File size: 24,594 Bytes
0bfbee7 |
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 |
import requests
from bs4 import BeautifulSoup
import pandas as pd
import time
import json
import re
import logging
from urllib.parse import urlparse
import fastavro
import msgpack
import os
from datasets import Dataset
# Function to scrape data and return as a Hugging Face Dataset
def scrape_data(url):
response = requests.get(url)
soup = BeautifulSoup(response.text, 'html.parser')
# Example scraping logic (to be customized)
data = []
for item in soup.find_all('div', class_='gpu-item'):
gpu_info = {
'gpu_name': item.find('h2').text,
'architecture': item.find('span', class_='architecture').text,
'memory_size': item.find('span', class_='memory-size').text,
# Add more fields as necessary
}
data.append(gpu_info)
# Convert to Hugging Face Dataset
return Dataset.from_list(data)
# Import additional libraries for new formats
import pyarrow as pa
import pyarrow.parquet as pq
import fastavro
import h5py
import sqlite3
import xml.etree.ElementTree as ET
import yaml
import pickle
from scipy.io import savemat
# Set up logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
logger = logging.getLogger(__name__)
# Try to import Selenium components - they'll be used if available
try:
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from webdriver_manager.chrome import ChromeDriverManager
SELENIUM_AVAILABLE = True
logger.info("Selenium is available and will be used for JavaScript-heavy sites")
except ImportError:
SELENIUM_AVAILABLE = False
logger.warning("Selenium not available. Install with: pip install selenium webdriver-manager")
class NvidiaGpuScraper:
def __init__(self, use_selenium=True):
self.use_selenium = use_selenium and SELENIUM_AVAILABLE
self.driver = self._setup_driver() if self.use_selenium else None
def _setup_driver(self):
"""Set up and return a Selenium WebDriver if available"""
if not SELENIUM_AVAILABLE:
return None
try:
options = Options()
options.add_argument('--headless')
options.add_argument('--no-sandbox')
options.add_argument('--disable-dev-shm-usage')
options.add_argument('--disable-gpu')
options.add_argument("--window-size=1920,1080")
options.add_argument("--user-agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36")
service = Service(ChromeDriverManager().install())
driver = webdriver.Chrome(service=service, options=options)
return driver
except Exception as e:
logger.error(f"Failed to initialize Selenium: {e}")
return None
def _fetch_with_selenium(self, url):
"""Fetch page content using Selenium for JavaScript-heavy sites"""
if self.driver is None:
return None
try:
logger.info(f"Fetching with Selenium: {url}")
self.driver.get(url)
# Wait for the page to load completely
WebDriverWait(self.driver, 20).until(
EC.presence_of_element_located((By.TAG_NAME, "body"))
)
# Scroll down to load lazy content
self.driver.execute_script("window.scrollTo(0, document.body.scrollHeight/2);")
time.sleep(1)
self.driver.execute_script("window.scrollTo(0, document.body.scrollHeight);")
time.sleep(2) # Additional wait for dynamic content
# Expand any "See more specifications" buttons if they exist
try:
see_more_buttons = self.driver.find_elements(By.XPATH,
"//button[contains(text(), 'See more') or contains(text(), 'specifications') or contains(text(), 'specs')]")
for button in see_more_buttons:
self.driver.execute_script("arguments[0].click();", button)
time.sleep(1)
except Exception as e:
logger.warning(f"Could not expand specification sections: {e}")
# Get the page source after JavaScript execution
page_source = self.driver.page_source
return BeautifulSoup(page_source, 'html.parser')
except Exception as e:
logger.error(f"Selenium error for {url}: {e}")
return None
def _fetch_with_requests(self, url):
"""Fetch page content using requests library"""
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
'Accept-Language': 'en-US,en;q=0.5',
'Referer': 'https://www.google.com/',
'DNT': '1',
'Connection': 'keep-alive',
'Upgrade-Insecure-Requests': '1',
'Cache-Control': 'max-age=0',
}
for attempt in range(3):
try:
logger.info(f"Fetching with requests: {url}")
response = requests.get(url, timeout=30, headers=headers)
response.raise_for_status()
return BeautifulSoup(response.content, 'html.parser')
except requests.exceptions.RequestException as e:
wait_time = 2 ** attempt
logger.warning(f"Request error for {url}: {e}. Retrying in {wait_time} seconds...")
time.sleep(wait_time)
return None
def fetch_page(self, url):
if not url:
raise ValueError("The URL provided is empty.")
"""Fetch page content, trying Selenium first if available"""
if self.use_selenium:
soup = self._fetch_with_selenium(url)
if soup:
return soup
# Fall back to requests if Selenium failed or isn't available
return self._fetch_with_requests(url)
def extract_gpu_specs(self, soup, url):
"""Extract GPU specifications from NVIDIA product pages"""
specs = {
'model': 'N/A',
'gpu_name': 'N/A',
'architecture': 'N/A',
'boost_clock': 'N/A',
'memory_size': 'N/A',
'memory_type': 'N/A',
'memory_interface': 'N/A',
'tdp': 'N/A',
'cuda_cores': 'N/A',
'tensor_cores': 'N/A',
'rt_cores': 'N/A',
'process_node': 'N/A',
'transistor_count': 'N/A',
'price': 'N/A',
'release_date': 'N/A',
'url': url,
}
try:
# Extract model name
for selector in ['h1', '.product-title', '.product-name', '.prod-title']:
title_element = soup.select_one(selector)
if title_element and title_element.text.strip():
specs['model'] = title_element.text.strip()
# Try to extract GPU name (e.g., RTX 4090)
gpu_match = re.search(r'(GTX|RTX|RTX\s+SUPER|GTX\s+SUPER)\s+(\d{4}\s*(?:Ti|SUPER)?)',
specs['model'], re.IGNORECASE)
if gpu_match:
specs['gpu_name'] = f"{gpu_match.group(1)} {gpu_match.group(2)}".strip()
break
# Field mapping dictionary - different ways NVIDIA might label each spec
field_mappings = {
'architecture': ['gpu architecture', 'architecture', 'nvidia architecture'],
'boost_clock': ['boost clock', 'gpu boost clock', 'clock speed', 'boost'],
'memory_size': ['memory size', 'standard memory config', 'memory configuration', 'video memory'],
'memory_type': ['memory type', 'memory spec', 'standard memory'],
'memory_interface': ['memory interface', 'memory bus', 'interface width', 'bit width'],
'tdp': ['graphics card power', 'tdp', 'total graphics power', 'power consumption', 'tgp', 'maximum power'],
'cuda_cores': ['cuda cores', 'cuda', 'nvidia cuda cores'],
'tensor_cores': ['tensor cores', 'tensor', 'ai cores'],
'rt_cores': ['rt cores', 'ray tracing cores', 'rt'],
'process_node': ['process', 'fabrication process', 'manufacturing process', 'fab'],
'transistor_count': ['transistor', 'transistor count', 'number of transistors'],
'price': ['price', 'msrp', 'suggested price', 'starting at'],
'release_date': ['release date', 'availability', 'launch date', 'available']
}
# Look for various specs sections
spec_sections = soup.select('.specs-section, .tech-specs, .product-specs, .specs, .spec-table, .spec, [class*="spec"]')
# If no dedicated sections found, look through the entire page
if not spec_sections:
spec_sections = [soup]
for section in spec_sections:
# Method 1: Look for labeled pairs or tables
self._extract_from_tables_and_pairs(section, specs, field_mappings)
# Method 2: Look for text patterns throughout the page
self._extract_from_text_patterns(section, specs)
# Extract from specification headings and adjacent elements
self._extract_from_spec_headings(soup, specs, field_mappings)
# Try to find any JSON-LD or structured data with specs
self._extract_from_json_ld(soup, specs)
# Clean and standardize specs
self._clean_specs(specs)
logger.info(f"Extracted NVIDIA GPU specs: {specs}")
return specs
except Exception as e:
logger.error(f"Error extracting GPU specs: {e}")
return specs
def _extract_from_tables_and_pairs(self, section, specs, field_mappings):
"""Extract specs from table-like structures or label-value pairs"""
# Check for table rows
rows = section.select('tr, .spec-row, .specs-row, [class*="row"]')
for row in rows:
cells = row.select('th, td, .spec-label, .spec-value, .specs-label, .specs-value')
if len(cells) >= 2:
header = cells[0].text.strip().lower()
value = cells[1].text.strip()
# Match header to our fields
for field, possible_headers in field_mappings.items():
if any(h in header for h in possible_headers):
specs[field] = value
# Check for definition lists
terms = section.select('dt, .term, .specs-term')
for term in terms:
header = term.text.strip().lower()
value_el = term.find_next_sibling(['dd', '.definition', '.specs-definition'])
if value_el:
value = value_el.text.strip()
# Match header to our fields
for field, possible_headers in field_mappings.items():
if any(h in header for h in possible_headers):
specs[field] = value
# Check for labeled pairs (common in NVIDIA's newer layout)
labels = section.select('.specs-label, .spec-label, .specs-name, .label, [class*="label"]')
for label in labels:
header = label.text.strip().lower()
# Try to find the adjacent value element
value_el = label.find_next_sibling('.specs-value, .spec-value, .specs-data, .value, [class*="value"]')
if value_el:
value = value_el.text.strip()
# Match header to our fields
for field, possible_headers in field_mappings.items():
if any(h in header for h in possible_headers):
specs[field] = value
def _extract_from_text_patterns(self, section, specs):
"""Extract specs using regex patterns in the page text"""
text = section.get_text(' ', strip=True)
# Extract CUDA cores
cuda_matches = re.search(r'(\d[\d,]+)\s*(?:nvidia)?\s*cuda\s*cores', text, re.IGNORECASE)
if cuda_matches and specs['cuda_cores'] == 'N/A':
specs['cuda_cores'] = cuda_matches.group(1)
# Extract Tensor cores
tensor_matches = re.search(r'(\d+)\s*(?:nvidia)?\s*tensor\s*cores', text, re.IGNORECASE)
if tensor_matches and specs['tensor_cores'] == 'N/A':
specs['tensor_cores'] = tensor_matches.group(1)
# Extract RT cores
rt_matches = re.search(r'(\d+)\s*(?:nvidia)?\s*rt\s*cores', text, re.IGNORECASE)
if rt_matches and specs['rt_cores'] == 'N/A':
specs['rt_cores'] = rt_matches.group(1)
# Extract memory size
mem_matches = re.search(r'(\d+)\s*GB\s*(?:G?DDR\d+[X]?)', text, re.IGNORECASE)
if mem_matches and specs['memory_size'] == 'N/A':
specs['memory_size'] = f"{mem_matches.group(1)} GB"
if specs['memory_type'] == 'N/A':
specs['memory_type'] = mem_matches.group(2)
# Extract boost clock
clock_matches = re.search(r'boost\s*clock\s*(?:up\s*to)?\s*:?\s*([\d.]+)\s*(?:MHz|GHz)', text, re.IGNORECASE)
if clock_matches and specs['boost_clock'] == 'N/A':
value = clock_matches.group(1)
unit = 'GHz' if float(value) < 100 else 'MHz' # Infer unit if not in match
specs['boost_clock'] = f"{value} {unit}"
# Extract memory interface
interface_matches = re.search(r'(\d+)[\s-]*bit(?:\s*memory)?\s*(?:interface|bus)', text, re.IGNORECASE)
if interface_matches and specs['memory_interface'] == 'N/A':
specs['memory_interface'] = f"{interface_matches.group(1)}-bit"
def _extract_from_spec_headings(self, soup, specs, field_mappings):
"""Extract specs from headings and their adjacent content"""
for field, terms in field_mappings.items():
if specs[field] != 'N/A': # Skip if already found
continue
for term in terms:
# Look for headings containing the term
headers = soup.select(f'h1:contains("{term}"), h2:contains("{term}"), h3:contains("{term}"), h4:contains("{term}"), h5:contains("{term}")')
for header in headers:
# Look at next sibling or child for the value
value_el = header.find_next()
if value_el:
specs[field] = value_el.text.strip()
break
def _extract_from_json_ld(self, soup, specs):
"""Extract specs from JSON-LD structured data if available"""
for script in soup.select('script[type="application/ld+json"]'):
try:
data = json.loads(script.string)
# Look for product data
if 'name' in data and specs['model'] == 'N/A':
specs['model'] = data['name']
# Check for specs in properties
if 'additionalProperty' in data:
for prop in data['additionalProperty']:
name = prop.get('name', '').lower()
value = prop.get('value', '')
if 'cuda' in name and specs['cuda_cores'] == 'N/A':
specs['cuda_cores'] = value
elif 'clock' in name and 'boost' in name and specs['boost_clock'] == 'N/A':
specs['boost_clock'] = value
elif 'memory' in name and 'size' in name and specs['memory_size'] == 'N/A':
specs['memory_size'] = value
# Add other mappings as needed
# Check for offer data
if 'offers' in data and specs['price'] == 'N/A':
if isinstance(data['offers'], list) and len(data['offers']) > 0:
specs['price'] = data['offers'][0].get('price', 'N/A')
elif isinstance(data['offers'], dict):
specs['price'] = data['offers'].get('price', 'N/A')
except:
pass
def _clean_specs(self, specs):
"""Clean and standardize the extracted specs"""
# Clean CUDA cores (remove commas)
if specs['cuda_cores'] != 'N/A':
specs['cuda_cores'] = specs['cuda_cores'].replace(',', '')
# Standardize memory size format
if specs['memory_size'] != 'N/A' and 'GB' not in specs['memory_size']:
if specs['memory_size'].isdigit():
specs['memory_size'] = f"{specs['memory_size']} GB"
# Standardize boost clock format
if specs['boost_clock'] != 'N/A':
# If it's just a number, add units
if re.match(r'^\d+(\.\d+)?$', specs['boost_clock']):
value = float(specs['boost_clock'])
if value > 100: # Likely MHz
specs['boost_clock'] = f"{value} MHz"
else: # Likely GHz
specs['boost_clock'] = f"{value} GHz"
def scrape_gpu(self, url):
if not url:
raise ValueError("The URL provided is empty.")
"""Scrape a single GPU product page"""
soup = self.fetch_page(url)
if not soup:
return {
'model': 'Failed to fetch',
'url': url
}
return self.extract_gpu_specs(soup, url)
def scrape_multiple_gpus(self, urls):
if not urls:
raise ValueError("The list of URLs is empty.")
"""Scrape multiple GPU product pages"""
results = []
for url in urls:
try:
specs = self.scrape_gpu(url)
results.append(specs)
# Be polite with a delay between requests
time.sleep(2)
except Exception as e:
logger.error(f"Error processing {url}: {e}")
results.append({
'model': f"Error: {str(e)[:50]}",
'url': url
})
return results
def cleanup(self):
"""Clean up resources"""
if self.driver:
self.driver.quit()
# Main execution function
def main():
# NVIDIA GPU product URLs - focused on specific product pages
nvidia_urls = [
"https://www.nvidia.com/en-us/geforce/graphics-cards/40-series/rtx-4090/",
"https://www.nvidia.com/en-us/geforce/graphics-cards/40-series/rtx-4080/",
"https://www.nvidia.com/en-us/geforce/graphics-cards/40-series/rtx-4070-ti-super/",
"https://www.nvidia.com/en-us/geforce/graphics-cards/40-series/rtx-4070/",
"https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3090-3090ti/",
"https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3080-3080ti/",
"https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3070-3070ti/",
]
# Create the scraper and run
scraper = NvidiaGpuScraper(use_selenium=SELENIUM_AVAILABLE)
try:
# Scrape the GPUs
results = scraper.scrape_multiple_gpus(nvidia_urls)
# Create and save DataFrame
df = pd.DataFrame(results)
df.to_csv('nvidia_gpus.csv', index=False)
df.to_json('nvidia_gpus.json', orient='records', lines=True)
df.to_excel('nvidia_gpus.xlsx', index=False)
# Save DataFrame in various formats
try:
df.to_parquet('nvidia_gpus.parquet')
except Exception as e:
logger.warning(f"Failed to save as Parquet: {e}")
try:
# Convert DataFrame to list of dictionaries
records = df.to_dict(orient='records')
# Define Avro schema
schema = {
'type': 'record',
'name': 'GPU',
'fields': [
{'name': col, 'type': ['string', 'null']} for col in df.columns
]
}
# Write to Avro file
with open('nvidia_gpus.avro', 'wb') as avro_file:
fastavro.writer(avro_file, schema, records)
except Exception as e:
logger.warning(f"Failed to save as Avro: {e}")
try:
df.to_orc('nvidia_gpus.orc')
except Exception as e:
logger.warning(f"Failed to save as ORC: {e}")
try:
df.to_hdf('nvidia_gpus.h5', key='df', mode='w')
except Exception as e:
logger.warning(f"Failed to save as HDF5: {e}")
try:
with sqlite3.connect('nvidia_gpus.db') as conn:
df.to_sql('gpus', conn, if_exists='replace', index=False)
except Exception as e:
logger.warning(f"Failed to save as SQLite: {e}")
try:
df.to_xml('nvidia_gpus.xml')
except Exception as e:
logger.warning(f"Failed to save as XML: {e}")
try:
with open('nvidia_gpus.yaml', 'w') as yaml_file:
yaml.dump(df.to_dict(orient='records'), yaml_file)
except Exception as e:
logger.warning(f"Failed to save as YAML: {e}")
try:
with open('nvidia_gpus.pkl', 'wb') as pickle_file:
pickle.dump(df, pickle_file)
except Exception as e:
logger.warning(f"Failed to save as Pickle: {e}")
try:
savemat('nvidia_gpus.mat', {'gpus': df.to_dict(orient='records')})
except Exception as e:
logger.warning(f"Failed to save as MAT: {e}")
try:
df.to_csv('nvidia_gpus.tsv', sep='\t', index=False)
except Exception as e:
logger.warning(f"Failed to save as TSV: {e}")
try:
df.to_json('nvidia_gpus.ndjson', orient='records', lines=True)
except Exception as e:
logger.warning(f"Failed to save as NDJSON: {e}")
try:
df.to_csv('nvidia_gpus.arff', index=False)
except Exception as e:
logger.warning(f"Failed to save as ARFF: {e}")
try:
# Convert DataFrame to dictionary
data = df.to_dict(orient='records')
# Write to MessagePack file
with open('nvidia_gpus.msgpack', 'wb') as msgpack_file:
msgpack.pack(data, msgpack_file)
except Exception as e:
logger.warning(f"Failed to save as MessagePack: {e}")
try:
df.to_pickle('nvidia_gpus.protobuf')
except Exception as e:
logger.warning(f"Failed to save as ProtoBuf: {e}")
try:
df.to_csv('nvidia_gpus.dta', index=False)
except Exception as e:
logger.warning(f"Failed to save as DTA: {e}")
try:
df.to_csv('nvidia_gpus.sas', index=False)
except Exception as e:
logger.warning(f"Failed to save as SAS: {e}")
try:
df.to_csv('nvidia_gpus.spss', index=False)
except Exception as e:
logger.warning(f"Failed to save as SPSS: {e}")
print("\nResults:")
print("\nResults:")
print(df)
# Print summary
successful = sum(1 for spec in results if spec.get('model') not in ['N/A', 'Failed to fetch'])
print(f"\nSummary: Successfully scraped {successful} out of {len(results)} NVIDIA GPUs")
return df
finally:
# Always clean up resources
scraper.cleanup()
if __name__ == "__main__":
main()
|