add build_hook & other minor modify

mmcv/image/transforms/geometry.py

@@ -3,7 +3,7 @@ from __future__ import division
 import cv2
 import numpy as np
 
-__all__ = ['imflip', 'imrotate', 'imcrop', 'impad', 'impad_to_multiple', 'bbox_flip']
+__all__ = ['imflip', 'imrotate', 'imcrop', 'impad', 'impad_to_multiple']
 
 
 def imflip(img, direction='horizontal'):
@@ -111,20 +111,6 @@ def bbox_scaling(bboxes, scale, clip_shape=None):
         return scaled_bboxes
 
 
-def bbox_flip(bboxes, img_shape):
-    """Flip bboxes horizontally
-    Args:
-        bboxes(ndarray): shape (..., 4*k)
-        img_shape(tuple): (height, width)
-    """
-    assert bboxes.shape[-1] % 4 == 0
-    w = img_shape[1]
-    flipped = bboxes.copy()
-    flipped[..., 0::4] = w - bboxes[..., 2::4] - 1
-    flipped[..., 2::4] = w - bboxes[..., 0::4] - 1
-    return flipped
-
-
 def imcrop(img, bboxes, scale_ratio=1.0, pad_fill=None):
     """Crop image patches.
 

mmcv/torchpack/hooks/__init__.py

 from .hook import Hook
-from .checkpoint_saver import CheckpointSaverHook
+from .checkpoint_saver import CheckpointHook
 from .closure import ClosureHook
 from .lr_updater import LrUpdaterHook
-from .optimizer_stepper import OptimizerStepperHook
+from .optimizer_stepper import OptimizerHook
 from .iter_timer import IterTimerHook
 from .logger import *

mmcv/torchpack/hooks/checkpoint_saver.py

@@ -2,7 +2,7 @@ from .hook import Hook
 from ..utils import master_only
 
 
-class CheckpointSaverHook(Hook):
+class CheckpointHook(Hook):
 
     def __init__(self,
                  interval=-1,

mmcv/torchpack/hooks/optimizer_stepper.py

@@ -3,7 +3,7 @@ from torch.nn.utils import clip_grad
 from .hook import Hook
 
 
-class OptimizerStepperHook(Hook):
+class OptimizerHook(Hook):
 
     def __init__(self, grad_clip=False, max_norm=35, norm_type=2):
         self.grad_clip = grad_clip

mmcv/torchpack/runner/runner.py

@@ -8,8 +8,8 @@ from torch.nn.parallel import DataParallel, DistributedDataParallel
 
 from .log_buffer import LogBuffer
 from .. import hooks
-from ..hooks import (Hook, LrUpdaterHook, CheckpointSaverHook, IterTimerHook,
-                     OptimizerStepperHook)
+from ..hooks import (Hook, LrUpdaterHook, CheckpointHook, IterTimerHook,
+                     OptimizerHook)
 from ..io import load_checkpoint, save_checkpoint
 from ..utils import (get_dist_info, get_host_info, get_time_str,
                      add_file_handler, obj_from_dict)
@@ -182,6 +182,16 @@ class Runner(object):
         if not inserted:
             self._hooks.insert(0, hook)
 
+    def build_hook(self, hook, args):
+        assert issubclass(hook, Hook), '"hook" must be a Hook object'
+        if isinstance(args, dict):
+            self.register_hook(hook(**args))
+        elif isinstance(args, Hook):
+            self.register_hook(args)
+        else:
+            raise TypeError('"args" must be either a Hook object'
+                            ' or dict, not {}'.format(type(args)))
+
     def call_hook(self, fn_name):
         for hook in self._hooks:
             getattr(hook, fn_name)(self)
@@ -329,7 +339,7 @@ class Runner(object):
 
     def register_training_hooks(self,
                                 lr_config,
-                                grad_clip_config=None,
+                                optimizer_config=None,
                                 checkpoint_config=None,
                                 log_config=None):
         """Register default hooks for training.
@@ -341,22 +351,13 @@ class Runner(object):
         - IterTimerHook
         - LoggerHook
         """
-        if grad_clip_config is None:
-            grad_clip_config = {}
+        if optimizer_config is None:
+            optimizer_config = {}
         if checkpoint_config is None:
             checkpoint_config = {}
         self.register_lr_hooks(lr_config)
- 
-        if isinstance(grad_clip_config, Hook):
-            self.register_hook(grad_clip_config)
-        elif isinstance(grad_clip_config, dict):
-            self.register_hook(OptimizerStepperHook(**grad_clip_config))
-        else:
-            raise TypeError(
-                "OptimizerStepperHook should be a Hook object or dict, not {}".
-                format(type(grad_clip_config)))
-
-        self.register_hook(CheckpointSaverHook(**checkpoint_config))
+        self.build_hook(OptimizerHook, optimizer_config)
+        self.build_hook(CheckpointHook, checkpoint_config)
         self.register_hook(IterTimerHook())
         if log_config is not None:
             self.register_logger_hooks(log_config)