if v.name.split(':')[0] in var_keep_dic: TypeError: argument of type 'NoneType' is not iterable

[hongyanghu]

Traceback (most recent call last): File "C:/Users/huhongyang/Desktop/Faster-RCNN-TensorFlow-Python3-master/train.py", line 216, in train.train() File "C:/Users/huhongyang/Desktop/Faster-RCNN-TensorFlow-Python3-master/train.py", line 123, in train variables_to_restore = self.net.get_variables_to_restore(variables, var_keep_dic) File "C:\Users\huhongyang\Desktop\Faster-RCNN-TensorFlow-Python3-master\lib\nets\vgg16.py", line 66, in get_variables_to_restore if v.name.split(':')[0] in var_keep_dic: TypeError: argument of type 'NoneType' is not iterable

[zhangyue678]

how to do?

[hongyanghu]

你把预训练的vgg_16.ckpt改为vgg16.ckpt。加波好友讨论一下吧------------------ 原始邮件 ------------------ 发件人: "zhangyue"[email protected] 发送时间: 2019年10月27日(星期天) 下午3:42 收件人: "dBeker/Faster-RCNN-TensorFlow-Python3"[email protected]; 抄送: "lincuncuncaogoudan"[email protected];"Author"[email protected]; 主题: Re: [dBeker/Faster-RCNN-TensorFlow-Python3] if v.name.split(':')[0]in var_keep_dic: TypeError: argument of type 'NoneType' is not iterable (#116)

how to do?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[chanstag]

For me this issue appeared because it was looking for vgg16.ckpt in /data/imagenet_weights but the file was saved as vgg_16.ckpt so I just copied it over to vgg16.ckpt.