[Feature] Support fp16 training (#178)

* change mmcls fp16 to mmcv hook

* support fp16

* clean unnessary stuff

configs/fp16/README.md

@@ -0,0 +1,14 @@
+# Mixed Precision Training
+
+## Introduction
+
+[OTHERS]
+
+```latex
+@article{micikevicius2017mixed,
+  title={Mixed precision training},
+  author={Micikevicius, Paulius and Narang, Sharan and Alben, Jonah and Diamos, Gregory and Elsen, Erich and Garcia, David and Ginsburg, Boris and Houston, Michael and Kuchaiev, Oleksii and Venkatesh, Ganesh and others},
+  journal={arXiv preprint arXiv:1710.03740},
+  year={2017}
+}
+```

configs/fp16/resnet50_b32x8_fp16_dynamic_imagenet.py

@@ -0,0 +1,4 @@
+_base_ = ['../resnet/resnet50_b32x8_imagenet.py']
+
+# fp16 settings
+fp16 = dict(loss_scale='dynamic')

configs/fp16/resnet50_b32x8_fp16_imagenet.py

@@ -0,0 +1,4 @@
+_base_ = ['../resnet/resnet50_b32x8_imagenet.py']
+
+# fp16 settings
+fp16 = dict(loss_scale=512.)

mmcls/apis/train.py

@@ -6,11 +6,18 @@ import torch
 from mmcv.parallel import MMDataParallel, MMDistributedDataParallel
 from mmcv.runner import DistSamplerSeedHook, build_optimizer, build_runner
 
-from mmcls.core import (DistEvalHook, DistOptimizerHook, EvalHook,
-                        Fp16OptimizerHook)
+from mmcls.core import DistEvalHook, DistOptimizerHook, EvalHook
 from mmcls.datasets import build_dataloader, build_dataset
 from mmcls.utils import get_root_logger
 
+# TODO import optimizer hook from mmcv and delete them from mmcls
+try:
+    from mmcv.runner import Fp16OptimizerHook
+except ImportError:
+    warnings.warn('FP16OptimizerHook from mmcls will be deprecated.'
+                  'Please install mmcv>=1.1.4.')
+    from mmcls.core import Fp16OptimizerHook
+
 
 def set_random_seed(seed, deterministic=False):
     """Set random seed.

mmcls/models/classifiers/base.py

@@ -10,12 +10,21 @@ import torch.nn as nn
 from mmcv import color_val
 from mmcv.utils import print_log
 
+# TODO import `auto_fp16` from mmcv and delete them from mmcls
+try:
+    from mmcv.runner import auto_fp16
+except ImportError:
+    warnings.warn('auto_fp16 from mmcls will be deprecated.'
+                  'Please install mmcv>=1.1.4.')
+    from mmcls.core import auto_fp16
+
 
 class BaseClassifier(nn.Module, metaclass=ABCMeta):
     """Base class for classifiers"""
 
     def __init__(self):
         super(BaseClassifier, self).__init__()
+        self.fp16_enabled = False
 
     @property
     def with_neck(self):
@@ -70,6 +79,7 @@ class BaseClassifier(nn.Module, metaclass=ABCMeta):
         else:
             raise NotImplementedError('aug_test has not been implemented')
 
+    @auto_fp16(apply_to=('img', ))
     def forward(self, img, return_loss=True, **kwargs):
         """
         Calls either forward_train or forward_test depending on whether

tools/test.py

@@ -10,10 +10,17 @@ from mmcv.parallel import MMDataParallel, MMDistributedDataParallel
 from mmcv.runner import get_dist_info, init_dist, load_checkpoint
 
 from mmcls.apis import multi_gpu_test, single_gpu_test
-from mmcls.core import wrap_fp16_model
 from mmcls.datasets import build_dataloader, build_dataset
 from mmcls.models import build_classifier
 
+# TODO import `wrap_fp16_model` from mmcv and delete them from mmcls
+try:
+    from mmcv.runner import wrap_fp16_model
+except ImportError:
+    warnings.warn('wrap_fp16_model from mmcls will be deprecated.'
+                  'Please install mmcv>=1.1.4.')
+    from mmcls.core import wrap_fp16_model
+
 
 def parse_args():
     parser = argparse.ArgumentParser(description='mmcls test model')