Spaces:
Sleeping
Sleeping
File size: 8,480 Bytes
9aa4daf | 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 | """
System Prerequisites Checker Module
This module provides functionality to check system prerequisites including:
- CUDA/GPU availability
- Environment dependencies
- Model download with progress tracking
"""
import os
import sys
import torch
import platform
from typing import Dict, Tuple, Optional
from pathlib import Path
import importlib.metadata
from huggingface_hub import hf_hub_download, snapshot_download
from tqdm import tqdm
class SystemChecker:
"""Check system prerequisites for MLOps platform."""
def __init__(self, models_dir: str = "models"):
"""
Initialize system checker.
Args:
models_dir: Directory to store downloaded models
"""
self.models_dir = Path(models_dir)
self.models_dir.mkdir(parents=True, exist_ok=True)
def check_cuda(self) -> Dict[str, any]:
"""
Check CUDA/GPU availability and information.
Returns:
Dict with CUDA status, device info, and specifications
"""
result = {
"available": torch.cuda.is_available(),
"device_count": 0,
"devices": [],
"cuda_version": None,
"cudnn_version": None
}
if result["available"]:
result["device_count"] = torch.cuda.device_count()
result["cuda_version"] = torch.version.cuda
result["cudnn_version"] = torch.backends.cudnn.version()
for i in range(result["device_count"]):
device_props = {
"id": i,
"name": torch.cuda.get_device_name(i),
"memory_total": torch.cuda.get_device_properties(i).total_memory / 1024**3, # GB
"compute_capability": f"{torch.cuda.get_device_properties(i).major}.{torch.cuda.get_device_properties(i).minor}"
}
result["devices"].append(device_props)
return result
def check_environment(self) -> Dict[str, any]:
"""
Check Python environment and required dependencies.
Returns:
Dict with Python version, package versions, and system info
"""
result = {
"python_version": sys.version,
"platform": platform.platform(),
"architecture": platform.machine(),
"packages": {},
"missing_packages": [],
"all_satisfied": True
}
# Required packages with minimum versions
required_packages = {
"torch": "2.0.0",
"transformers": "4.36.0",
"streamlit": "1.28.0",
"pandas": "2.0.0",
"numpy": "1.24.0",
"plotly": "5.18.0",
"scikit-learn": "1.3.0"
}
for package, min_version in required_packages.items():
try:
version = importlib.metadata.version(package)
result["packages"][package] = {
"installed": version,
"required": f">={min_version}",
"satisfied": True # Simple check, could add version comparison
}
except importlib.metadata.PackageNotFoundError:
result["packages"][package] = {
"installed": None,
"required": f">={min_version}",
"satisfied": False
}
result["missing_packages"].append(package)
result["all_satisfied"] = False
return result
def download_model(
self,
model_name: str,
progress_callback: Optional[callable] = None
) -> Tuple[bool, str, str]:
"""
Download model from HuggingFace Hub to local cache.
Args:
model_name: HuggingFace model identifier (e.g., "roberta-base")
progress_callback: Optional callback function for progress updates
Returns:
Tuple of (success: bool, model_path: str, message: str)
"""
try:
model_cache_path = self.models_dir / model_name.replace("/", "_")
# Check if model already exists
if model_cache_path.exists() and any(model_cache_path.iterdir()):
return True, str(model_cache_path), f"Model '{model_name}' already exists in cache"
# Download model
if progress_callback:
progress_callback(f"Downloading {model_name}...", 0.1)
# Use snapshot_download to get all model files
cache_dir = snapshot_download(
repo_id=model_name,
cache_dir=str(self.models_dir),
local_dir=str(model_cache_path),
local_dir_use_symlinks=False
)
if progress_callback:
progress_callback(f"Downloaded {model_name} successfully", 1.0)
return True, str(model_cache_path), f"Model '{model_name}' downloaded successfully"
except Exception as e:
error_msg = f"Failed to download model '{model_name}': {str(e)}"
if progress_callback:
progress_callback(error_msg, 0.0)
return False, "", error_msg
def get_model_info(self, model_name: str) -> Dict[str, any]:
"""
Get information about a model (local or remote).
Args:
model_name: Model identifier
Returns:
Dict with model information
"""
model_cache_path = self.models_dir / model_name.replace("/", "_")
info = {
"name": model_name,
"local_path": str(model_cache_path),
"exists_locally": model_cache_path.exists() and any(model_cache_path.iterdir()),
"size_mb": 0
}
if info["exists_locally"]:
# Calculate total size
total_size = sum(
f.stat().st_size
for f in model_cache_path.rglob('*')
if f.is_file()
)
info["size_mb"] = total_size / (1024 * 1024)
return info
def format_bytes(bytes_size: float) -> str:
"""Format bytes to human-readable string."""
for unit in ['B', 'KB', 'MB', 'GB']:
if bytes_size < 1024.0:
return f"{bytes_size:.2f} {unit}"
bytes_size /= 1024.0
return f"{bytes_size:.2f} TB"
def get_system_summary() -> str:
"""Get a formatted summary of system capabilities."""
checker = SystemChecker()
cuda_info = checker.check_cuda()
env_info = checker.check_environment()
summary = []
summary.append("=" * 60)
summary.append("SYSTEM SUMMARY")
summary.append("=" * 60)
# Python & Platform
summary.append(f"\nPython: {env_info['python_version'].split()[0]}")
summary.append(f"Platform: {env_info['platform']}")
summary.append(f"Architecture: {env_info['architecture']}")
# CUDA
summary.append(f"\nCUDA Available: {'Yes' if cuda_info['available'] else 'No (CPU only)'}")
if cuda_info['available']:
summary.append(f"CUDA Version: {cuda_info['cuda_version']}")
summary.append(f"Number of GPUs: {cuda_info['device_count']}")
for device in cuda_info['devices']:
summary.append(f" - GPU {device['id']}: {device['name']}")
summary.append(f" Memory: {device['memory_total']:.2f} GB")
summary.append(f" Compute: {device['compute_capability']}")
# Packages
summary.append(f"\n📦 Required Packages: {'✅ All Satisfied' if env_info['all_satisfied'] else '⚠️ Missing Packages'}")
if env_info['missing_packages']:
summary.append(f" Missing: {', '.join(env_info['missing_packages'])}")
summary.append("=" * 60)
return "\n".join(summary)
if __name__ == "__main__":
# Test the system checker
print(get_system_summary())
checker = SystemChecker()
# Test model download (small model for testing)
print("\n\nTesting model download...")
success, path, msg = checker.download_model(
"distilbert-base-uncased",
progress_callback=lambda msg, progress: print(f"Progress: {progress*100:.0f}% - {msg}")
)
print(f"Result: {msg}")
|