sudhasubramaniam

This might be because u-net architecture consists of contracting path(conv, max pooling) and expansion path (upsampling, concatenation), when we down sample image of shape [(None,25,25,512)] we will get [None,12,12,512], on...

Hey did you got any solution for this, I'm also facing same issue.

Thanks a lot. Still I'm facing same issue. Code s available in the link below https://github.com/sudhasubramaniam/segmentation

No change. Still facing same issue.

Thankyou, It working now. If you see my github repository my training dataset has training image and label. Please let me know whether I can extract boundary information as in...

Thank you. Let me try as such.

Hi Sai, I trained with boundary information but still my perdiction is not good. (epoch-300). Please comment on causes https://github.com/sudhasubramaniam/segmentation I used your code for training.

Thank you for your comments.

model.py import tensorflow as tf class MNIST_CNN: def __init__(self, name='MNIST_CNN'): self.name = name def convlayer(self, input, shape, name): w_conv = tf.Variable(tf.truncated_normal(shape=shape, dtype=tf.float32, stddev=0.1), name='w_{0}'.format(name)) b_conv = tf.Variable(tf.constant(0.0, shape=shape[-1:], dtype=tf.float32), name='b_{0}'.format(name))...

train.py program from utils import DataGenerator, MNISTLoader from model import MNIST_CNN import tensorflow as tf logdir = './logs/' chkpt = './logs/model.ckpt' n_epochs = 20 batch_size = 20 class Trainer: def...