Recommended way to change input and output size

jakeret / tf_unet

Generic U-Net Tensorflow implementation for image segmentation

GNU General Public License v3.0

1.9k stars 748 forks source link

Closed I-CANT-CODE closed 5 years ago

I-CANT-CODE commented 5 years ago

Hello,

I am attempting to train on 128*128 input and output images. What is the recommended way to modify the architecture to support this type of data?

Edit: forgot to specify, both the input and output are Black and white images.

jakeret commented 5 years ago

In principle you don't have to change anything in the network architecture. The implementation if capable of dealing with any sized image data

I-CANT-CODE commented 5 years ago

Oh awesome! Thanks.