fix #1043

examples/FasterRCNN/dataset.py

@@ -7,7 +7,6 @@ import tqdm
 import json
 
 from tensorpack.utils import logger
-from tensorpack.utils.argtools import log_once
 from tensorpack.utils.timer import timed_operation
 
 from config import config as cfg
@@ -121,7 +120,7 @@ class COCODetection(object):
         valid_objs = []
         width = img['width']
         height = img['height']
-        for obj in objs:
+        for objid, obj in objs:
             if obj.get('ignore', 0) == 1:
                 continue
             x1, y1, w, h = obj['bbox']
@@ -145,8 +144,10 @@ class COCODetection(object):
                         obj['segmentation'] = None
                     else:
                         valid_segs = [np.asarray(p).reshape(-1, 2).astype('float32') for p in segs if len(p) >= 6]
-                        if len(valid_segs) < len(segs):
-                            log_once("Image {} has invalid polygons!".format(img['file_name']), 'warn')
+                        if len(valid_segs) == 0:
+                            logger.error("Object {} in image {} has no valid polygons!".format(objid, img['file_name']))
+                        elif len(valid_segs) < len(segs):
+                            logger.warn("Object {} in image {} has invalid polygons!".format(objid, img['file_name']))
 
                         obj['segmentation'] = valid_segs
 

examples/ResNet/load-resnet.py

@@ -47,10 +47,10 @@ class Model(ModelDesc):
             logits = (LinearWrap(image)
                       .Conv2D('conv0', 64, 7, strides=2, activation=BNReLU, padding='VALID')
                       .MaxPooling('pool0', 3, strides=2, padding='SAME')
-                      .apply(resnet_group, 'group0', bottleneck, 64, blocks[0], 1)
-                      .apply(resnet_group, 'group1', bottleneck, 128, blocks[1], 2)
-                      .apply(resnet_group, 'group2', bottleneck, 256, blocks[2], 2)
-                      .apply(resnet_group, 'group3', bottleneck, 512, blocks[3], 2)
+                      .apply2(resnet_group, 'group0', bottleneck, 64, blocks[0], 1)
+                      .apply2(resnet_group, 'group1', bottleneck, 128, blocks[1], 2)
+                      .apply2(resnet_group, 'group2', bottleneck, 256, blocks[2], 2)
+                      .apply2(resnet_group, 'group3', bottleneck, 512, blocks[3], 2)
                       .GlobalAvgPooling('gap')
                       .FullyConnected('linear', 1000)())
         tf.nn.softmax(logits, name='prob')

tensorpack/models/linearwrap.py

@@ -80,6 +80,9 @@ class LinearWrap(object):
         Apply a function on the wrapped tensor. The tensor
         will be the second argument of func.
 
+        This is because many symbolic functions
+        (such as tensorpack's layers) takes 'scope' as the first argument.
+
         Returns:
             LinearWrap: ``LinearWrap(func(args[0], self.tensor(), *args[1:], **kwargs))``.
         """