Created
January 6, 2017 05:28
-
-
Save SamuelMarks/17e968288545042da4e718e886e458e3 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
diff --git a/tutorials/image/cifar10/cifar10.py b/tutorials/image/cifar10/cifar10.py | |
index d99ffb9..4edcf68 100644 | |
--- a/tutorials/image/cifar10/cifar10.py | |
+++ b/tutorials/image/cifar10/cifar10.py | |
@@ -90,8 +90,8 @@ def _activation_summary(x): | |
# Remove 'tower_[0-9]/' from the name in case this is a multi-GPU training | |
# session. This helps the clarity of presentation on tensorboard. | |
tensor_name = re.sub('%s_[0-9]*/' % TOWER_NAME, '', x.op.name) | |
- tf.contrib.deprecated.histogram_summary(tensor_name + '/activations', x) | |
- tf.contrib.deprecated.scalar_summary(tensor_name + '/sparsity', | |
+ tf.summary.histogram_summary(tensor_name + '/activations', x) | |
+ tf.summary.scalar_summary(tensor_name + '/sparsity', | |
tf.nn.zero_fraction(x)) | |
@@ -316,8 +316,8 @@ def _add_loss_summaries(total_loss): | |
for l in losses + [total_loss]: | |
# Name each loss as '(raw)' and name the moving average version of the loss | |
# as the original loss name. | |
- tf.contrib.deprecated.scalar_summary(l.op.name + ' (raw)', l) | |
- tf.contrib.deprecated.scalar_summary(l.op.name, loss_averages.average(l)) | |
+ tf.summary.scalar_summary(l.op.name + ' (raw)', l) | |
+ tf.summary.scalar_summary(l.op.name, loss_averages.average(l)) | |
return loss_averages_op | |
@@ -345,7 +345,7 @@ def train(total_loss, global_step): | |
decay_steps, | |
LEARNING_RATE_DECAY_FACTOR, | |
staircase=True) | |
- tf.contrib.deprecated.scalar_summary('learning_rate', lr) | |
+ tf.summary.scalar_summary('learning_rate', lr) | |
# Generate moving averages of all losses and associated summaries. | |
loss_averages_op = _add_loss_summaries(total_loss) | |
@@ -360,12 +360,12 @@ def train(total_loss, global_step): | |
# Add histograms for trainable variables. | |
for var in tf.trainable_variables(): | |
- tf.contrib.deprecated.histogram_summary(var.op.name, var) | |
+ tf.summary.histogram_summary(var.op.name, var) | |
# Add histograms for gradients. | |
for grad, var in grads: | |
if grad is not None: | |
- tf.contrib.deprecated.histogram_summary(var.op.name + '/gradients', grad) | |
+ tf.summary.histogram_summary(var.op.name + '/gradients', grad) | |
# Track the moving averages of all trainable variables. | |
variable_averages = tf.train.ExponentialMovingAverage( | |
diff --git a/tutorials/image/cifar10/cifar10_input.py b/tutorials/image/cifar10/cifar10_input.py | |
index 7bfcb2e..38861d1 100644 | |
--- a/tutorials/image/cifar10/cifar10_input.py | |
+++ b/tutorials/image/cifar10/cifar10_input.py | |
@@ -84,13 +84,13 @@ def read_cifar10(filename_queue): | |
# The first bytes represent the label, which we convert from uint8->int32. | |
result.label = tf.cast( | |
- tf.strided_slice(record_bytes, [0], [label_bytes]), tf.int32) | |
+ tf.strided_slice(record_bytes, [0], [label_bytes], [1,1]), tf.int32) | |
# The remaining bytes after the label represent the image, which we reshape | |
# from [depth * height * width] to [depth, height, width]. | |
depth_major = tf.reshape( | |
tf.strided_slice(record_bytes, [label_bytes], | |
- [label_bytes + image_bytes]), | |
+ [label_bytes + image_bytes], [1,1]), | |
[result.depth, result.height, result.width]) | |
# Convert from [depth, height, width] to [height, width, depth]. | |
result.uint8image = tf.transpose(depth_major, [1, 2, 0]) | |
@@ -132,7 +132,7 @@ def _generate_image_and_label_batch(image, label, min_queue_examples, | |
capacity=min_queue_examples + 3 * batch_size) | |
# Display the training images in the visualizer. | |
- tf.contrib.deprecated.image_summary('images', images) | |
+ tf.summary.image_summary('images', images) | |
return images, tf.reshape(label_batch, [batch_size]) | |
diff --git a/tutorials/image/cifar10/cifar10_multi_gpu_train.py b/tutorials/image/cifar10/cifar10_multi_gpu_train.py | |
index ed0ac6f..c83580e 100644 | |
--- a/tutorials/image/cifar10/cifar10_multi_gpu_train.py | |
+++ b/tutorials/image/cifar10/cifar10_multi_gpu_train.py | |
@@ -93,7 +93,7 @@ def tower_loss(scope): | |
# Remove 'tower_[0-9]/' from the name in case this is a multi-GPU training | |
# session. This helps the clarity of presentation on tensorboard. | |
loss_name = re.sub('%s_[0-9]*/' % cifar10.TOWER_NAME, '', l.op.name) | |
- tf.contrib.deprecated.scalar_summary(loss_name, l) | |
+ tf.summary.scalar_summary(loss_name, l) | |
return total_loss | |
@@ -187,13 +187,13 @@ def train(): | |
grads = average_gradients(tower_grads) | |
# Add a summary to track the learning rate. | |
- summaries.append(tf.contrib.deprecated.scalar_summary('learning_rate', lr)) | |
+ summaries.append(tf.summary.scalar_summary('learning_rate', lr)) | |
# Add histograms for gradients. | |
for grad, var in grads: | |
if grad is not None: | |
summaries.append( | |
- tf.contrib.deprecated.histogram_summary(var.op.name + '/gradients', | |
+ tf.summary.histogram_summary(var.op.name + '/gradients', | |
grad)) | |
# Apply the gradients to adjust the shared variables. | |
@@ -202,7 +202,7 @@ def train(): | |
# Add histograms for trainable variables. | |
for var in tf.trainable_variables(): | |
summaries.append( | |
- tf.contrib.deprecated.histogram_summary(var.op.name, var)) | |
+ tf.summary.histogram_summary(var.op.name, var)) | |
# Track the moving averages of all trainable variables. | |
variable_averages = tf.train.ExponentialMovingAverage( | |
@@ -216,7 +216,7 @@ def train(): | |
saver = tf.train.Saver(tf.global_variables()) | |
# Build the summary operation from the last tower summaries. | |
- summary_op = tf.contrib.deprecated.merge_summary(summaries) | |
+ summary_op = tf.summary.merge_summary(summaries) | |
# Build an initialization operation to run below. | |
init = tf.global_variables_initializer() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment