lucasjinreal/keras_frcnn

Issues

• Question for batch training

  #132 opened 2 years ago by kunzeng-ch
  0

• Pytorch vision

  #119 opened 2 years ago by YilinGao-SHU
  0

• How to download dataset kitti ?

  #34 opened 6 years ago by ThorPham
  3

• Bounding boxes dictionary empty

  #63 opened 5 years ago by drprajapati
  3

• Heatmap visualization issue

  #95 opened 4 years ago by SiwarBG-2019
  0

• Can anyone provides your model file or before model file in this work ? Thanks !

  #90 opened 4 years ago by DongLiangTu
  0

• Can not install Keras 2.0.3

  #52 opened 6 years ago by DeZhao-Zhang
  2

• I cannot train the model, and always got error like this! Any info will be appreciated!

  #45 opened 6 years ago by xincxiong
  3

• AttributeError: Can't get attribute '"Config"'

  #80 opened 5 years ago by kongzhiyou
  1

• learning rate: 1e-5?

  #76 opened 5 years ago by Try2ChangeX
  0

• How long does it take to train this model for 2000 epochs on GTX1080Ti? I have trained 10 days, but only 1121/2000.

  #75 opened 5 years ago by Try2ChangeX
  0

• No boxes when predict

  #74 opened 5 years ago by AntoineMath
  0

• How many epochs do I need to train?

  #73 opened 5 years ago by tossy-yossy
  0

• 图片存在，txt文件满足训练格式，但训练时报错NoneType' object has no attribute 'shape，这是为什么？

  #71 opened 5 years ago by jypython
  2

• where is the test_frcnn_kitti.py

  #72 opened 5 years ago by StephenLiubjtu
  1

• How to adjust the batchsize to train?

  #62 opened 5 years ago by ryosism
  1

• Unable to locate file test_frcnn_kitti.py

  #65 opened 5 years ago by NaveenGandla
  3

• 关于训练数据一张图片不止一类的问题

  #70 opened 5 years ago by jypython
  0

• the usage of data_gen_val

  #42 opened 6 years ago by lxy5513
  1

• ValueError: Shape must be rank 1 but is rank 0 for 'bn_conv1/Reshape_4' (op: 'Reshape') with input shapes: [1,1,1,64], [].

  #58 opened 6 years ago by 2017TJM
  2

• please how to use measrue_map.py

  #69 opened 5 years ago by zhangyufei1995
  2

• 这个config配置文件中self.std_scaling是用来做什么的?

  #49 opened 6 years ago by 17S051031
  2

• How to use measure_map.py?

  #36 opened 6 years ago by kelisiya
  7

• 请问有计算recall、mAP

  #56 opened 6 years ago by huanglihegn
  1

• 请问有计算mAP的代码吗？

  #57 opened 6 years ago by huanglihegn
  1

• 为啥跑的图片尺度不一样会跑的比较慢？

  #61 opened 5 years ago by Larry-Zheng
  2

• The bbox may not have positive RPN region. In data_generators.py, line 208, num_anchors_for_bbox[idx] can be more than 0 and the idx-th bbox does not have the correponding anchor.

  #68 opened 5 years ago by ironshangki
  0

• Extracting region proposals from an image.

  #67 opened 5 years ago by djongpo
  0

• Anchor out of image size

  #64 opened 5 years ago by syifalaili26
  0

• only png? not support jpg?

  #60 opened 5 years ago by zhangyufei1995
  0

• 有预训练权重吗？

  #59 opened 5 years ago by xianfenggao
  0

• train own dataset but cbf find hdf5

  #48 opened 6 years ago by ycAlex11
  3

• 照抄别人的代码，就不能加个引用吗？可恶

  #51 opened 6 years ago by jackros1022
  1

• 求助怎样进行预测？

  #55 opened 6 years ago by sheldonnnn
  0

• Is it possible that the rpn training loss or classifier training loss is smaller than 0?

  #53 opened 6 years ago by white2018
  0

• 小目标检测

  #54 opened 6 years ago by Yangget
  0

• Exception: Error when checking target: expected rpn_out_class to have shape (None, None, None, 9) but got array with shape (1, 38, 124, 18)

  #38 opened 6 years ago by moses1994
  1

• The test image can not detect object?

  #31 opened 6 years ago by longzeyilang
  1

• 这份代码跑VOC2007或者VOC2012最后指标能跑到多少？

  #50 opened 6 years ago by ljquality
  0

• 请问如果自己的数据集中一张图片上出现的物体不止一个的情况下，标签是什么样子的呢

  #41 opened 6 years ago by Bigwode
  1

• How's the performance on Pascal VOC dataset?

  #47 opened 6 years ago by zbxzc35
  0

• Exception: a must be non-empty

  #39 opened 6 years ago by lengyue1996
  1

• Mask R-#

  #46 opened 6 years ago by MjdMahasneh
  0

• Get feature-map of a given region in the image from the ground truth

  #43 opened 6 years ago by praisethemoon
  0

• background detection in roi_helpers.py

  #33 opened 6 years ago by chiayuhsu
  1

• Validation Accuracy or Early Stopping

  #40 opened 6 years ago by nassarofficial
  0

• Is there any difference between the code and original paper?

  #37 opened 6 years ago by ccnankai
  0

• X, Y = np.meshgrid(np.arange(cols), np.arange(rows)) why not offset from center?

  #35 opened 6 years ago by ZHANGKEON
  0

• Where to find the kiti data set you talk about ?

  #32 opened 6 years ago by trineldave
  0

• How do we train an individual image without changing the default path of the test_frcnn directory? Also how to set path to multiple images? My image is saved in the designated folder as image.png

  #30 opened 6 years ago by ShreshthKhare
  0