ValueError: No variables to save

[PiroAntoine]

Hello,

I would like to create a dataset (the same way Pascal VOC) and train a res101 on it.

I wrote a cpp code to create my dataset and call the train_faster_rcnn.sh

But I meet this error : "ValueError: No variables to save"

Here is my log file :

• echo Logging output to tf-faster-rcnn/experiments/logs/res101_voc_2007_trainval__res101.txt.2018-01-12_16-01-23 Logging output to tf-faster-rcnn/experiments/logs/res101_voc_2007_trainval__res101.txt.2018-01-12_16-01-23
• set +x
• '[' '!' -f tf-faster-rcnn/output/res101/voc_2007_trainval/default/res101_faster_rcnn_iter_70000.ckpt.index ']'
• [[ ! -z '' ]]
• CUDA_VISIBLE_DEVICES=0
• time python2 ./tf-faster-rcnn/tools/trainval_net.py --weight /home/antoine/Documents/RCNNProject/DataFolder/imagenet_weights/res101.ckpt Called with args: Namespace(cfg_file=None, imdb_name='voc_2007_trainval', imdbval_name='voc_2007_test', max_iters=70000, net='res50', set_cfgs=None, tag=None, weight='/home/antoine/Documents/RCNNProject/DataFolder/imagenet_weights/res101.ckpt') Using config: {'ANCHOR_RATIOS': [0.5, 1, 2], 'ANCHOR_SCALES': [8, 16, 32], 'DATA_DIR': '/home/antoine/Documents/RCNNProject/DataFolder/', 'EXP_DIR': 'default', 'MATLAB': 'matlab', 'MOBILENET': {'DEPTH_MULTIPLIER': 1.0, 'FIXED_LAYERS': 5, 'REGU_DEPTH': False, 'WEIGHT_DECAY': 4e-05}, 'PIXEL_MEANS': array([[[ 102.9801, 115.9465, 122.7717]]]), 'POOLING_MODE': 'crop', 'POOLING_SIZE': 7, 'RESNET': {'FIXED_BLOCKS': 1, 'MAX_POOL': False}, 'RNG_SEED': 3, 'ROOT_DIR': '/home/antoine/Documents/RCNNProject/project-build/tf-faster-rcnn', 'RPN_CHANNELS': 512, 'TEST': {'BBOX_REG': True, 'HAS_RPN': False, 'MAX_SIZE': 1000, 'MODE': 'nms', 'NMS': 0.3, 'PROPOSAL_METHOD': 'gt', 'RPN_NMS_THRESH': 0.7, 'RPN_POST_NMS_TOP_N': 300, 'RPN_PRE_NMS_TOP_N': 6000, 'RPN_TOP_N': 5000, 'SCALES': [600], 'SVM': False}, 'TRAIN': {'ASPECT_GROUPING': False, 'BATCH_SIZE': 128, 'BBOX_INSIDE_WEIGHTS': [1.0, 1.0, 1.0, 1.0], 'BBOX_NORMALIZE_MEANS': [0.0, 0.0, 0.0, 0.0], 'BBOX_NORMALIZE_STDS': [0.1, 0.1, 0.2, 0.2], 'BBOX_NORMALIZE_TARGETS': True, 'BBOX_NORMALIZE_TARGETS_PRECOMPUTED': True, 'BBOX_REG': True, 'BBOX_THRESH': 0.5, 'BG_THRESH_HI': 0.5, 'BG_THRESH_LO': 0.1, 'BIAS_DECAY': False, 'DISPLAY': 10, 'DOUBLE_BIAS': True, 'FG_FRACTION': 0.25, 'FG_THRESH': 0.5, 'GAMMA': 0.1, 'HAS_RPN': True, 'IMS_PER_BATCH': 1, 'LEARNING_RATE': 0.001, 'MAX_SIZE': 1000, 'MOMENTUM': 0.9, 'PROPOSAL_METHOD': 'gt', 'RPN_BATCHSIZE': 256, 'RPN_BBOX_INSIDE_WEIGHTS': [1.0, 1.0, 1.0, 1.0], 'RPN_CLOBBER_POSITIVES': False, 'RPN_FG_FRACTION': 0.5, 'RPN_NEGATIVE_OVERLAP': 0.3, 'RPN_NMS_THRESH': 0.7, 'RPN_POSITIVE_OVERLAP': 0.7, 'RPN_POSITIVE_WEIGHT': -1.0, 'RPN_POST_NMS_TOP_N': 2000, 'RPN_PRE_NMS_TOP_N': 12000, 'SCALES': [600], 'SNAPSHOT_ITERS': 5000, 'SNAPSHOT_KEPT': 3, 'SNAPSHOT_PREFIX': 'res101_faster_rcnn', 'STEPSIZE': [30000], 'SUMMARY_INTERVAL': 180, 'TRUNCATED': False, 'USE_ALL_GT': True, 'USE_FLIPPED': True, 'USE_GT': False, 'WEIGHT_DECAY': 0.0001}, 'USE_GPU_NMS': True} Loaded dataset voc_2007_trainval for training Set proposal method: gt Appending horizontally-flipped training examples... wrote gt roidb to /home/antoine/Documents/RCNNProject/DataFolder/cache/voc_2007_trainval_gt_roidb.pkl done Preparing training data... done 2348 roidb entries Output will be saved to /home/antoine/Documents/RCNNProject/project-build/tf-faster-rcnn/output/default/voc_2007_trainval/default TensorFlow summaries will be saved to /home/antoine/Documents/RCNNProject/project-build/tf-faster-rcnn/tensorboard/default/voc_2007_trainval/default Loaded dataset voc_2007_test for training Set proposal method: gt Preparing training data... wrote gt roidb to /home/antoine/Documents/RCNNProject/DataFolder/cache/voc_2007_test_gt_roidb.pkl done 293 validation roidb entries Filtered 0 roidb entries: 2348 -> 2348 Filtered 0 roidb entries: 293 -> 293 2018-01-12 16:01:26.192323: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA Solving... /home/antoine/.local/lib/python2.7/site-packages/tensorflow/python/ops/gradients_impl.py:96: 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 /home/antoine/Documents/RCNNProject/DataFolder/imagenet_weights/res101.ckpt Traceback (most recent call last): File "./tf-faster-rcnn/tools/trainval_net.py", line 139, in max_iters=args.max_iters) File "/home/antoine/Documents/RCNNProject/project-build/tf-faster-rcnn/tools/../lib/model/train_val.py", line 391, in train_net sw.train_model(sess, max_iters) File "/home/antoine/Documents/RCNNProject/project-build/tf-faster-rcnn/tools/../lib/model/train_val.py", line 268, in train_model sess) File "/home/antoine/Documents/RCNNProject/project-build/tf-faster-rcnn/tools/../lib/model/train_val.py", line 201, in initialize restorer = tf.train.Saver(variables_to_restore) File "/home/antoine/.local/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1218, in init self.build() File "/home/antoine/.local/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1227, in build self._build(self._filename, build_save=True, build_restore=True) File "/home/antoine/.local/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1251, in _build raise ValueError("No variables to save") ValueError: No variables to save Command exited with non-zero status 1 8.54user 1.10system 0:07.82elapsed 123%CPU (0avgtext+0avgdata 724440maxresident)k 0inputs+8320outputs (0major+174129minor)pagefaults 0swaps

Can someone tell me what can be going on ? :)

Thank you

[minda163]

Did you solve it？I have the same issue

[cnuzh]

Did you solve it？I have the same issue Hi, I have the same issue. Did you solve it?

[devendraswamy]

!python /content/drive/MyDrive/tf-faster-rcnn/tools/trainval_net.py --net vgg16 --weight /content/drive/MyDrive/tf-faster-rcnn/data/imagenet_weights/vgg16.ckpt --imdb voc_2007_trainval --imdbval voc_2007_test

please make sure the --net and --weight is vgg16 & vgg.ckpt are like this. thank you