Hi, I don't know how to do it. Can you tell me

[ytgcljj]

importError:No module named object_detection.utils

[yyll008]

https://github.com/tensorflow/models

Maybe you not install this.

[duxiaoyang06]

hi，where is the model config file?

[priya-dwivedi]

Please check this link

https://github.com/tensorflow/models/tree/master/research/object_detection/samples/configs

On Sun, Oct 15, 2017 at 1:29 PM, by.sun [email protected] wrote:

hi，where is the model config file?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/priya-dwivedi/Deep-Learning/issues/1#issuecomment-336693495, or mute the thread https://github.com/notifications/unsubscribe-auth/AUHGOAcstoZZAmXgcZdcOdZ1hw3Y0fuJks5ssbtxgaJpZM4OZt2w .

[duxiaoyang06]

Thanks . I know the default config files,but I am not sure what prefer params to config.And when I transfer any ckpt to pb file it always raise error 'FailedPreconditionError: Attempting to use uninitialized value FirstStageFeatureExtractor...' ,I don‘t know why?The training process can finish automatically or must be stopped manually?I must use the last ckpt file?

[priya-dwivedi]

You can start training with default params and adjust based on performance. As for the error, no idea. Please post on stackoverflow.

On Mon, Oct 16, 2017 at 12:07 PM, by.sun [email protected] wrote:

Thanks . I know the default config files,but I am not sure what prefer params to config.And when I transfer any ckpt to pb file it always raise error 'FailedPreconditionError: Attempting to use uninitialized value FirstStageFeatureExtractor...' ,I don‘t know why?The training process can finish automatically or must be stopped manually?I must use the last ckpt file?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/priya-dwivedi/Deep-Learning/issues/1#issuecomment-336793508, or mute the thread https://github.com/notifications/unsubscribe-auth/AUHGODsAiVUn_VWOfnJIETI_mvFKDrhSks5ssvm2gaJpZM4OZt2w .

[duxiaoyang06]

Thanks a lot～

[wenchengxucool]

download https://github.com/tensorflow/models . and copy to the directory of MarkRCNN?

[priya-dwivedi]

please ensure tensorflow version is 1.5. That’s the one I used.

Sent from my iPhone

On Apr 21, 2018, at 4:24 AM, Wei Chen [email protected] wrote:

download https://github.com/tensorflow/models . and copy to the directory of MarkRCNN?

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

[wenchengxucool]

sorry, I didn't notice the comment "# This is needed since the notebook is stored in the object_detection folder."....