SSD-Tensorflow
SSD-Tensorflow copied to clipboard
Single Shot MultiBox Detector in TensorFlow
I clone the project, and run the ssd minimal example in docker, but it had error.As follows: DataLossError: Unable to open table file ../checkpoints/ssd_300_vgg.ckpt: Failed precondition: ../checkpoints/ssd_300_vgg.ckpt: perhaps your file...
SPLITS_TO_SIZES = { 'train': 5000, 'test': 1921, } what this code meaning? i cannot find the usage of it, do anyone can help me?
when i use original model, i can got a good bbox results, but with my model, i can get feedforward outputs like this, rpredictions [array([[[[[ 0.99024117, 0.00975887], [ 0.98918754, 0.01081248],...
The set parameters: {'adadelta_rho': 0.95, 'adagrad_initial_accumulator_value': 0.1, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'batch_size': 16, 'checkpoint_exclude_scopes': None, 'checkpoint_model_scope': 'ssd_300_vgg', 'checkpoint_path': './checkpoints/ssd_300_vgg.ckpt/ssd_300_vgg.ckpt', 'clone_on_cpu': False, 'dataset_dir': './tfrecords_voc', 'dataset_name': 'pascalvoc_2007', 'dataset_split_name': 'train', 'end_learning_rate': 1e-06, 'ftrl_initial_accumulator_value':...
When i test my new model,it display an error: Failed to run optimizer ArithmeticOptimizer, stage HoistCommonFactor . Error: Node average_precision_voc07/ArithmeticOptimizer/HoistCommonFactor_Add_AddN is missing output properties at position :0 (num_outputs=0) I really...
zhe@amax1:~/SSD-Tensorflow-master$ python eval_ssd_network.py --eval_dir=./logs --dataset_dir=./test --datset_name=pascalvoc_2007 --dataset_split_name=test --model_name=ssd_300_vgg --checkpoint_path=./checkpoints/ssd_300_vgg.ckpt --batch_size=1 /home/fancy/program/anaconda2/lib/python2.7/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be...
The shell raises the error that " FileNotFoundError: [Errno 2] No such file or directory: './VOC2007/test/Annotations/' ". How can I solve this problem?
Hi, First of all, Thanks for your excellent implementation ! I already made VOC2007 TFRecord files. Then, I have a question how to make VOC 07+12 TFRecord that is union...
I am wondering if anyone can reproduce the same mAP as the original paper from this repo's training script from ImageNet pretrained weights. I find it really difficult to converge...