[Enchance] Set a random seed when the user does not set a seed. (#554)

mmcls/apis/__init__.py

@@ -1,9 +1,10 @@
 # Copyright (c) OpenMMLab. All rights reserved.
 from .inference import inference_model, init_model, show_result_pyplot
 from .test import multi_gpu_test, single_gpu_test
-from .train import set_random_seed, train_model
+from .train import init_random_seed, set_random_seed, train_model
 
 __all__ = [
     'set_random_seed', 'train_model', 'init_model', 'inference_model',
-    'multi_gpu_test', 'single_gpu_test', 'show_result_pyplot'
+    'multi_gpu_test', 'single_gpu_test', 'show_result_pyplot',
+    'init_random_seed'
 ]

mmcls/apis/train.py

@@ -4,8 +4,10 @@ import warnings
 
 import numpy as np
 import torch
+import torch.distributed as dist
 from mmcv.parallel import MMDataParallel, MMDistributedDataParallel
-from mmcv.runner import DistSamplerSeedHook, build_optimizer, build_runner
+from mmcv.runner import (DistSamplerSeedHook, build_optimizer, build_runner,
+                         get_dist_info)
 
 from mmcls.core import DistOptimizerHook
 from mmcls.datasets import build_dataloader, build_dataset
@@ -29,6 +31,39 @@ except ImportError:
     from mmcls.core import Fp16OptimizerHook
 
 
+def init_random_seed(seed=None, device='cuda'):
+    """Initialize random seed.
+
+    If the seed is not set, the seed will be automatically randomized,
+    and then broadcast to all processes to prevent some potential bugs.
+
+    Args:
+        seed (int, Optional): The seed. Default to None.
+        device (str): The device where the seed will be put on.
+            Default to 'cuda'.
+
+    Returns:
+        int: Seed to be used.
+    """
+    if seed is not None:
+        return seed
+
+    # Make sure all ranks share the same random seed to prevent
+    # some potential bugs. Please refer to
+    # https://github.com/open-mmlab/mmdetection/issues/6339
+    rank, world_size = get_dist_info()
+    seed = np.random.randint(2**31)
+    if world_size == 1:
+        return seed
+
+    if rank == 0:
+        random_num = torch.tensor(seed, dtype=torch.int32, device=device)
+    else:
+        random_num = torch.tensor(0, dtype=torch.int32, device=device)
+    dist.broadcast(random_num, src=0)
+    return random_num.item()
+
+
 def set_random_seed(seed, deterministic=False):
     """Set random seed.
 

tools/train.py

@@ -12,7 +12,7 @@ from mmcv import Config, DictAction
 from mmcv.runner import get_dist_info, init_dist
 
 from mmcls import __version__
-from mmcls.apis import set_random_seed, train_model
+from mmcls.apis import init_random_seed, set_random_seed, train_model
 from mmcls.datasets import build_dataset
 from mmcls.models import build_classifier
 from mmcls.utils import collect_env, get_root_logger
@@ -143,12 +143,12 @@ def main():
     logger.info(f'Config:\n{cfg.pretty_text}')
 
     # set random seeds
-    if args.seed is not None:
-        logger.info(f'Set random seed to {args.seed}, '
-                    f'deterministic: {args.deterministic}')
-        set_random_seed(args.seed, deterministic=args.deterministic)
-    cfg.seed = args.seed
-    meta['seed'] = args.seed
+    seed = init_random_seed(args.seed)
+    logger.info(f'Set random seed to {seed}, '
+                f'deterministic: {args.deterministic}')
+    set_random_seed(seed, deterministic=args.deterministic)
+    cfg.seed = seed
+    meta['seed'] = seed
 
     model = build_classifier(cfg.model)
     model.init_weights()