Is your maxpooling layer wrong?

orobix / retina-unet

Retina blood vessel segmentation with a convolutional neural network

1.26k stars 468 forks source link

Is your maxpooling layer wrong? #43

Open eliottbrion opened 6 years ago

eliottbrion commented 6 years ago

Hello,

I checked both shapes of conv1 and pool1 and they don't seem to be coherent: K.int_shape(conv1) (None, 32, 48, 48) K.int_shape(pool1) (None, 16, 24, 48) It looks like you're doing the maxpooling operation on the feature map.

In retina-unet/src/retinaNN_training.py, don't you think that you should replace the 37th line pool1 = MaxPooling2D((2, 2))(conv1) by pool1 = MaxPooling2D((2, 2),data_format='channels_first')(conv1)?

Eliott

itachi1232gg commented 6 years ago

@eliottbrion I think you are right, I found this error as well. What interesting is, it does not affect the results a lot.

liuzhibo-delta commented 4 years ago

I agree with you , the UpSampling Layer as well as the Maxpooling Layer are Wrong . Because of the parameter data_format='channels_first' is missing . You can correct it .