YunYang1994/TensorFlow2.0-Examples

Issues

• 这个没有download怎么下载

  #147 opened a year ago by terrydy9527
  0

• Can we train with multiple scale training procedure?

  #115 opened 4 years ago by tranleanh
  1

• 请问为什么求conf_loss和prob_loss时的logits用的是feature map的输出而不用decode后得到的概率值，感谢！

  #145 opened 2 years ago by Kerwin-hanhan
  0

• YOLOv3 model doesn't converge

  #127 opened 3 years ago by JVD9kh96
  7

• 运行video_demo.py时，只有第一帧预测正确后面都是nan，也就是后面没有预测框

  #138 opened 3 years ago by mastergao57
  1

• Error when training with my own dataset using train.py: Failed to get convolution algorithm.

  #140 opened 3 years ago by Guanzhao-hub
  0

• yolov3 mAP on PASCAL VOC2012

  #139 opened 3 years ago by Phoenix0qi
  0

• YOLOv3中训练自己的数据集时可以调用官方的yolov3.weights,在它的基础上进行训练吗？

  #87 opened 5 years ago by monoloxo
  8

• 运行test.py只有第一张图片有测试结果，剩下的图片都没有检测框出现

  #122 opened 4 years ago by taojianggit
  3

• test.py nan

  #121 opened 4 years ago by huhuhuhuha
  2

• Problem about ResNet.

  #134 opened 3 years ago by A-cunminF
  0

• 本人小白，请教迁移学习，初始化权重训练自己的数据！

  #133 opened 3 years ago by jsago
  0

• Question about the image_preprocess function

  #132 opened 3 years ago by JVD9kh96
  0

• 关于SSD卷积代替全连接层的请教

  #131 opened 3 years ago by junnyin
  0

• Confused about convolutional layer implementation in YOLOv3

  #130 opened 3 years ago by s-ilic
  0

• 请问大家是怎么自制数据集的呢

  #118 opened 4 years ago by NiceRingNode
  1

• InvalidArgumentError: ConcatOp : Dimensions of inputs should match: shape[0] = [2704,54] vs. shape[1] = [8112,18] [Op:ConcatV2] name: concat

  #126 opened 3 years ago by VedVyapak
  0

• 训练自己的模型时，运行image_demo.py,找不到yolov3.weights文件

  #112 opened 4 years ago by xuxiaodong248
  1

• StartAbort Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above

  #99 opened 4 years ago by clemente0731
  3

• 运行image_demo.py报错failed to read all data

  #110 opened 4 years ago by futureflsl
  1

• 请问如何计算FLOPs?

  #123 opened 4 years ago by taojianggit
  0

• 关于 postprocess_boxes(pred_bbox, org_img_shape, input_size, score_threshold)

  #120 opened 4 years ago by StarStar-666
  0

• 为什么这里没有开启多尺度训练了？

  #119 opened 4 years ago by StarStar-666
  0

• anxious for transformer

  #117 opened 4 years ago by VV20192019
  0

• yolov3模型定义方法与其他检测模型不同

  #113 opened 4 years ago by hnuhchen
  2

• Training on custom dataset

  #114 opened 4 years ago by sampreets3
  0

• 请教 FCN训练时出错

  #93 opened 4 years ago by huihua7882
  3

• 在yolov3.py文件中第177行，iou = bbox_iou(pred_xywh[:, :, :, :, np.newaxis, :], bboxes[:, np.newaxis, np.newaxis, np.newaxis, :, :])是不是有错误？bboxes的维度传进来的时候是（self.max_bbox_per_scale, 4)，但在此处，显然是认为有三个维度。还有pred_xywh的第零维和bboxes的第零维不相同(pred_xywh.shape[0] != bboxes.shape[0])，还能做iou吗？

  #102 opened 4 years ago by dujifeng
  1

• TF2.0_YOLOv3 main.py不存在

  #109 opened 4 years ago by dengss33
  1

• yyminst应该放到哪个文件夹下？谢谢，小白一枚

  #96 opened 4 years ago by litianzhu123
  1

• yolov3 evaluation

  #108 opened 4 years ago by SeHwanJoo
  0

• YoloV3 saving weights

  #86 opened 5 years ago by FishJyz
  4

• YOLOv3模型训练好并保存后，再加载模型预测的输出shape变了(直接编译运行错误了)

  #106 opened 4 years ago by WorstCodeWay
  0

• FCN training problem

  #105 opened 4 years ago by CJAPPLE5
  0

• 训练中，giou_loss 震荡不下降，徘徊在 12 左右

  #98 opened 4 years ago by forever208
  6

• 2080ti batchsize最大只能选12

  #103 opened 4 years ago by pp1bb
  0

• RPN for multi class object detection

  #101 opened 4 years ago
  0

• make_data.py 报list index out of range

  #100 opened 4 years ago by ShirleyHe2020
  0

• yolov3 support tf-gpu ==2.0 ?

  #97 opened 4 years ago by gyshi
  0

• 请问出现这个是为什么呢？image = cv2.resize(image, (int(28*ratio), int(28*ratio))) cv2.error: OpenCV(4.1.1) C:\projects\opencv-python\opencv\modules\imgproc\src\resize.cpp:3720: error: (-215:Assertion failed) !ssize.empty() in function 'cv::resize'

  #95 opened 4 years ago by litianzhu123
  1

• YOLOv3的程序中在train.py的model.save_weights()函数里保存成.h5文件就会出错这是为什么

  #94 opened 4 years ago by pp1bb
  0

• yolov3 giou loss计算的问题

  #92 opened 5 years ago by hbo-lambda
  2

• RPN loss compute

  #90 opened 5 years ago by haigh1510
  0

• Can't predict anything when running test.py

  #89 opened 5 years ago by warriors30n11
  1

• 请问，修改输入图片尺寸[256, 512]的话，需要怎么改？多次尝试都失败

  #88 opened 5 years ago by bytehuang0912
  1

• the position of activation in residual_block (残差网络中激活层位置的问题)

  #84 opened 5 years ago by gorgeous369
  0

• Faster R-CNN page 404

  #83 opened 5 years ago by FishJyz
  0

• Zeropadding in common.py

  #82 opened 5 years ago by karan-shr
  0

• YOLO strange point in implementation!

  #81 opened 5 years ago by aif2017
  0

• About ResNet

  #80 opened 5 years ago by Jian-Yin-Shine
  0