[FasterRCNN] also support deeper resnet

examples/FasterRCNN/basemodel.py

@@ -91,11 +91,11 @@ def pretrained_resnet_conv4(image, num_blocks):
     return l
 
 
-def resnet_conv5(image):
+def resnet_conv5_gap(image, num_block):
     with argscope([Conv2D, GlobalAvgPooling, BatchNorm], data_format='NCHW'), \
             argscope(Conv2D, nl=tf.identity, use_bias=False), \
             argscope(BatchNorm, use_local_stat=False):
         # 14x14:
-        l = resnet_group(image, 'group3', resnet_bottleneck, 512, 3, stride=2)
+        l = resnet_group(image, 'group3', resnet_bottleneck, 512, num_block, stride=2)
         l = GlobalAvgPooling('gap', l)
         return l

examples/FasterRCNN/config.py

@@ -10,6 +10,9 @@ TRAIN_DATASET = ['train2014', 'valminusminival2014']
 VAL_DATASET = 'minival2014'   # only support evaluation on one dataset
 NUM_CLASS = 81
 
+# basemodel ----------------------
+RESNET_NUM_BLOCK = [3, 4, 6, 3]     # resnet50
+
 # preprocessing --------------------
 SHORT_EDGE_SIZE = 600
 MAX_SIZE = 1024

examples/FasterRCNN/train.py

@@ -25,7 +25,7 @@ from tensorpack.utils.gpu import get_nr_gpu
 
 from coco import COCODetection
 from basemodel import (
-    image_preprocess, pretrained_resnet_conv4, resnet_conv5)
+    image_preprocess, pretrained_resnet_conv4, resnet_conv5_gap)
 from model import (
     rpn_head, rpn_losses,
     decode_bbox_target, encode_bbox_target,
@@ -87,8 +87,7 @@ class Model(ModelDesc):
 
         fm_anchors = self._get_anchors(image)
         anchor_boxes_encoded = encode_bbox_target(anchor_boxes, fm_anchors)
-        # resnet50
-        featuremap = pretrained_resnet_conv4(image, [3, 4, 6])
+        featuremap = pretrained_resnet_conv4(image, config.RESNET_NUM_BLOCK[:3])
         rpn_label_logits, rpn_box_logits = rpn_head(featuremap, 1024, config.NR_ANCHOR)
         rpn_label_loss, rpn_box_loss = rpn_losses(
             anchor_labels, anchor_boxes_encoded, rpn_label_logits, rpn_box_logits)
@@ -104,7 +103,7 @@ class Model(ModelDesc):
                 proposal_boxes, gt_boxes, gt_labels)
             boxes_on_featuremap = rcnn_sampled_boxes * (1.0 / config.ANCHOR_STRIDE)
             roi_resized = roi_align(featuremap, boxes_on_featuremap, 14)
-            feature_fastrcnn = resnet_conv5(roi_resized)    # nxc
+            feature_fastrcnn = resnet_conv5_gap(roi_resized, config.RESNET_NUM_BLOCK[-1])    # nxc
             fastrcnn_label_logits, fastrcnn_box_logits = fastrcnn_head(feature_fastrcnn, config.NUM_CLASS)
 
             fastrcnn_label_loss, fastrcnn_box_loss = fastrcnn_losses(
@@ -123,7 +122,7 @@ class Model(ModelDesc):
                 add_moving_summary(k)
         else:
             roi_resized = roi_align(featuremap, proposal_boxes * (1.0 / config.ANCHOR_STRIDE), 14)
-            feature_fastrcnn = resnet_conv5(roi_resized)    # nxc
+            feature_fastrcnn = resnet_conv5_gap(roi_resized, config.RESNET_NUM_BLOCK[-1])    # nxc
             label_logits, fastrcnn_box_logits = fastrcnn_head(feature_fastrcnn, config.NUM_CLASS)
             label_probs = tf.nn.softmax(label_logits, name='fastrcnn_all_probs')  # NP,
             labels = tf.argmax(label_logits, axis=1)

tensorpack/tfutils/sessinit.py

@@ -273,4 +273,5 @@ def TryResumeTraining():
     path = os.path.join(logger.get_logger_dir(), 'checkpoint')
     if not tf.gfile.Exists(path):
         return JustCurrentSession()
+    logger.info("Found checkpoint at {}.".format(path))
     return SaverRestore(path)