File size: 9,525 Bytes
18a82fb | 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 | import torch
import numpy as np
import random
import os
from pathlib import Path
import json
import yaml
from typing import Dict, Any, Optional, List
import hashlib
import shutil
from datetime import datetime
def set_seed(seed: int = 42):
"""
Set random seed for reproducibility
Args:
seed: Random seed value
"""
random.seed(seed)
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed(seed)
torch.cuda.manual_seed_all(seed)
torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False
os.environ['PYTHONHASHSEED'] = str(seed)
def get_device(gpu_id: Optional[int] = None) -> torch.device:
"""
Get torch device
Args:
gpu_id: Specific GPU ID to use
Returns:
torch.device
"""
if torch.cuda.is_available():
if gpu_id is not None:
device = torch.device(f'cuda:{gpu_id}')
else:
device = torch.device('cuda')
else:
device = torch.device('cpu')
return device
def count_parameters(model: torch.nn.Module) -> Dict[str, int]:
"""
Count model parameters
Args:
model: PyTorch model
Returns:
Dictionary with parameter counts
"""
total_params = sum(p.numel() for p in model.parameters())
trainable_params = sum(p.numel() for p in model.parameters() if p.requires_grad)
non_trainable_params = total_params - trainable_params
return {
'total': total_params,
'trainable': trainable_params,
'non_trainable': non_trainable_params,
'total_mb': total_params * 4 / 1024 / 1024, # Assuming float32
'trainable_mb': trainable_params * 4 / 1024 / 1024
}
def save_config(config: Dict[str, Any], save_path: Path, format: str = 'json'):
"""
Save configuration to file
Args:
config: Configuration dictionary
save_path: Path to save file
format: File format ('json' or 'yaml')
"""
save_path = Path(save_path)
save_path.parent.mkdir(parents=True, exist_ok=True)
if format == 'json':
with open(save_path, 'w') as f:
json.dump(config, f, indent=2)
elif format == 'yaml':
with open(save_path, 'w') as f:
yaml.dump(config, f, default_flow_style=False)
else:
raise ValueError(f"Unknown format: {format}")
def load_config(config_path: Path) -> Dict[str, Any]:
"""
Load configuration from file
Args:
config_path: Path to config file
Returns:
Configuration dictionary
"""
config_path = Path(config_path)
if config_path.suffix == '.json':
with open(config_path, 'r') as f:
config = json.load(f)
elif config_path.suffix in ['.yaml', '.yml']:
with open(config_path, 'r') as f:
config = yaml.safe_load(f)
else:
raise ValueError(f"Unknown config format: {config_path.suffix}")
return config
def merge_configs(base_config: Dict, update_config: Dict) -> Dict:
"""
Recursively merge two configuration dictionaries
Args:
base_config: Base configuration
update_config: Configuration with updates
Returns:
Merged configuration
"""
merged = base_config.copy()
for key, value in update_config.items():
if key in merged and isinstance(merged[key], dict) and isinstance(value, dict):
merged[key] = merge_configs(merged[key], value)
else:
merged[key] = value
return merged
def get_timestamp() -> str:
"""Get current timestamp string"""
return datetime.now().strftime('%Y%m%d_%H%M%S')
def get_experiment_name(
model_name: str,
dataset_name: str = 'decade',
timestamp: bool = True
) -> str:
"""
Generate experiment name
Args:
model_name: Name of the model
dataset_name: Name of the dataset
timestamp: Whether to include timestamp
Returns:
Experiment name
"""
name_parts = [model_name, dataset_name]
if timestamp:
name_parts.append(get_timestamp())
return '_'.join(name_parts)
def compute_file_hash(file_path: Path, hash_algo: str = 'md5') -> str:
"""
Compute hash of a file
Args:
file_path: Path to file
hash_algo: Hash algorithm to use
Returns:
Hex digest of file hash
"""
hash_func = getattr(hashlib, hash_algo)()
with open(file_path, 'rb') as f:
for chunk in iter(lambda: f.read(4096), b''):
hash_func.update(chunk)
return hash_func.hexdigest()
def create_experiment_structure(base_dir: Path, experiment_name: str) -> Dict[str, Path]:
"""
Create directory structure for an experiment
Args:
base_dir: Base directory for experiments
experiment_name: Name of the experiment
Returns:
Dictionary mapping directory names to paths
"""
exp_dir = base_dir / experiment_name
dirs = {
'root': exp_dir,
'checkpoints': exp_dir / 'checkpoints',
'logs': exp_dir / 'logs',
'visualizations': exp_dir / 'visualizations',
'predictions': exp_dir / 'predictions',
'configs': exp_dir / 'configs'
}
for dir_path in dirs.values():
dir_path.mkdir(parents=True, exist_ok=True)
return dirs
def backup_code(src_dir: Path, backup_dir: Path, extensions: List[str] = None):
"""
Backup source code to experiment directory
Args:
src_dir: Source directory
backup_dir: Backup destination
extensions: List of file extensions to backup
"""
if extensions is None:
extensions = ['.py', '.yaml', '.yml', '.json', '.txt', '.md']
backup_dir = Path(backup_dir)
backup_dir.mkdir(parents=True, exist_ok=True)
for file_path in src_dir.rglob('*'):
if file_path.is_file() and file_path.suffix in extensions:
relative_path = file_path.relative_to(src_dir)
backup_path = backup_dir / relative_path
backup_path.parent.mkdir(parents=True, exist_ok=True)
shutil.copy2(file_path, backup_path)
def format_time(seconds: float) -> str:
"""
Format time in seconds to human-readable string
Args:
seconds: Time in seconds
Returns:
Formatted time string
"""
hours = int(seconds // 3600)
minutes = int((seconds % 3600) // 60)
seconds = int(seconds % 60)
if hours > 0:
return f"{hours}h {minutes}m {seconds}s"
elif minutes > 0:
return f"{minutes}m {seconds}s"
else:
return f"{seconds}s"
def get_gpu_memory_usage() -> Dict[str, float]:
"""
Get GPU memory usage
Returns:
Dictionary with memory usage in MB
"""
if not torch.cuda.is_available():
return {'allocated': 0, 'reserved': 0}
return {
'allocated': torch.cuda.memory_allocated() / 1024 / 1024,
'reserved': torch.cuda.memory_reserved() / 1024 / 1024
}
def clean_checkpoint(checkpoint_path: Path, keep_keys: List[str] = None):
"""
Clean checkpoint file by keeping only specified keys
Args:
checkpoint_path: Path to checkpoint
keep_keys: Keys to keep (default: model_state_dict only)
"""
if keep_keys is None:
keep_keys = ['model_state_dict']
checkpoint = torch.load(checkpoint_path, map_location='cpu')
cleaned_checkpoint = {k: v for k, v in checkpoint.items() if k in keep_keys}
# Save cleaned checkpoint
output_path = checkpoint_path.parent / f"{checkpoint_path.stem}_cleaned.pth"
torch.save(cleaned_checkpoint, output_path)
# Print size reduction
original_size = checkpoint_path.stat().st_size / 1024 / 1024
new_size = output_path.stat().st_size / 1024 / 1024
print(f"Cleaned checkpoint: {original_size:.1f}MB → {new_size:.1f}MB")
return output_path
class AverageMeter:
"""Computes and stores the average and current value"""
def __init__(self):
self.reset()
def reset(self):
self.val = 0
self.avg = 0
self.sum = 0
self.count = 0
def update(self, val, n=1):
self.val = val
self.sum += val * n
self.count += n
self.avg = self.sum / self.count
class EarlyStopping:
"""Early stopping to stop training when validation loss doesn't improve"""
def __init__(self, patience=7, verbose=False, delta=0):
self.patience = patience
self.verbose = verbose
self.counter = 0
self.best_score = None
self.early_stop = False
self.val_loss_min = np.Inf
self.delta = delta
def __call__(self, val_loss, model=None):
score = -val_loss
if self.best_score is None:
self.best_score = score
self.save_checkpoint(val_loss, model)
elif score < self.best_score + self.delta:
self.counter += 1
if self.verbose:
print(f'EarlyStopping counter: {self.counter} out of {self.patience}')
if self.counter >= self.patience:
self.early_stop = True
else:
self.best_score = score
self.save_checkpoint(val_loss, model)
self.counter = 0
def save_checkpoint(self, val_loss, model):
"""Saves model when validation loss decrease"""
if self.verbose:
print(f'Validation loss decreased ({self.val_loss_min:.6f} --> {val_loss:.6f})')
self.val_loss_min = val_loss |