the error messages of running demo_toy_problem.ipynb about tf_unet HOT 5 CLOSED

Weird! This seems to be caused somewhere deep in Tensorflow. What version are you using?

The code works fine on my MacBook Pro, Tensorflow '0.10.0rc0' and Python '3.4.4'

from tf_unet.

Hi Jakeret, this is the version information I am using:

python -c 'import tensorflow as tf; print(tf.version)'
0.8.0

import sys
print(sys.version)
3.4.4 (default, May 19 2016, 07:45:08)
[GCC 5.3.0]

from tf_unet.

Hi Jacket, it looks to me that this line of code gradients = sess.run((self.optimizer, self.net.cost, self.learning_rate_node, self.net.gradients_node), feed_dict={self.net.x: batch_x,self.net.y: util.crop_to_shape(batch_y,pred_shape), self.net.keep_prob: dropout}) does not work for the version before 0.10. I changed it to _,loss,lr,gradients = sess.run([self.optimizer, self.net.cost, self.learning_rate_node]+self.net.gradients_node,feed_dict={self.net.x: batch_x,self.net.y: util.crop_to_shape(batch_y, pred_shape),self.net.keep_prob: dropout}) , but it right now gives the error message such as ValueError: too many values to unpack (expected 4)

Just curious, are there any approaches to solve this problem without upgrading to the version you are using. Thank you very much.

from tf_unet.

Sorry I'm traveling at the moment. However, I don't have any experience
with Tensorflow < 0.9.0. So I don't know if it's runnable with your version
On Oct 14, 2016 4:32 AM, "wenouyang" [email protected] wrote:

Hi Jacket, it looks to me that this line of code gradients =
sess.run((self.optimizer, self.net.cost, self.learning_rate_node,
self.net.gradients_node), feed_dict={self.net.x: batch_x,self.net.y:
util.crop_to_shape(batch_y,pred_shape), self.net.keep_prob: dropout})
does not work for the version before 0.10. I changed it to _,loss,lr,gradients
= sess.run([self.optimizer, self.net.cost, self.learning_rate_node]+self.
net.gradients_node,feed_dict={self.net.x: batch_x,self.net.y:
util.crop_to_shape(batch_y, pred_shape),self.net.keep_prob: dropout})
, but it right now gives the error message such as ValueError: too many
values to unpack (expected 4)

Just curious, are there any approaches to solve this problem without
upgrading to the version you are using. Thank you very much.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#1 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/ALSFvwwnKkg2Vq_Irj6ZTR9fNTQ0T7Xwks5qzpVagaJpZM4KKjR_
.

from tf_unet.

Thank you, with some help, I got this figured out.

from tf_unet.