information
st.markdown("### Model Information")
if 'model' in st.session_state:
st.success("β Model loaded")
st.info("""
This model was trained on travel plans for:
- 5 destinations
- 1-14 days duration
- 3 budget levels
- 5 interest combinations
""")
# Load or train model
if 'model' not in st.session_state:
with st.spinner("Loading AI model... Please wait..."):
model, tokenizer = load_or_train_model()
if model is None or tokenizer is None:
st.error("Failed to load/train the AI model. Please try again.")
return
st.session_state.model = model
st.session_state.tokenizer = tokenizer
# Create two columns for input form
col1, col2 = st.columns([2, 1])
with col1:
# Input form in a card-like container
with st.container():
st.markdown("### π― Plan Your Trip")
# Destination and Duration row
dest_col, days_col = st.columns(2)
with dest_col:
destination = st.text_input(
"π Destination",
placeholder="e.g., Paris, Tokyo, New York...",
help="Enter the city you want to visit"
)
with days_col:
days = st.slider(
"π
Number of days",
min_value=1,
max_value=14,
value=3,
help="Select the duration of your trip"
)
# Budget and Interests row
budget_col, interests_col = st.columns(2)
with budget_col:
budget = st.selectbox(
"π° Budget Level",
["Budget", "Moderate", "Luxury"],
help="Select your preferred budget level"
)
with interests_col:
interests = st.multiselect(
"π― Interests",
["Culture", "History", "Food", "Nature", "Shopping",
"Adventure", "Relaxation", "Art", "Museums"],
["Culture", "Food"],
help="Select up to three interests to personalize your plan"
)
with col2:
# Tips and information
st.markdown("### π‘ Travel Tips")
st.info("""
- Choose up to 3 interests for best results
- Consider your travel season
- Budget levels affect activity suggestions
- Plans are customizable after generation
""")
# Generate button centered
col1, col2, col3 = st.columns([1, 2, 1])
with col2:
generate_button = st.button(
"π¨ Generate Travel Plan",
type="primary",
use_container_width=True
)
if generate_button:
if not destination:
st.error("Please enter a destination!")
return
if not interests:
st.error("Please select at least one interest!")
return
if len(interests) > 3:
st.warning("For best results, please select up to 3 interests.")
with st.spinner("π€ Creating your personalized travel plan..."):
travel_plan = generate_travel_plan(
destination,
days,
interests,
budget,
st.session_state.model,
st.session_state.tokenizer
)
# Create an expander for the success message with trip overview
with st.expander("β¨ Your travel plan is ready! Click to see trip overview", expanded=True):
col1, col2, col3 = st.columns(3)
with col1:
st.metric("Destination", destination)
with col2:
if days == 1:
st.metric("Duration", f"{days} day")
else:
st.metric("Duration", f"{days} days")
with col3:
st.metric("Budget", budget)
st.write("**Selected Interests:**", ", ".join(interests))
# Display the plan in tabs with improved styling
plan_tab, summary_tab = st.tabs(["π Detailed Itinerary", "βΉοΈ Trip Summary"])
with plan_tab:
# Add a container for better spacing
with st.container():
# Add trip title
st.markdown(f"## π {days}-Day Trip to {destination}")
st.markdown("---")
# Display the formatted plan
st.markdown(travel_plan)
# Add export options in a nice container
with st.container():
st.markdown("---")
col1, col2 = st.columns([1, 4])
with col1:
st.download_button(
label="π₯ Download Plan",
data=travel_plan,
file_name=f"travel_plan_{destination.lower().replace(' ', '_')}.md",
mime="text/markdown",
use_container_width=True
)
with summary_tab:
# Create three columns for summary information with cards
with st.container():
st.markdown("## Trip Overview")
sum_col1, sum_col2, sum_col3 = st.columns(3)
with sum_col1:
with st.container():
st.markdown("### π Destination Details")
st.markdown(f"**Location:** {destination}")
if days == 1:
st.markdown(f"**Duration:** {days} day")
else:
st.markdown(f"**Duration:** {days} days")
st.markdown(f"**Budget Level:** {budget}")
with sum_col2:
with st.container():
st.markdown("### π― Trip Focus")
st.markdown("**Selected Interests:**")
for interest in interests:
st.markdown(f"- {interest}")
with sum_col3:
with st.container():
st.markdown("### β οΈ Travel Tips")
st.info(
"β’ Verify opening hours\n"
"β’ Check current prices\n"
"β’ Confirm availability\n"
"β’ Consider seasonal factors"
)
if __name__ == "__main__":
main() |