depth_estimation train with my images

[Samad08]

Hello,

I am working with your depth_estimation code. I downloaded all the required files and kept them as zip-file as mentioned. The training works well with the dataset given in the repository. If I try to replace the training images (nyu_data.zip) with my 20 images (I have them as RGB and Depth in a zip file as needed in the code), it doesn't work and gives a 'Key Error' for one of the images. Do you have an idea how can I solve this?

p.s: I'm a new to this

Regards,
Samad