HiKapok/SSD.TensorFlow

Issues

• voc_eval.py 211行出现keyerror: File "voc_eval.py", line 211, in voc_eval R = class_recs[image_ids[d]] KeyError: '000000117121'

  #140 opened 4 years ago by yangguoqing666
  3

• How can I use multi-GPUs to train the ssd model?

  #124 opened 5 years ago by cemiboou
  1

• 怎样把Tensor转换成NDArray

  #141 opened 3 years ago by MSTK2014
  0

• TypeError: `default_name` type (<class 'list'>) is not a string type. When run train_ssd.py, I meet a TpyeError.

  #117 opened 5 years ago by CaptainKeer
  10

• tensorflow.python.framework.errors_impl.NotFoundError: Could not find directory

  #142 opened 3 years ago by ZhaolinWang0110
  0

• 请问我在执行eval_ssd.py遇到的问题，下面是报错内容

  #139 opened 4 years ago by chenjjjjuu
  0

• 请问我在执行eval——

  #138 opened 4 years ago by chenjjjjuu
  0

• InvalidArgumentError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error: Assign requires shapes of both tensors to match. lhs shape= [1,1,1024,1024] rhs shape= [1,1,4096,4096] [[node save/Assign_29 (defined at E:\Infrared_Deeplearning\SSD-AbsoluteCoord\utility\scaffolds.py:86) ]]

  #126 opened 5 years ago by navy63
  2

• Hello, can this network run under win10 + CPU environment?

  #137 opened 4 years ago by meng0li
  0

• may i have a script for transferring .ckpt to .pb

  #136 opened 4 years ago by feifeiwei
  0

• How to continue the last training？

  #133 opened 5 years ago by 18846935350
  1

• TypeError: Expected binary or unicode string, got None

  #135 opened 4 years ago by zimozhou
  0

• 请问如何在GPU上运行？

  #130 opened 5 years ago by chenjjjjuu
  1

• eval_ssd.py 生成的预测结果文件中没有filename

  #134 opened 5 years ago by leileileidan
  1

• voc_eval.py on my own dataset

  #120 opened 5 years ago by Kasyene
  0

• How to train my own Dataset?

  #132 opened 5 years ago by zhang295498
  1

• the pixel of train_image_size is 256

  #125 opened 5 years ago by tristehai
  0

• How to train my own

  #131 opened 5 years ago by zhang295498
  0

• ERROR:Expected binary or unicode string, got None

  #121 opened 5 years ago by yang1015661763
  2

• Why there are no celoss ,locloss and loss images after I finish training on my own dataset?

  #129 opened 5 years ago by Taylor0424
  0

• 为什么多GPU并行报OOM

  #128 opened 5 years ago by DataXujing
  0

• 为什么开始训练后执行到一定情况就会卡住，然后CPU占用率满，但是gpu为0

  #127 opened 5 years ago by bubbliiiing
  1

• Is the coordinate position of the detection box upper left corner, lower right corner or upper left corner with width and height?

  #123 opened 5 years ago by AndyHon
  0

• I can't get the weights of the model.

  #122 opened 5 years ago by WangFengtu1996
  1

• train_ssd.py loss=NaN or The program does not continue or stop

  #119 opened 5 years ago by chenaifang
  3

• 我在自己的样本上测试时，使用了没有groundtruth的XML文件，所以在执行eval_ssd.py时报错了

  #118 opened 5 years ago by fanfan00
  3

• Detect an image by trained model?

  #116 opened 5 years ago by xi3299
  1

• would you explain the modified_smooth_l1?

  #115 opened 5 years ago by xiaxia111
  0

• Input nodes missing in frozen graph

  #114 opened 5 years ago by sjmjk429
  1

• why not any info printed in traning?

  #112 opened 5 years ago by zmhik
  0

• TypeError: Expected binary or unicode string, got None

  #103 opened 5 years ago by ydhongHIT
  2

• Unhandled cases???

  #111 opened 5 years ago by lloydnguyen96
  0

• lr_rate

  #110 opened 5 years ago by ydhongHIT
  0

• questions about voc_eval.py

  #107 opened 5 years ago by ydhongHIT
  1

• when complete this project

  #109 opened 5 years ago by zhuxiaoyangcute
  0

• Exception thrown if inference with batch size other than 1

  #108 opened 5 years ago by mandy-li
  0

• How can I train from scratch?

  #106 opened 5 years ago by chengcchn
  0

• join some no labeled image to train dataset

  #97 opened 5 years ago by duoduo1234
  2

• How to train without pretrained checkpoint

  #105 opened 5 years ago by ydhongHIT
  0

• some questions about the eval_ssd.py

  #104 opened 5 years ago by ydhongHIT
  1

• Bug: Using images out of ImageSets\Main\trainval.txt for training

  #102 opened 5 years ago by jamiechoi1995
  1

• gpu fraction

  #101 opened 5 years ago by zhengduoru
  2

• gpu fraction

  #100 opened 5 years ago by zhengduoru
  0

• How to run train from scratch (without pretrained model) ?

  #99 opened 5 years ago by jfpalngipang
  1

• train_ssd Error

  #98 opened 5 years ago by fuaiguo
  5

• Training stopping by itself after (250/batch_size) steps

  #94 opened 5 years ago by tomrichardon
  7

• the loss is aroung 20 for a long time

  #96 opened 5 years ago by jamiechoi1995
  1

• why is the select_threshold 2

  #95 opened 5 years ago by tutu123133
  3

• 关于hard negative mining过程的疑问，代码疑似有问题

  #92 opened 5 years ago by okfang
  2

• Default MaxPoolingOp only supports NHWC on device type CPU

  #93 opened 5 years ago by wangjing60755
  1