search

pkmital / CADL

ARCHIVED: Contains historical course materials/Homework materials for the FREE MOOC course on "Creative Applications of Deep Learning w/ Tensorflow" #CADL
https://www.kadenze.com/courses/creative-applications-of-deep-learning-with-tensorflow/info
Apache License 2.0
1.48k stars 732 forks source link

Fixed kernel width in utils.deconv2d #57

Closed sunsided closed 7 years ago

sunsided commented 7 years ago

In the utils.deconv2d() function, the parameter k_w was ignored and the kernel height was used in its place:

W = tf.get_variable(
    name='W',
    shape=[k_h, k_h, n_output_ch, n_input_ch or x.get_shape()[-1]],
    initializer=tf.contrib.layers.xavier_initializer_conv2d())

This has been changed to:

W = tf.get_variable(
    name='W',
    shape=[k_h, k_w, n_output_ch, n_input_ch or x.get_shape()[-1]],
    initializer=tf.contrib.layers.xavier_initializer_conv2d())
pkmital commented 7 years ago

Oops good catch! TY!

sunsided commented 7 years ago

Very welcome. :slightly_smiling_face: