Question about image data format

DeepTrial / Retina-VesselNet

A Simple U-net model for Retinal Blood Vessel Segmentation based on tensorflow2

303 stars 76 forks source link

Question about image data format #13

Closed ZiqianXie closed 5 years ago

ZiqianXie commented 5 years ago

I was looking at your dense_unet model, the input shape is (patch_width, patch_height, 1), so it is channel-last ordering, but then the batch normalization is done on axis=1, which corresponds to channel-first ordering. Is this a bug or did I miss something?

DeepTrial commented 5 years ago

i have to admit that it's bug.I pushed the wrong model which written in theano.thx

ZiqianXie commented 5 years ago

Thanks for your good work, can you also update the pretrained model?

DeepTrial commented 5 years ago

oh,of course! I will update as soon as possible

DeepTrial commented 5 years ago

I already update the whole project(including new download-link of pretrained model ),please check

ZiqianXie commented 5 years ago

I really appreciate it, thank you!