蒸馏报错has no is_crowd field
Created by: thunder95
我训练了个faster-rcnn目标检测模型,想用它蒸馏yolov3_mobilenet_v1
运行命令如下:
!python slim/distillation/distill.py
-c myconfigs/yolov3_mobilenet_v1_voc_distill.yml
-t myconfigs/faster_rcnn_dcn_r50_fpn_1x.yml
-o YOLOv3Loss.batch_size=32
--teacher_pretrained output/faster_rcnn_dcn_r50_fpn_1x/model_final
报错如下: Traceback (most recent call last): File "slim/distillation/distill.py", line 401, in main() File "slim/distillation/distill.py", line 211, in main train_fetches = model.train(teacher_feed_vars) File "/home/aistudio/work/PaddleDetection/ppdet/modeling/architectures/faster_rcnn.py", line 241, in train return self.build(feed_vars, 'train') File "/home/aistudio/work/PaddleDetection/ppdet/modeling/architectures/faster_rcnn.py", line 73, in build self._input_check(required_fields, feed_vars) File "/home/aistudio/work/PaddleDetection/ppdet/modeling/architectures/faster_rcnn.py", line 190, in _input_check "{} has no {} field".format(feed_vars, var) AssertionError: OrderedDict([('image', name: "image