when i use jetson-interface DIGIDTS train on PC

When I chose the dataset, the web page told me that the folder did not exist, but clearly existed, I did follow the instructions.

https://github.com/dusty-nv/jetson-inference/blob/master/docs/imagenet-training.md#customizing-the-object-classes

i download pic datasets,and use the imagenet-subset.sh make some file in 12_classes

Then ,i fill the location of 12_classes to digits web page,the web tell me “Floder does not exist”

i dont know, why?
please help me 。thanks

Hi,

It looks like you meet the similar issue of this one:
https://github.com/NVIDIA/DIGITS/issues/429

Could you check if the reply inside helps?
Thanks.

yes,same problem.
I want to ask you how to solve the problem mentioned in the following sentence. Now the problem is known, but I don’t know how to solve it. Can you help me? Thank you

Permissions
The DIGITS server runs as www-data, so keep in mind that prebuilt LMDB datasets used for generic models need to be readable by the www-data user. In particular, the entire chain of directories from / to your data must be readable by www-data.

What is your docker run command?

You might want to see this post helps for running the imagenet-subset.sh script inside of your Docker environment:
https://github.com/dusty-nv/jetson-inference/issues/202#issuecomment-435798621