SungHwan Kim commited to 258d1f0 January 15

implemented guided gradcam (#38)

Current Status:

Analysed View logs
View on GitHub
Not up to standards.

This commit quality could be better.

  • +8 Issues
  • = Duplication
  • - Complexity
  • - Coverage
download-models.sh
Tips depend on target shell and yours is unknown. Add a shebang.
test_model_dir="./test/data"
test/test_gradcam.py
Unused variable 'net'
net, end_points = resnet_v1.resnet_v1_50(self.inputs, self.nbclasses, is_training=False)
Unused variable 'end_points'
net, end_points = vgg.vgg_16(self.inputs, self.nbclasses, is_training=False)
test/test_gradcam_dangling.py
Unused variable 'net'
net, end_points = resnet_v1.resnet_v1_50(inputs, self.nbclasses, is_training=False)
Trailing whitespace
test/test_gradcam_guided_backprop.py
Unused variable 'y'
y = tf.placeholder(tf.int32, name='y_placeholder', shape=[1, 2])
Unused variable 'logits'
logits = tf.nn.softmax(top)
test/test_gradcam_util.py
Unused variable 'y'
x, y, cross_entropy = nn_graph(activation=tf.nn.relu)
More
You didn't fix any issue!

Loading...

Loading...

Loading...

Loading...