refine code to avoid some problem (#1772)

configs/_base_/datasets/coco.yml

@@ -6,7 +6,6 @@ TrainDataset:
     image_dir: train2017
     anno_path: annotations/instances_train2017.json
     dataset_dir: dataset/coco
-    mixup_epoch: 250
 
 EvalDataset:
   !COCODataSet

configs/_base_/readers/yolov3_reader.yml

@@ -21,6 +21,7 @@ TrainReader:
   batch_size: 8
   shuffle: true
   drop_last: true
+  mixup_epoch: 250
 
 
 EvalReader:

ppdet/data/reader.py

@@ -30,6 +30,7 @@ class Compose(object):
                 if hasattr(op_cls, 'num_classes'):
                     op_cls.num_classes = num_classes
 
+                # TODO: should be refined in the future
                 if op_cls in [
                         transform.Gt2YoloTargetOp, transform.Gt2YoloTarget
                 ]:
@@ -89,7 +90,8 @@ class BaseDataLoader(object):
                  drop_last=False,
                  drop_empty=True,
                  num_classes=81,
-                 with_background=True):
+                 with_background=True,
+                 **kwargs):
         # out fields 
         self._fields = inputs_def['fields'] if inputs_def else None
         # sample transform
@@ -107,6 +109,7 @@ class BaseDataLoader(object):
         self.shuffle = shuffle
         self.drop_last = drop_last
         self.with_background = with_background
+        self.kwargs = kwargs
 
     def __call__(self,
                  dataset,
@@ -120,6 +123,8 @@ class BaseDataLoader(object):
         # get data
         self._dataset.set_out(self._sample_transforms,
                               copy.deepcopy(self._fields))
+        # set kwargs
+        self._dataset.set_kwargs(**self.kwargs)
         # batch sampler
         if batch_sampler is None:
             self._batch_sampler = DistributedBatchSampler(
@@ -154,10 +159,12 @@ class TrainReader(BaseDataLoader):
                  drop_last=True,
                  drop_empty=True,
                  num_classes=81,
-                 with_background=True):
-        super(TrainReader, self).__init__(
-            inputs_def, sample_transforms, batch_transforms, batch_size,
-            shuffle, drop_last, drop_empty, num_classes, with_background)
+                 with_background=True,
+                 **kwargs):
+        super(TrainReader, self).__init__(inputs_def, sample_transforms,
+                                          batch_transforms, batch_size, shuffle,
+                                          drop_last, drop_empty, num_classes,
+                                          with_background, **kwargs)
 
 
 @register
@@ -171,10 +178,12 @@ class EvalReader(BaseDataLoader):
                  drop_last=True,
                  drop_empty=True,
                  num_classes=81,
-                 with_background=True):
-        super(EvalReader, self).__init__(
-            inputs_def, sample_transforms, batch_transforms, batch_size,
-            shuffle, drop_last, drop_empty, num_classes, with_background)
+                 with_background=True,
+                 **kwargs):
+        super(EvalReader, self).__init__(inputs_def, sample_transforms,
+                                         batch_transforms, batch_size, shuffle,
+                                         drop_last, drop_empty, num_classes,
+                                         with_background, **kwargs)
 
 
 @register
@@ -188,7 +197,9 @@ class TestReader(BaseDataLoader):
                  drop_last=False,
                  drop_empty=True,
                  num_classes=81,
-                 with_background=True):
-        super(TestReader, self).__init__(
-            inputs_def, sample_transforms, batch_transforms, batch_size,
-            shuffle, drop_last, drop_empty, num_classes, with_background)
+                 with_background=True,
+                 **kwargs):
+        super(TestReader, self).__init__(inputs_def, sample_transforms,
+                                         batch_transforms, batch_size, shuffle,
+                                         drop_last, drop_empty, num_classes,
+                                         with_background, **kwargs)

ppdet/data/source/coco.py

@@ -28,18 +28,9 @@ class COCODataSet(DetDataset):
                  dataset_dir=None,
                  image_dir=None,
                  anno_path=None,
-                 mixup_epoch=-1,
-                 cutmix_epoch=-1,
-                 mosaic_epoch=-1,
                  sample_num=-1):
-        super(COCODataSet, self).__init__(
-            dataset_dir,
-            image_dir,
-            anno_path,
-            sample_num,
-            mixup_epoch=mixup_epoch,
-            cutmix_epoch=cutmix_epoch,
-            mosaic_epoch=mosaic_epoch)
+        super(COCODataSet, self).__init__(dataset_dir, image_dir, anno_path,
+                                          sample_num)
         self.load_image_only = False
         self.load_semantic = False
 

ppdet/data/source/dataset.py

@@ -33,9 +33,6 @@ class DetDataset(Dataset):
                  anno_path=None,
                  sample_num=-1,
                  use_default_label=None,
-                 mixup_epoch=-1,
-                 cutmix_epoch=-1,
-                 mosaic_epoch=-1,
                  **kwargs):
         super(DetDataset, self).__init__()
         self.dataset_dir = dataset_dir if dataset_dir is not None else ''
@@ -44,9 +41,6 @@ class DetDataset(Dataset):
         self.sample_num = sample_num
         self.use_default_label = use_default_label
         self.epoch = 0
-        self.mixup_epoch = mixup_epoch
-        self.cutmix_epoch = cutmix_epoch
-        self.mosaic_epoch = mosaic_epoch
 
     def __len__(self, ):
         return len(self.roidbs)
@@ -77,6 +71,11 @@ class DetDataset(Dataset):
             out[k] = roidb[k]
         return out.values()
 
+    def set_kwargs(self, **kwargs):
+        self.mixup_epoch = kwargs.get('mixup_epoch', -1)
+        self.cutmix_epoch = kwargs.get('cutmix_epoch', -1)
+        self.mosaic_epoch = kwargs.get('mosaic_epoch', -1)
+
     def set_out(self, sample_transform, fields):
         self.transform = sample_transform
         self.fields = fields

ppdet/modeling/backbone/darknet.py

@@ -4,7 +4,7 @@ import paddle.nn.functional as F
 from paddle import ParamAttr
 from paddle.regularizer import L2Decay
 from ppdet.core.workspace import register, serializable
-from ppdet.modeling.ops import BatchNorm
+from ppdet.modeling.ops import batch_norm
 
 __all__ = ['DarkNet', 'ConvBNLayer']
 
@@ -31,7 +31,7 @@ class ConvBNLayer(nn.Layer):
             groups=groups,
             weight_attr=ParamAttr(name=name + '.conv.weights'),
             bias_attr=False)
-        self.batch_norm = BatchNorm(ch_out, norm_type=norm_type, name=name)
+        self.batch_norm = batch_norm(ch_out, norm_type=norm_type, name=name)
         self.act = act
 
     def forward(self, inputs):

ppdet/modeling/ops.py

@@ -29,40 +29,26 @@ import numpy as np
 from functools import reduce
 
 __all__ = [
-    'roi_pool',
-    'roi_align',
-    'prior_box',
-    'anchor_generator',
-    'generate_proposals',
-    'iou_similarity',
-    'box_coder',
-    'yolo_box',
-    'multiclass_nms',
-    'distribute_fpn_proposals',
-    'collect_fpn_proposals',
-    'matrix_nms',
-    'BatchNorm',
+    'roi_pool', 'roi_align', 'prior_box', 'anchor_generator',
+    'generate_proposals', 'iou_similarity', 'box_coder', 'yolo_box',
+    'multiclass_nms', 'distribute_fpn_proposals', 'collect_fpn_proposals',
+    'matrix_nms', 'batch_norm'
 ]
 
 
-class BatchNorm(nn.Layer):
-    def __init__(self, ch, norm_type='bn', name=None):
-        super(BatchNorm, self).__init__()
-        bn_name = name + '.bn'
-        if norm_type == 'sync_bn':
-            batch_norm = nn.SyncBatchNorm
-        else:
-            batch_norm = nn.BatchNorm2D
-
-        self.batch_norm = batch_norm(
-            ch,
-            weight_attr=ParamAttr(
-                name=bn_name + '.scale', regularizer=L2Decay(0.)),
-            bias_attr=ParamAttr(
-                name=bn_name + '.offset', regularizer=L2Decay(0.)))
-
-    def forward(self, x):
-        return self.batch_norm(x)
+def batch_norm(ch, norm_type='bn', name=None):
+    bn_name = name + '.bn'
+    if norm_type == 'sync_bn':
+        batch_norm = nn.SyncBatchNorm
+    else:
+        batch_norm = nn.BatchNorm2D
+
+    return batch_norm(
+        ch,
+        weight_attr=ParamAttr(
+            name=bn_name + '.scale', regularizer=L2Decay(0.)),
+        bias_attr=ParamAttr(
+            name=bn_name + '.offset', regularizer=L2Decay(0.)))
 
 
 def roi_pool(input,

tools/train.py

@@ -156,7 +156,7 @@ def run(FLAGS, cfg, place):
     start_epoch = optimizer.state_dict()['LR_Scheduler']['last_epoch']
     for epoch_id in range(int(cfg.epoch)):
         cur_eid = epoch_id + start_epoch
-        train_loader.dataset.epoch = epoch_id
+        train_loader.dataset.epoch = cur_eid
         for iter_id, data in enumerate(train_loader):
             start_time = end_time
             end_time = time.time()