Issues
- 0
loss出现Nan改了很久,希望您帮忙看一下
#45 opened by Wangcancan1127 - 0
- 0
训练成功,想咨询您预测的问题
#43 opened by Wangcancan1127 - 12
- 0
自己的数据集。 在数据加载的时候。出现这个问题‘NoneType' object has no attribute 'find'’。 验证集和测试集加载不了
#42 opened by MARUIHAN123 - 1
你好,loss一直为nan,自己的数据集。什么问题呀
#41 opened by MARUIHAN123 - 3
- 2
- 1
@MARUIHAN123 就是1.19.0
#39 opened by MARUIHAN123 - 1
@dengchubo 我用的是TensorFlow 1.14.0
#38 opened by MARUIHAN123 - 7
rcnn_class_loss训练到一半就会变成nan
#37 opened by huangxudong - 2
- 9
您好,预训练模型连接失效了,可以更新一下吗,感谢
#35 opened by Wangcancan1127 - 11
执行train.py一直OOM
#32 opened by BngThea - 1
没有画出框
#34 opened by huangxudong - 0
- 4
请问如何训练不带gtbox的图片
#31 opened by ashipiling - 5
坐标归一化
#30 opened by Crispus1119 - 1
为啥我聚类聚出负数来了
#28 opened by shadow12138 - 1
acc
#27 opened by liby542698 - 2
报错 'Model' object has no attribute 'add_metric'
#26 opened by dengchubo - 2
请问我该怎样把basenet改成VGG16呢
#25 opened by Faterjames - 5
训练的时候感觉loss也很低,大概在2左右,但是做evaluate mAP 基本是0 。
#18 opened by sibadakesi - 2
epoch
#23 opened by liby542698 - 1
有人对这个代码增加FPN吗?
#24 opened by sibadakesi - 0
训练完了没有发现已经保存的 model
#22 opened by hzcforever - 0
运行 inference.py 后程序终止
#21 opened by hzcforever - 3
rcnn_miss_gt_num非常高? 有没有什么解决办法
#20 opened by sibadakesi - 1
关于map_fn
#19 opened by sibadakesi - 1
RPN网络的class输出没有经过softmax直接进行计算交叉熵损失了?
#17 opened by dishofchicken - 7
请问这个可以训练自己的数据集吗,该修改哪些地方吗
#2 opened by huangzhenjie - 4
- 3
evaluate error
#16 opened by NiFangBaAGe - 1
运行inference.py进程异常退出
#14 opened by 836304831 - 4
- 6
- 2
voc_path = '/opt/dataset/VOCdevkit'目錄找不到?
#11 opened by kaonick - 5
- 9
- 1
点赞支持
#9 opened by qingmeizhujiu - 6
How about the performance?
#1 opened by bityangke - 2
- 2
train.rcnn.log
#5 opened by wudi00 - 2
- 0
tensorflow.python.framework.errors_impl.NotFoundError: Resource __per_step_8/_tensor_arraysrpn2proposals/map/TensorArray_2_17/N10tensorflow11TensorArrayE does not exist.
#3 opened by yizt