raise ValueError("No variables to save")

[junjunwang2016]

When i run ./train_faster_rcnn.sh 0 coco res101_fusion EXP_DIR coco_flip_0001_bilinear_new, i meet the problem :
Solving...
/media/junjun/work/ICDE/detection/RGB-N-master/envir/lib/python3.5/site-packages/tensorflow/python/ops/gradients_impl.py:91: UserWarning: Converting sparse IndexedSlices to a dense Tensor of unknown shape. This may consume a large amount of memory.
"Converting sparse IndexedSlices to a dense Tensor of unknown shape. "
Loading initial model weights from /media/junjun/work/ICDE/detection/RGB-N-master/imagenet_weights/vgg16.ckpt
Traceback (most recent call last):
File "./tools/trainval_net.py", line 173, in
max_iters=args.max_iters)
File "/media/junjun/work/ICDE/detection/RGB-N-master/tools/../lib/model/train_val.py", line 356, in train_net
sw.train_model(sess, max_iters)
File "/media/junjun/work/ICDE/detection/RGB-N-master/tools/../lib/model/train_val.py", line 170, in train_model
restorer = tf.train.Saver(variables_to_restore)
File "/media/junjun/work/ICDE/detection/RGB-N-master/envir/lib/python3.5/site-packages/tensorflow/python/training/saver.py", line 1000, in init
self.build()
File "/media/junjun/work/ICDE/detection/RGB-N-master/envir/lib/python3.5/site-packages/tensorflow/python/training/saver.py", line 1021, in build
raise ValueError("No variables to save")