Skip to content

GitLab

Commit 4dcd5116 authored by Jing Li's avatar Jing Li
Browse files

Removed deprecated research/gan. Please visit https://github.com/tensorflow/gan.

parent f673f7a8
Hide whitespace changes
Inline Side-by-side
Showing with 0 additions and 1901 deletions
research/gan/image_compression/train_test.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for image_compression.train."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl import flags
from absl.testing import parameterized
import numpy as np
import tensorflow as tf
import train
FLAGS = flags.FLAGS
mock = tf.test.mock
class TrainTest(tf.test.TestCase, parameterized.TestCase):
@parameterized.named_parameters(
('NoAdversarialLoss', 0.0),
('AdversarialLoss', 1.0))
def test_build_graph(self, weight_factor):
FLAGS.max_number_of_steps = 0
FLAGS.weight_factor = weight_factor
batch_size = 3
patch_size = 16
FLAGS.batch_size = batch_size
FLAGS.patch_size = patch_size
mock_imgs = np.zeros([batch_size, patch_size, patch_size, 3],
dtype=np.float32)
with mock.patch.object(train, 'data_provider') as mock_data_provider:
mock_data_provider.provide_data.return_value = mock_imgs
train.main(None)
if __name__ == '__main__':
tf.test.main()
research/gan/mnist/conditional_eval.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Evaluates a conditional TFGAN trained MNIST model."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl import app
from absl import flags
import tensorflow as tf
import data_provider
import networks
import util
tfgan = tf.contrib.gan
flags.DEFINE_string('checkpoint_dir', '/tmp/mnist/',
'Directory where the model was written to.')
flags.DEFINE_string('eval_dir', '/tmp/mnist/',
'Directory where the results are saved to.')
flags.DEFINE_integer('num_images_per_class', 10,
'Number of images to generate per class.')
flags.DEFINE_integer('noise_dims', 64,
'Dimensions of the generator noise vector')
flags.DEFINE_string('classifier_filename', None,
'Location of the pretrained classifier. If `None`, use '
'default.')
flags.DEFINE_integer('max_number_of_evaluations', None,
'Number of times to run evaluation. If `None`, run '
'forever.')
flags.DEFINE_boolean('write_to_disk', True, 'If `True`, run images to disk.')
FLAGS = flags.FLAGS
NUM_CLASSES = 10
def main(_, run_eval_loop=True):
with tf.name_scope('inputs'):
noise, one_hot_labels = _get_generator_inputs(
FLAGS.num_images_per_class, NUM_CLASSES, FLAGS.noise_dims)
# Generate images.
with tf.variable_scope('Generator'): # Same scope as in train job.
images = networks.conditional_generator(
(noise, one_hot_labels), is_training=False)
# Visualize images.
reshaped_img = tfgan.eval.image_reshaper(
images, num_cols=FLAGS.num_images_per_class)
tf.summary.image('generated_images', reshaped_img, max_outputs=1)
# Calculate evaluation metrics.
tf.summary.scalar('MNIST_Classifier_score',
util.mnist_score(images, FLAGS.classifier_filename))
tf.summary.scalar('MNIST_Cross_entropy',
util.mnist_cross_entropy(
images, one_hot_labels, FLAGS.classifier_filename))
# Write images to disk.
image_write_ops = None
if FLAGS.write_to_disk:
image_write_ops = tf.write_file(
'%s/%s'% (FLAGS.eval_dir, 'conditional_gan.png'),
tf.image.encode_png(data_provider.float_image_to_uint8(
reshaped_img[0])))
# For unit testing, use `run_eval_loop=False`.
if not run_eval_loop: return
tf.contrib.training.evaluate_repeatedly(
FLAGS.checkpoint_dir,
hooks=[tf.contrib.training.SummaryAtEndHook(FLAGS.eval_dir),
tf.contrib.training.StopAfterNEvalsHook(1)],
eval_ops=image_write_ops,
max_number_of_evaluations=FLAGS.max_number_of_evaluations)
def _get_generator_inputs(num_images_per_class, num_classes, noise_dims):
# Since we want a grid of numbers for the conditional generator, manually
# construct the desired class labels.
num_images_generated = num_images_per_class * num_classes
noise = tf.random_normal([num_images_generated, noise_dims])
labels = [lbl for lbl in range(num_classes) for _
in range(num_images_per_class)]
one_hot_labels = tf.one_hot(tf.constant(labels), num_classes)
return noise, one_hot_labels
if __name__ == '__main__':
app.run(main)
research/gan/mnist/conditional_eval_test.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for tfgan.examples.mnist.conditional_eval."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl.testing import absltest
import conditional_eval
class ConditionalEvalTest(absltest.TestCase):
def test_build_graph(self):
conditional_eval.main(None, run_eval_loop=False)
if __name__ == '__main__':
absltest.main()
research/gan/mnist/data_provider.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Contains code for loading and preprocessing the MNIST data."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import tensorflow as tf
from slim.datasets import dataset_factory as datasets
slim = tf.contrib.slim
def provide_data(split_name, batch_size, dataset_dir, num_readers=1,
num_threads=1):
"""Provides batches of MNIST digits.
Args:
split_name: Either 'train' or 'test'.
batch_size: The number of images in each batch.
dataset_dir: The directory where the MNIST data can be found.
num_readers: Number of dataset readers.
num_threads: Number of prefetching threads.
Returns:
images: A `Tensor` of size [batch_size, 28, 28, 1]
one_hot_labels: A `Tensor` of size [batch_size, mnist.NUM_CLASSES], where
each row has a single element set to one and the rest set to zeros.
num_samples: The number of total samples in the dataset.
Raises:
ValueError: If `split_name` is not either 'train' or 'test'.
"""
dataset = datasets.get_dataset('mnist', split_name, dataset_dir=dataset_dir)
provider = slim.dataset_data_provider.DatasetDataProvider(
dataset,
num_readers=num_readers,
common_queue_capacity=2 * batch_size,
common_queue_min=batch_size,
shuffle=(split_name == 'train'))
[image, label] = provider.get(['image', 'label'])
# Preprocess the images.
image = (tf.to_float(image) - 128.0) / 128.0
# Creates a QueueRunner for the pre-fetching operation.
images, labels = tf.train.batch(
[image, label],
batch_size=batch_size,
num_threads=num_threads,
capacity=5 * batch_size)
one_hot_labels = tf.one_hot(labels, dataset.num_classes)
return images, one_hot_labels, dataset.num_samples
def float_image_to_uint8(image):
"""Convert float image in [-1, 1) to [0, 255] uint8.
Note that `1` gets mapped to `0`, but `1 - epsilon` gets mapped to 255.
Args:
image: An image tensor. Values should be in [-1, 1).
Returns:
Input image cast to uint8 and with integer values in [0, 255].
"""
image = (image * 128.0) + 128.0
return tf.cast(image, tf.uint8)
research/gan/mnist/data_provider_test.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for data_provider."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
from absl import flags
import tensorflow as tf
import data_provider
class DataProviderTest(tf.test.TestCase):
def test_mnist_data_reading(self):
dataset_dir = os.path.join(
flags.FLAGS.test_srcdir,
'google3/third_party/tensorflow_models/gan/mnist/testdata')
batch_size = 5
images, labels, num_samples = data_provider.provide_data(
'test', batch_size, dataset_dir)
self.assertEqual(num_samples, 10000)
with self.test_session() as sess:
with tf.contrib.slim.queues.QueueRunners(sess):
images, labels = sess.run([images, labels])
self.assertEqual(images.shape, (batch_size, 28, 28, 1))
self.assertEqual(labels.shape, (batch_size, 10))
if __name__ == '__main__':
tf.test.main()
research/gan/mnist/eval.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Evaluates a TFGAN trained MNIST model."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl import app
from absl import flags
import tensorflow as tf
import data_provider
import networks
import util
FLAGS = flags.FLAGS
tfgan = tf.contrib.gan
flags.DEFINE_string('checkpoint_dir', '/tmp/mnist/',
'Directory where the model was written to.')
flags.DEFINE_string('eval_dir', '/tmp/mnist/',
'Directory where the results are saved to.')
flags.DEFINE_string('dataset_dir', None, 'Location of data.')
flags.DEFINE_integer('num_images_generated', 1000,
'Number of images to generate at once.')
flags.DEFINE_boolean('eval_real_images', False,
'If `True`, run Inception network on real images.')
flags.DEFINE_integer('noise_dims', 64,
'Dimensions of the generator noise vector')
flags.DEFINE_string('classifier_filename', None,
'Location of the pretrained classifier. If `None`, use '
'default.')
flags.DEFINE_integer('max_number_of_evaluations', None,
'Number of times to run evaluation. If `None`, run '
'forever.')
flags.DEFINE_boolean('write_to_disk', True, 'If `True`, run images to disk.')
def main(_, run_eval_loop=True):
# Fetch real images.
with tf.name_scope('inputs'):
real_images, _, _ = data_provider.provide_data(
'train', FLAGS.num_images_generated, FLAGS.dataset_dir)
image_write_ops = None
if FLAGS.eval_real_images:
tf.summary.scalar('MNIST_Classifier_score',
util.mnist_score(real_images, FLAGS.classifier_filename))
else:
# In order for variables to load, use the same variable scope as in the
# train job.
with tf.variable_scope('Generator'):
images = networks.unconditional_generator(
tf.random_normal([FLAGS.num_images_generated, FLAGS.noise_dims]),
is_training=False)
tf.summary.scalar('MNIST_Frechet_distance',
util.mnist_frechet_distance(
real_images, images, FLAGS.classifier_filename))
tf.summary.scalar('MNIST_Classifier_score',
util.mnist_score(images, FLAGS.classifier_filename))
if FLAGS.num_images_generated >= 100 and FLAGS.write_to_disk:
reshaped_images = tfgan.eval.image_reshaper(
images[:100, ...], num_cols=10)
uint8_images = data_provider.float_image_to_uint8(reshaped_images)
image_write_ops = tf.write_file(
'%s/%s'% (FLAGS.eval_dir, 'unconditional_gan.png'),
tf.image.encode_png(uint8_images[0]))
# For unit testing, use `run_eval_loop=False`.
if not run_eval_loop: return
tf.contrib.training.evaluate_repeatedly(
FLAGS.checkpoint_dir,
hooks=[tf.contrib.training.SummaryAtEndHook(FLAGS.eval_dir),
tf.contrib.training.StopAfterNEvalsHook(1)],
eval_ops=image_write_ops,
max_number_of_evaluations=FLAGS.max_number_of_evaluations)
if __name__ == '__main__':
app.run(main)
research/gan/mnist/eval_test.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for tfgan.examples.mnist.eval."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl import flags
from absl.testing import absltest
from absl.testing import parameterized
import eval # pylint:disable=redefined-builtin
class EvalTest(parameterized.TestCase):
@parameterized.named_parameters(
('RealData', True),
('GeneratedData', False))
def test_build_graph(self, eval_real_images):
flags.FLAGS.eval_real_images = eval_real_images
eval.main(None, run_eval_loop=False)
if __name__ == '__main__':
absltest.main()
research/gan/mnist/infogan_eval.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Evaluates an InfoGAN TFGAN trained MNIST model.
The image visualizations, as in https://arxiv.org/abs/1606.03657, show the
effect of varying a specific latent variable on the image. Each visualization
focuses on one of the three structured variables. Columns have two of the three
variables fixed, while the third one is varied. Different rows have different
random samples from the remaining latents.
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl import app
from absl import flags
import numpy as np
import tensorflow as tf
import data_provider
import networks
import util
tfgan = tf.contrib.gan
flags.DEFINE_string('checkpoint_dir', '/tmp/mnist/',
'Directory where the model was written to.')
flags.DEFINE_string('eval_dir', '/tmp/mnist/',
'Directory where the results are saved to.')
flags.DEFINE_integer('noise_samples', 6,
'Number of samples to draw from the continuous structured '
'noise.')
flags.DEFINE_integer('unstructured_noise_dims', 62,
'The number of dimensions of the unstructured noise.')
flags.DEFINE_integer('continuous_noise_dims', 2,
'The number of dimensions of the continuous noise.')
flags.DEFINE_string('classifier_filename', None,
'Location of the pretrained classifier. If `None`, use '
'default.')
flags.DEFINE_integer('max_number_of_evaluations', None,
'Number of times to run evaluation. If `None`, run '
'forever.')
flags.DEFINE_boolean('write_to_disk', True, 'If `True`, run images to disk.')
CAT_SAMPLE_POINTS = np.arange(0, 10)
CONT_SAMPLE_POINTS = np.linspace(-2.0, 2.0, 10)
FLAGS = flags.FLAGS
def main(_, run_eval_loop=True):
with tf.name_scope('inputs'):
noise_args = (FLAGS.noise_samples, CAT_SAMPLE_POINTS, CONT_SAMPLE_POINTS,
FLAGS.unstructured_noise_dims, FLAGS.continuous_noise_dims)
# Use fixed noise vectors to illustrate the effect of each dimension.
display_noise1 = util.get_eval_noise_categorical(*noise_args)
display_noise2 = util.get_eval_noise_continuous_dim1(*noise_args)
display_noise3 = util.get_eval_noise_continuous_dim2(*noise_args)
_validate_noises([display_noise1, display_noise2, display_noise3])
# Visualize the effect of each structured noise dimension on the generated
# image.
def generator_fn(inputs):
return networks.infogan_generator(
inputs, len(CAT_SAMPLE_POINTS), is_training=False)
with tf.variable_scope('Generator') as genscope: # Same scope as in training.
categorical_images = generator_fn(display_noise1)
reshaped_categorical_img = tfgan.eval.image_reshaper(
categorical_images, num_cols=len(CAT_SAMPLE_POINTS))
tf.summary.image('categorical', reshaped_categorical_img, max_outputs=1)
with tf.variable_scope(genscope, reuse=True):
continuous1_images = generator_fn(display_noise2)
reshaped_continuous1_img = tfgan.eval.image_reshaper(
continuous1_images, num_cols=len(CONT_SAMPLE_POINTS))
tf.summary.image('continuous1', reshaped_continuous1_img, max_outputs=1)
with tf.variable_scope(genscope, reuse=True):
continuous2_images = generator_fn(display_noise3)
reshaped_continuous2_img = tfgan.eval.image_reshaper(
continuous2_images, num_cols=len(CONT_SAMPLE_POINTS))
tf.summary.image('continuous2', reshaped_continuous2_img, max_outputs=1)
# Evaluate image quality.
all_images = tf.concat(
[categorical_images, continuous1_images, continuous2_images], 0)
tf.summary.scalar('MNIST_Classifier_score',
util.mnist_score(all_images, FLAGS.classifier_filename))
# Write images to disk.
image_write_ops = []
if FLAGS.write_to_disk:
image_write_ops.append(_get_write_image_ops(
FLAGS.eval_dir, 'categorical_infogan.png', reshaped_categorical_img[0]))
image_write_ops.append(_get_write_image_ops(
FLAGS.eval_dir, 'continuous1_infogan.png', reshaped_continuous1_img[0]))
image_write_ops.append(_get_write_image_ops(
FLAGS.eval_dir, 'continuous2_infogan.png', reshaped_continuous2_img[0]))
# For unit testing, use `run_eval_loop=False`.
if not run_eval_loop: return
tf.contrib.training.evaluate_repeatedly(
FLAGS.checkpoint_dir,
hooks=[tf.contrib.training.SummaryAtEndHook(FLAGS.eval_dir),
tf.contrib.training.StopAfterNEvalsHook(1)],
eval_ops=image_write_ops,
max_number_of_evaluations=FLAGS.max_number_of_evaluations)
def _validate_noises(noises):
"""Sanity check on constructed noise tensors.
Args:
noises: List of 3-tuples of noise vectors.
"""
assert isinstance(noises, (list, tuple))
for noise_l in noises:
assert len(noise_l) == 3
assert isinstance(noise_l[0], np.ndarray)
batch_dim = noise_l[0].shape[0]
for i, noise in enumerate(noise_l):
assert isinstance(noise, np.ndarray)
# Check that batch dimensions are all the same.
assert noise.shape[0] == batch_dim
# Check that shapes for corresponding noises are the same.
assert noise.shape == noises[0][i].shape
def _get_write_image_ops(eval_dir, filename, images):
"""Create Ops that write images to disk."""
return tf.write_file(
'%s/%s'% (eval_dir, filename),
tf.image.encode_png(data_provider.float_image_to_uint8(images)))
if __name__ == '__main__':
app.run(main)
research/gan/mnist/infogan_eval_test.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for tfgan.examples.mnist.infogan_eval."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl.testing import absltest
import infogan_eval
class MnistInfoGANEvalTest(absltest.TestCase):
def test_build_graph(self):
infogan_eval.main(None, run_eval_loop=False)
if __name__ == '__main__':
absltest.main()
research/gan/mnist/launch_jobs.sh deleted 100755 → 0
View file @ f673f7a8
# Copyright 2016 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
#!/bin/bash
#
# This script performs the following operations:
# 1. Downloads the MNIST dataset.
# 2. Trains an unconditional, conditional, or InfoGAN model on the MNIST
# training set.
# 3. Evaluates the models and writes sample images to disk.
#
# These examples are intended to be fast. For better final results, tune
# hyperparameters or train longer.
#
# NOTE: Each training step takes about 0.5 second with a batch size of 32 on
# CPU. On GPU, it takes ~5 milliseconds.
#
# With the default batch size and number of steps, train times are:
#
# unconditional: CPU: 800 steps, ~10 min GPU: 800 steps, ~1 min
# conditional: CPU: 2000 steps, ~20 min GPU: 2000 steps, ~2 min
# infogan: CPU: 3000 steps, ~20 min GPU: 3000 steps, ~6 min
#
# Usage:
# cd models/research/gan/mnist
# ./launch_jobs.sh ${gan_type} ${git_repo}
set -e
# Type of GAN to run. Right now, options are `unconditional`, `conditional`, or
# `infogan`.
gan_type=$1
if ! [[ "$gan_type" =~ ^(unconditional|conditional|infogan) ]]; then
echo "'gan_type' must be one of: 'unconditional', 'conditional', 'infogan'."
exit
fi
# Location of the git repository.
git_repo=$2
if [[ "$git_repo" == "" ]]; then
echo "'git_repo' must not be empty."
exit
fi
# Base name for where the checkpoint and logs will be saved to.
TRAIN_DIR=/tmp/mnist-model
# Base name for where the evaluation images will be saved to.
EVAL_DIR=/tmp/mnist-model/eval
# Where the dataset is saved to.
DATASET_DIR=/tmp/mnist-data
# Location of the classifier frozen graph used for evaluation.
FROZEN_GRAPH="${git_repo}/research/gan/mnist/data/classify_mnist_graph_def.pb"
export PYTHONPATH=$PYTHONPATH:$git_repo:$git_repo/research:$git_repo/research/slim
# A helper function for printing pretty output.
Banner () {
local text=$1
local green='\033[0;32m'
local nc='\033[0m' # No color.
echo -e "${green}${text}${nc}"
}
# Download the dataset.
python "${git_repo}/research/slim/download_and_convert_data.py" \
--dataset_name=mnist \
--dataset_dir=${DATASET_DIR}
# Run unconditional GAN.
if [[ "$gan_type" == "unconditional" ]]; then
UNCONDITIONAL_TRAIN_DIR="${TRAIN_DIR}/unconditional"
UNCONDITIONAL_EVAL_DIR="${EVAL_DIR}/unconditional"
NUM_STEPS=3000
# Run training.
Banner "Starting training unconditional GAN for ${NUM_STEPS} steps..."
python "${git_repo}/research/gan/mnist/train.py" \
--train_log_dir=${UNCONDITIONAL_TRAIN_DIR} \
--dataset_dir=${DATASET_DIR} \
--max_number_of_steps=${NUM_STEPS} \
--gan_type="unconditional" \
--alsologtostderr
Banner "Finished training unconditional GAN ${NUM_STEPS} steps."
# Run evaluation.
Banner "Starting evaluation of unconditional GAN..."
python "${git_repo}/research/gan/mnist/eval.py" \
--checkpoint_dir=${UNCONDITIONAL_TRAIN_DIR} \
--eval_dir=${UNCONDITIONAL_EVAL_DIR} \
--dataset_dir=${DATASET_DIR} \
--eval_real_images=false \
--classifier_filename=${FROZEN_GRAPH} \
--max_number_of_evaluations=1
Banner "Finished unconditional evaluation. See ${UNCONDITIONAL_EVAL_DIR} for output images."
fi
# Run conditional GAN.
if [[ "$gan_type" == "conditional" ]]; then
CONDITIONAL_TRAIN_DIR="${TRAIN_DIR}/conditional"
CONDITIONAL_EVAL_DIR="${EVAL_DIR}/conditional"
NUM_STEPS=3000
# Run training.
Banner "Starting training conditional GAN for ${NUM_STEPS} steps..."
python "${git_repo}/research/gan/mnist/train.py" \
--train_log_dir=${CONDITIONAL_TRAIN_DIR} \
--dataset_dir=${DATASET_DIR} \
--max_number_of_steps=${NUM_STEPS} \
--gan_type="conditional" \
--alsologtostderr
Banner "Finished training conditional GAN ${NUM_STEPS} steps."
# Run evaluation.
Banner "Starting evaluation of conditional GAN..."
python "${git_repo}/research/gan/mnist/conditional_eval.py" \
--checkpoint_dir=${CONDITIONAL_TRAIN_DIR} \
--eval_dir=${CONDITIONAL_EVAL_DIR} \
--classifier_filename=${FROZEN_GRAPH} \
--max_number_of_evaluations=1
Banner "Finished conditional evaluation. See ${CONDITIONAL_EVAL_DIR} for output images."
fi
# Run InfoGAN.
if [[ "$gan_type" == "infogan" ]]; then
INFOGAN_TRAIN_DIR="${TRAIN_DIR}/infogan"
INFOGAN_EVAL_DIR="${EVAL_DIR}/infogan"
NUM_STEPS=3000
# Run training.
Banner "Starting training infogan GAN for ${NUM_STEPS} steps..."
python "${git_repo}/research/gan/mnist/train.py" \
--train_log_dir=${INFOGAN_TRAIN_DIR} \
--dataset_dir=${DATASET_DIR} \
--max_number_of_steps=${NUM_STEPS} \
--gan_type="infogan" \
--alsologtostderr
Banner "Finished training InfoGAN ${NUM_STEPS} steps."
# Run evaluation.
Banner "Starting evaluation of infogan..."
python "${git_repo}/research/gan/mnist/infogan_eval.py" \
--checkpoint_dir=${INFOGAN_TRAIN_DIR} \
--eval_dir=${INFOGAN_EVAL_DIR} \
--classifier_filename=${FROZEN_GRAPH} \
--max_number_of_evaluations=1
Banner "Finished InfoGAN evaluation. See ${INFOGAN_EVAL_DIR} for output images."
fi
research/gan/mnist/networks.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Networks for MNIST example using TFGAN."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import tensorflow as tf
ds = tf.contrib.distributions
layers = tf.contrib.layers
tfgan = tf.contrib.gan
def _generator_helper(
noise, is_conditional, one_hot_labels, weight_decay, is_training):
"""Core MNIST generator.
This function is reused between the different GAN modes (unconditional,
conditional, etc).
Args:
noise: A 2D Tensor of shape [batch size, noise dim].
is_conditional: Whether to condition on labels.
one_hot_labels: Optional labels for conditioning.
weight_decay: The value of the l2 weight decay.
is_training: If `True`, batch norm uses batch statistics. If `False`, batch
norm uses the exponential moving average collected from population
statistics.
Returns:
A generated image in the range [-1, 1].
"""
with tf.contrib.framework.arg_scope(
[layers.fully_connected, layers.conv2d_transpose],
activation_fn=tf.nn.relu, normalizer_fn=layers.batch_norm,
weights_regularizer=layers.l2_regularizer(weight_decay)):
with tf.contrib.framework.arg_scope(
[layers.batch_norm], is_training=is_training):
net = layers.fully_connected(noise, 1024)
if is_conditional:
net = tfgan.features.condition_tensor_from_onehot(net, one_hot_labels)
net = layers.fully_connected(net, 7 * 7 * 128)
net = tf.reshape(net, [-1, 7, 7, 128])
net = layers.conv2d_transpose(net, 64, [4, 4], stride=2)
net = layers.conv2d_transpose(net, 32, [4, 4], stride=2)
# Make sure that generator output is in the same range as `inputs`
# ie [-1, 1].
net = layers.conv2d(
net, 1, [4, 4], normalizer_fn=None, activation_fn=tf.tanh)
return net
def unconditional_generator(noise, weight_decay=2.5e-5, is_training=True):
"""Generator to produce unconditional MNIST images.
Args:
noise: A single Tensor representing noise.
weight_decay: The value of the l2 weight decay.
is_training: If `True`, batch norm uses batch statistics. If `False`, batch
norm uses the exponential moving average collected from population
statistics.
Returns:
A generated image in the range [-1, 1].
"""
return _generator_helper(noise, False, None, weight_decay, is_training)
def conditional_generator(inputs, weight_decay=2.5e-5, is_training=True):
"""Generator to produce MNIST images conditioned on class.
Args:
inputs: A 2-tuple of Tensors (noise, one_hot_labels).
weight_decay: The value of the l2 weight decay.
is_training: If `True`, batch norm uses batch statistics. If `False`, batch
norm uses the exponential moving average collected from population
statistics.
Returns:
A generated image in the range [-1, 1].
"""
noise, one_hot_labels = inputs
return _generator_helper(
noise, True, one_hot_labels, weight_decay, is_training)
def infogan_generator(inputs, categorical_dim, weight_decay=2.5e-5,
is_training=True):
"""InfoGAN generator network on MNIST digits.
Based on a paper https://arxiv.org/abs/1606.03657, their code
https://github.com/openai/InfoGAN, and code by pooleb@.
Args:
inputs: A 3-tuple of Tensors (unstructured_noise, categorical structured
noise, continuous structured noise). `inputs[0]` and `inputs[2]` must be
2D, and `inputs[1]` must be 1D. All must have the same first dimension.
categorical_dim: Dimensions of the incompressible categorical noise.
weight_decay: The value of the l2 weight decay.
is_training: If `True`, batch norm uses batch statistics. If `False`, batch
norm uses the exponential moving average collected from population
statistics.
Returns:
A generated image in the range [-1, 1].
"""
unstructured_noise, cat_noise, cont_noise = inputs
cat_noise_onehot = tf.one_hot(cat_noise, categorical_dim)
all_noise = tf.concat(
[unstructured_noise, cat_noise_onehot, cont_noise], axis=1)
return _generator_helper(all_noise, False, None, weight_decay, is_training)
_leaky_relu = lambda x: tf.nn.leaky_relu(x, alpha=0.01)
def _discriminator_helper(img, is_conditional, one_hot_labels, weight_decay):
"""Core MNIST discriminator.
This function is reused between the different GAN modes (unconditional,
conditional, etc).
Args:
img: Real or generated MNIST digits. Should be in the range [-1, 1].
is_conditional: Whether to condition on labels.
one_hot_labels: Labels to optionally condition the network on.
weight_decay: The L2 weight decay.
Returns:
Final fully connected discriminator layer. [batch_size, 1024].
"""
with tf.contrib.framework.arg_scope(
[layers.conv2d, layers.fully_connected],
activation_fn=_leaky_relu, normalizer_fn=None,
weights_regularizer=layers.l2_regularizer(weight_decay),
biases_regularizer=layers.l2_regularizer(weight_decay)):
net = layers.conv2d(img, 64, [4, 4], stride=2)
net = layers.conv2d(net, 128, [4, 4], stride=2)
net = layers.flatten(net)
if is_conditional:
net = tfgan.features.condition_tensor_from_onehot(net, one_hot_labels)
net = layers.fully_connected(net, 1024, normalizer_fn=layers.layer_norm)
return net
def unconditional_discriminator(img, unused_conditioning, weight_decay=2.5e-5):
"""Discriminator network on unconditional MNIST digits.
Args:
img: Real or generated MNIST digits. Should be in the range [-1, 1].
unused_conditioning: The TFGAN API can help with conditional GANs, which
would require extra `condition` information to both the generator and the
discriminator. Since this example is not conditional, we do not use this
argument.
weight_decay: The L2 weight decay.
Returns:
Logits for the probability that the image is real.
"""
net = _discriminator_helper(img, False, None, weight_decay)
return layers.linear(net, 1)
def conditional_discriminator(img, conditioning, weight_decay=2.5e-5):
"""Conditional discriminator network on MNIST digits.
Args:
img: Real or generated MNIST digits. Should be in the range [-1, 1].
conditioning: A 2-tuple of Tensors representing (noise, one_hot_labels).
weight_decay: The L2 weight decay.
Returns:
Logits for the probability that the image is real.
"""
_, one_hot_labels = conditioning
net = _discriminator_helper(img, True, one_hot_labels, weight_decay)
return layers.linear(net, 1)
def infogan_discriminator(img, unused_conditioning, weight_decay=2.5e-5,
categorical_dim=10, continuous_dim=2):
"""InfoGAN discriminator network on MNIST digits.
Based on a paper https://arxiv.org/abs/1606.03657, their code
https://github.com/openai/InfoGAN, and code by pooleb@.
Args:
img: Real or generated MNIST digits. Should be in the range [-1, 1].
unused_conditioning: The TFGAN API can help with conditional GANs, which
would require extra `condition` information to both the generator and the
discriminator. Since this example is not conditional, we do not use this
argument.
weight_decay: The L2 weight decay.
categorical_dim: Dimensions of the incompressible categorical noise.
continuous_dim: Dimensions of the incompressible continuous noise.
Returns:
Logits for the probability that the image is real, and a list of posterior
distributions for each of the noise vectors.
"""
net = _discriminator_helper(img, False, None, weight_decay)
logits_real = layers.fully_connected(net, 1, activation_fn=None)
# Recognition network for latent variables has an additional layer
with tf.contrib.framework.arg_scope([layers.batch_norm], is_training=False):
encoder = layers.fully_connected(net, 128, normalizer_fn=layers.batch_norm,
activation_fn=_leaky_relu)
# Compute logits for each category of categorical latent.
logits_cat = layers.fully_connected(
encoder, categorical_dim, activation_fn=None)
q_cat = ds.Categorical(logits_cat)
# Compute mean for Gaussian posterior of continuous latents.
mu_cont = layers.fully_connected(encoder, continuous_dim, activation_fn=None)
sigma_cont = tf.ones_like(mu_cont)
q_cont = ds.Normal(loc=mu_cont, scale=sigma_cont)
return logits_real, [q_cat, q_cont]
research/gan/mnist/train.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Trains a generator on MNIST data."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import functools
from absl import flags
from absl import logging
import tensorflow as tf
import data_provider
import networks
import util
tfgan = tf.contrib.gan
flags.DEFINE_integer('batch_size', 32, 'The number of images in each batch.')
flags.DEFINE_string('train_log_dir', '/tmp/mnist/',
'Directory where to write event logs.')
flags.DEFINE_string('dataset_dir', None, 'Location of data.')
flags.DEFINE_integer('max_number_of_steps', 20000,
'The maximum number of gradient steps.')
flags.DEFINE_string(
'gan_type', 'unconditional',
'Either `unconditional`, `conditional`, or `infogan`.')
flags.DEFINE_integer(
'grid_size', 5, 'Grid size for image visualization.')
flags.DEFINE_integer(
'noise_dims', 64, 'Dimensions of the generator noise vector.')
FLAGS = flags.FLAGS
def _learning_rate(gan_type):
# First is generator learning rate, second is discriminator learning rate.
return {
'unconditional': (1e-3, 1e-4),
'conditional': (1e-5, 1e-4),
'infogan': (0.001, 9e-5),
}[gan_type]
def main(_):
if not tf.gfile.Exists(FLAGS.train_log_dir):
tf.gfile.MakeDirs(FLAGS.train_log_dir)
# Force all input processing onto CPU in order to reserve the GPU for
# the forward inference and back-propagation.
with tf.name_scope('inputs'):
with tf.device('/cpu:0'):
images, one_hot_labels, _ = data_provider.provide_data(
'train', FLAGS.batch_size, FLAGS.dataset_dir, num_threads=4)
# Define the GANModel tuple. Optionally, condition the GAN on the label or
# use an InfoGAN to learn a latent representation.
if FLAGS.gan_type == 'unconditional':
gan_model = tfgan.gan_model(
generator_fn=networks.unconditional_generator,
discriminator_fn=networks.unconditional_discriminator,
real_data=images,
generator_inputs=tf.random_normal(
[FLAGS.batch_size, FLAGS.noise_dims]))
elif FLAGS.gan_type == 'conditional':
noise = tf.random_normal([FLAGS.batch_size, FLAGS.noise_dims])
gan_model = tfgan.gan_model(
generator_fn=networks.conditional_generator,
discriminator_fn=networks.conditional_discriminator,
real_data=images,
generator_inputs=(noise, one_hot_labels))
elif FLAGS.gan_type == 'infogan':
cat_dim, cont_dim = 10, 2
generator_fn = functools.partial(
networks.infogan_generator, categorical_dim=cat_dim)
discriminator_fn = functools.partial(
networks.infogan_discriminator, categorical_dim=cat_dim,
continuous_dim=cont_dim)
unstructured_inputs, structured_inputs = util.get_infogan_noise(
FLAGS.batch_size, cat_dim, cont_dim, FLAGS.noise_dims)
gan_model = tfgan.infogan_model(
generator_fn=generator_fn,
discriminator_fn=discriminator_fn,
real_data=images,
unstructured_generator_inputs=unstructured_inputs,
structured_generator_inputs=structured_inputs)
tfgan.eval.add_gan_model_image_summaries(gan_model, FLAGS.grid_size)
# Get the GANLoss tuple. You can pass a custom function, use one of the
# already-implemented losses from the losses library, or use the defaults.
with tf.name_scope('loss'):
mutual_information_penalty_weight = (1.0 if FLAGS.gan_type == 'infogan'
else 0.0)
gan_loss = tfgan.gan_loss(
gan_model,
gradient_penalty_weight=1.0,
mutual_information_penalty_weight=mutual_information_penalty_weight,
add_summaries=True)
tfgan.eval.add_regularization_loss_summaries(gan_model)
# Get the GANTrain ops using custom optimizers.
with tf.name_scope('train'):
gen_lr, dis_lr = _learning_rate(FLAGS.gan_type)
train_ops = tfgan.gan_train_ops(
gan_model,
gan_loss,
generator_optimizer=tf.train.AdamOptimizer(gen_lr, 0.5),
discriminator_optimizer=tf.train.AdamOptimizer(dis_lr, 0.5),
summarize_gradients=True,
aggregation_method=tf.AggregationMethod.EXPERIMENTAL_ACCUMULATE_N)
# Run the alternating training loop. Skip it if no steps should be taken
# (used for graph construction tests).
status_message = tf.string_join(
['Starting train step: ',
tf.as_string(tf.train.get_or_create_global_step())],
name='status_message')
if FLAGS.max_number_of_steps == 0: return
tfgan.gan_train(
train_ops,
hooks=[tf.train.StopAtStepHook(num_steps=FLAGS.max_number_of_steps),
tf.train.LoggingTensorHook([status_message], every_n_iter=10)],
logdir=FLAGS.train_log_dir,
get_hooks_fn=tfgan.get_joint_train_hooks())
if __name__ == '__main__':
logging.set_verbosity(logging.INFO)
tf.app.run()
research/gan/mnist/train_test.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for mnist.train."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from absl import flags
from absl.testing import parameterized
import numpy as np
import tensorflow as tf
import train
FLAGS = flags.FLAGS
mock = tf.test.mock
class TrainTest(tf.test.TestCase, parameterized.TestCase):
@mock.patch.object(train, 'data_provider', autospec=True)
def test_run_one_train_step(self, mock_data_provider):
FLAGS.max_number_of_steps = 1
FLAGS.gan_type = 'unconditional'
FLAGS.batch_size = 5
FLAGS.grid_size = 1
tf.set_random_seed(1234)
# Mock input pipeline.
mock_imgs = np.zeros([FLAGS.batch_size, 28, 28, 1], dtype=np.float32)
mock_lbls = np.concatenate(
(np.ones([FLAGS.batch_size, 1], dtype=np.int32),
np.zeros([FLAGS.batch_size, 9], dtype=np.int32)), axis=1)
mock_data_provider.provide_data.return_value = (mock_imgs, mock_lbls, None)
train.main(None)
@parameterized.named_parameters(
('Unconditional', 'unconditional'),
('Conditional', 'conditional'),
('InfoGAN', 'infogan'))
def test_build_graph(self, gan_type):
FLAGS.max_number_of_steps = 0
FLAGS.gan_type = gan_type
# Mock input pipeline.
mock_imgs = np.zeros([FLAGS.batch_size, 28, 28, 1], dtype=np.float32)
mock_lbls = np.concatenate(
(np.ones([FLAGS.batch_size, 1], dtype=np.int32),
np.zeros([FLAGS.batch_size, 9], dtype=np.int32)), axis=1)
with mock.patch.object(train, 'data_provider') as mock_data_provider:
mock_data_provider.provide_data.return_value = (
mock_imgs, mock_lbls, None)
train.main(None)
if __name__ == '__main__':
tf.test.main()
research/gan/mnist/util.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""A utility for evaluating MNIST generative models.
These functions use a pretrained MNIST classifier with ~99% eval accuracy to
measure various aspects of the quality of generated MNIST digits.
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import numpy as np
from six.moves import xrange # pylint: disable=redefined-builtin
import tensorflow as tf
ds = tf.contrib.distributions
tfgan = tf.contrib.gan
__all__ = [
'mnist_score',
'mnist_frechet_distance',
'mnist_cross_entropy',
'get_eval_noise_categorical',
'get_eval_noise_continuous_dim1',
'get_eval_noise_continuous_dim2',
'get_infogan_noise',
]
# Prepend `../`, since paths start from `third_party/tensorflow`.
MODEL_GRAPH_DEF = '../tensorflow_models/gan/mnist/data/classify_mnist_graph_def.pb'
INPUT_TENSOR = 'inputs:0'
OUTPUT_TENSOR = 'logits:0'
def mnist_score(images, graph_def_filename=None, input_tensor=INPUT_TENSOR,
output_tensor=OUTPUT_TENSOR, num_batches=1):
"""Get MNIST logits of a fully-trained classifier.
Args:
images: A minibatch tensor of MNIST digits. Shape must be
[batch, 28, 28, 1].
graph_def_filename: Location of a frozen GraphDef binary file on disk. If
`None`, uses a default graph.
input_tensor: GraphDef's input tensor name.
output_tensor: GraphDef's output tensor name.
num_batches: Number of batches to split `generated_images` in to in order to
efficiently run them through Inception.
Returns:
A logits tensor of [batch, 10].
"""
images.shape.assert_is_compatible_with([None, 28, 28, 1])
graph_def = _graph_def_from_par_or_disk(graph_def_filename)
mnist_classifier_fn = lambda x: tfgan.eval.run_image_classifier( # pylint: disable=g-long-lambda
x, graph_def, input_tensor, output_tensor)
score = tfgan.eval.classifier_score(
images, mnist_classifier_fn, num_batches)
score.shape.assert_is_compatible_with([])
return score
def mnist_frechet_distance(real_images, generated_images,
graph_def_filename=None, input_tensor=INPUT_TENSOR,
output_tensor=OUTPUT_TENSOR, num_batches=1):
"""Frechet distance between real and generated images.
This technique is described in detail in https://arxiv.org/abs/1706.08500.
Please see TFGAN for implementation details
(https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/gan/
python/eval/python/classifier_metrics_impl.py).
Args:
real_images: Real images to use to compute Frechet Inception distance.
generated_images: Generated images to use to compute Frechet Inception
distance.
graph_def_filename: Location of a frozen GraphDef binary file on disk. If
`None`, uses a default graph.
input_tensor: GraphDef's input tensor name.
output_tensor: GraphDef's output tensor name.
num_batches: Number of batches to split images into in order to
efficiently run them through the classifier network.
Returns:
The Frechet distance. A floating-point scalar.
"""
real_images.shape.assert_is_compatible_with([None, 28, 28, 1])
generated_images.shape.assert_is_compatible_with([None, 28, 28, 1])
graph_def = _graph_def_from_par_or_disk(graph_def_filename)
mnist_classifier_fn = lambda x: tfgan.eval.run_image_classifier( # pylint: disable=g-long-lambda
x, graph_def, input_tensor, output_tensor)
frechet_distance = tfgan.eval.frechet_classifier_distance(
real_images, generated_images, mnist_classifier_fn, num_batches)
frechet_distance.shape.assert_is_compatible_with([])
return frechet_distance
def mnist_cross_entropy(images, one_hot_labels, graph_def_filename=None,
input_tensor=INPUT_TENSOR, output_tensor=OUTPUT_TENSOR):
"""Returns the cross entropy loss of the classifier on images.
Args:
images: A minibatch tensor of MNIST digits. Shape must be
[batch, 28, 28, 1].
one_hot_labels: The one hot label of the examples. Tensor size is
[batch, 10].
graph_def_filename: Location of a frozen GraphDef binary file on disk. If
`None`, uses a default graph embedded in the par file.
input_tensor: GraphDef's input tensor name.
output_tensor: GraphDef's output tensor name.
Returns:
A scalar Tensor representing the cross entropy of the image minibatch.
"""
graph_def = _graph_def_from_par_or_disk(graph_def_filename)
logits = tfgan.eval.run_image_classifier(
images, graph_def, input_tensor, output_tensor)
return tf.losses.softmax_cross_entropy(
one_hot_labels, logits, loss_collection=None)
# (joelshor): Refactor the `eval_noise` functions to reuse code.
def get_eval_noise_categorical(
noise_samples, categorical_sample_points, continuous_sample_points,
unstructured_noise_dims, continuous_noise_dims):
"""Create noise showing impact of categorical noise in InfoGAN.
Categorical noise is constant across columns. Other noise is constant across
rows.
Args:
noise_samples: Number of non-categorical noise samples to use.
categorical_sample_points: Possible categorical noise points to sample.
continuous_sample_points: Possible continuous noise points to sample.
unstructured_noise_dims: Dimensions of the unstructured noise.
continuous_noise_dims: Dimensions of the continuous noise.
Returns:
Unstructured noise, categorical noise, continuous noise numpy arrays. Each
should have shape [noise_samples, ?].
"""
rows, cols = noise_samples, len(categorical_sample_points)
# Take random draws for non-categorical noise, making sure they are constant
# across columns.
unstructured_noise = []
for _ in xrange(rows):
cur_sample = np.random.normal(size=[1, unstructured_noise_dims])
unstructured_noise.extend([cur_sample] * cols)
unstructured_noise = np.concatenate(unstructured_noise)
continuous_noise = []
for _ in xrange(rows):
cur_sample = np.random.choice(
continuous_sample_points, size=[1, continuous_noise_dims])
continuous_noise.extend([cur_sample] * cols)
continuous_noise = np.concatenate(continuous_noise)
# Increase categorical noise from left to right, making sure they are constant
# across rows.
categorical_noise = np.tile(categorical_sample_points, rows)
return unstructured_noise, categorical_noise, continuous_noise
def get_eval_noise_continuous_dim1(
noise_samples, categorical_sample_points, continuous_sample_points,
unstructured_noise_dims, continuous_noise_dims): # pylint:disable=unused-argument
"""Create noise showing impact of first dim continuous noise in InfoGAN.
First dimension of continuous noise is constant across columns. Other noise is
constant across rows.
Args:
noise_samples: Number of non-categorical noise samples to use.
categorical_sample_points: Possible categorical noise points to sample.
continuous_sample_points: Possible continuous noise points to sample.
unstructured_noise_dims: Dimensions of the unstructured noise.
continuous_noise_dims: Dimensions of the continuous noise.
Returns:
Unstructured noise, categorical noise, continuous noise numpy arrays.
"""
rows, cols = noise_samples, len(continuous_sample_points)
# Take random draws for non-first-dim-continuous noise, making sure they are
# constant across columns.
unstructured_noise = []
for _ in xrange(rows):
cur_sample = np.random.normal(size=[1, unstructured_noise_dims])
unstructured_noise.extend([cur_sample] * cols)
unstructured_noise = np.concatenate(unstructured_noise)
categorical_noise = []
for _ in xrange(rows):
cur_sample = np.random.choice(categorical_sample_points)
categorical_noise.extend([cur_sample] * cols)
categorical_noise = np.array(categorical_noise)
cont_noise_dim2 = []
for _ in xrange(rows):
cur_sample = np.random.choice(continuous_sample_points, size=[1, 1])
cont_noise_dim2.extend([cur_sample] * cols)
cont_noise_dim2 = np.concatenate(cont_noise_dim2)
# Increase first dimension of continuous noise from left to right, making sure
# they are constant across rows.
cont_noise_dim1 = np.expand_dims(np.tile(continuous_sample_points, rows), 1)
continuous_noise = np.concatenate((cont_noise_dim1, cont_noise_dim2), 1)
return unstructured_noise, categorical_noise, continuous_noise
def get_eval_noise_continuous_dim2(
noise_samples, categorical_sample_points, continuous_sample_points,
unstructured_noise_dims, continuous_noise_dims): # pylint:disable=unused-argument
"""Create noise showing impact of second dim of continuous noise in InfoGAN.
Second dimension of continuous noise is constant across columns. Other noise
is constant across rows.
Args:
noise_samples: Number of non-categorical noise samples to use.
categorical_sample_points: Possible categorical noise points to sample.
continuous_sample_points: Possible continuous noise points to sample.
unstructured_noise_dims: Dimensions of the unstructured noise.
continuous_noise_dims: Dimensions of the continuous noise.
Returns:
Unstructured noise, categorical noise, continuous noise numpy arrays.
"""
rows, cols = noise_samples, len(continuous_sample_points)
# Take random draws for non-first-dim-continuous noise, making sure they are
# constant across columns.
unstructured_noise = []
for _ in xrange(rows):
cur_sample = np.random.normal(size=[1, unstructured_noise_dims])
unstructured_noise.extend([cur_sample] * cols)
unstructured_noise = np.concatenate(unstructured_noise)
categorical_noise = []
for _ in xrange(rows):
cur_sample = np.random.choice(categorical_sample_points)
categorical_noise.extend([cur_sample] * cols)
categorical_noise = np.array(categorical_noise)
cont_noise_dim1 = []
for _ in xrange(rows):
cur_sample = np.random.choice(continuous_sample_points, size=[1, 1])
cont_noise_dim1.extend([cur_sample] * cols)
cont_noise_dim1 = np.concatenate(cont_noise_dim1)
# Increase first dimension of continuous noise from left to right, making sure
# they are constant across rows.
cont_noise_dim2 = np.expand_dims(np.tile(continuous_sample_points, rows), 1)
continuous_noise = np.concatenate((cont_noise_dim1, cont_noise_dim2), 1)
return unstructured_noise, categorical_noise, continuous_noise
def get_infogan_noise(batch_size, categorical_dim, structured_continuous_dim,
total_continuous_noise_dims):
"""Get unstructured and structured noise for InfoGAN.
Args:
batch_size: The number of noise vectors to generate.
categorical_dim: The number of categories in the categorical noise.
structured_continuous_dim: The number of dimensions of the uniform
continuous noise.
total_continuous_noise_dims: The number of continuous noise dimensions. This
number includes the structured and unstructured noise.
Returns:
A 2-tuple of structured and unstructured noise. First element is the
unstructured noise, and the second is a 2-tuple of
(categorical structured noise, continuous structured noise).
"""
# Get unstructurd noise.
unstructured_noise = tf.random_normal(
[batch_size, total_continuous_noise_dims - structured_continuous_dim])
# Get categorical noise Tensor.
categorical_dist = ds.Categorical(logits=tf.zeros([categorical_dim]))
categorical_noise = categorical_dist.sample([batch_size])
# Get continuous noise Tensor.
continuous_dist = ds.Uniform(-tf.ones([structured_continuous_dim]),
tf.ones([structured_continuous_dim]))
continuous_noise = continuous_dist.sample([batch_size])
return [unstructured_noise], [categorical_noise, continuous_noise]
def _graph_def_from_par_or_disk(filename):
if filename is None:
return tfgan.eval.get_graph_def_from_resource(MODEL_GRAPH_DEF)
else:
return tfgan.eval.get_graph_def_from_disk(filename)
research/gan/mnist/util_test.py deleted 100644 → 0
View file @ f673f7a8
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for mnist.util."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import numpy as np
import tensorflow as tf
from google3.third_party.tensorflow_models.gan.mnist import util
# pylint: disable=line-too-long
# This is a real digit `5` from MNIST.
REAL_DIGIT = [[-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.9765625, -0.859375, -0.859375, -0.859375, -0.015625, 0.0625, 0.3671875, -0.796875, 0.296875, 0.9921875, 0.9296875, -0.0078125, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.765625, -0.71875, -0.265625, 0.203125, 0.328125, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.7578125, 0.34375, 0.9765625, 0.890625, 0.5234375, -0.5, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.6171875, 0.859375, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.9609375, -0.2734375, -0.359375, -0.359375, -0.5625, -0.6953125, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.859375, 0.7109375, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.546875, 0.421875, 0.9296875, 0.8828125, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.375, 0.21875, -0.1640625, 0.9765625, 0.9765625, 0.6015625, -0.9140625, -1.0, -0.6640625, 0.203125, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.890625, -0.9921875, 0.203125, 0.9765625, -0.296875, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, 0.0859375, 0.9765625, 0.484375, -0.984375, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.9140625, 0.484375, 0.9765625, -0.453125, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.7265625, 0.8828125, 0.7578125, 0.25, -0.15625, -0.9921875, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.3671875, 0.875, 0.9765625, 0.9765625, -0.0703125, -0.8046875, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.6484375, 0.453125, 0.9765625, 0.9765625, 0.171875, -0.7890625, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.875, -0.2734375, 0.96875, 0.9765625, 0.4609375, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, 0.9453125, 0.9765625, 0.9453125, -0.5, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.640625, 0.015625, 0.4296875, 0.9765625, 0.9765625, 0.6171875, -0.984375, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.6953125, 0.15625, 0.7890625, 0.9765625, 0.9765625, 0.9765625, 0.953125, 0.421875, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.8125, -0.109375, 0.7265625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.5703125, -0.390625, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.8203125, -0.484375, 0.6640625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.546875, -0.3671875, -0.984375, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -0.859375, 0.3359375, 0.7109375, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.5234375, -0.375, -0.9296875, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -0.5703125, 0.34375, 0.765625, 0.9765625, 0.9765625, 0.9765625, 0.9765625, 0.90625, 0.0390625, -0.9140625, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, 0.0625, 0.9765625, 0.9765625, 0.9765625, 0.65625, 0.0546875, 0.03125, -0.875, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0], [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0, -1.0]]
ONE_HOT = [[0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0]]
# Uniform noise in [-1, 1].
FAKE_DIGIT = [[0.778958797454834, 0.8792028427124023, 0.07099628448486328, 0.8518857955932617, -0.3541288375854492, -0.7431280612945557, -0.12607860565185547, 0.17328786849975586, 0.6749839782714844, -0.5402040481567383, 0.9034252166748047, 0.2420203685760498, 0.3455841541290283, 0.1937558650970459, 0.9989571571350098, 0.9039363861083984, -0.955411434173584, 0.6228537559509277, -0.33131909370422363, 0.9653763771057129, 0.864208459854126, -0.05056142807006836, 0.12686634063720703, -0.09225749969482422, 0.49758028984069824, 0.08698725700378418, 0.5533185005187988, 0.20227980613708496], [0.8400616645812988, 0.7409703731536865, -0.6215496063232422, -0.53228759765625, -0.20184636116027832, -0.8568699359893799, -0.8662903308868408, -0.8735041618347168, -0.11022663116455078, -0.8418543338775635, 0.8193502426147461, -0.901512622833252, -0.7680232524871826, 0.6209826469421387, 0.06459426879882812, 0.5341305732727051, -0.4078702926635742, -0.13658642768859863, 0.6602437496185303, 0.848508358001709, -0.23431801795959473, 0.5995683670043945, -0.9807922840118408, 0.2657158374786377, -0.8068397045135498, 0.2438051700592041, -0.2116842269897461, 0.011460304260253906], [0.00040912628173828125, -0.058798789978027344, 0.3239307403564453, 0.5040378570556641, -0.03192305564880371, -0.4816470146179199, -0.14559340476989746, -0.9231269359588623, -0.6602556705474854, -0.2537086009979248, -0.11059761047363281, -0.8174862861633301, 0.6180260181427002, 0.7245023250579834, 0.5007762908935547, -0.1575303077697754, -0.0167086124420166, 0.7173266410827637, 0.1126704216003418, -0.9878268241882324, 0.4538843631744385, -0.4422755241394043, -0.7899672985076904, 0.7349567413330078, -0.4448075294494629, -0.7548923492431641, -0.5739786624908447, 0.30504918098449707], [-0.8488152027130127, 0.43424463272094727, 0.7724254131317139, 0.43314504623413086, 0.7352848052978516, -0.26010799407958984, 0.43951940536499023, -0.7642686367034912, -0.657184362411499, -0.9933960437774658, -0.47258639335632324, 0.10390830039978027, 0.11454653739929199, -0.6156411170959473, -0.23431062698364258, -0.6897118091583252, -0.5721850395202637, -0.3574075698852539, 0.13927006721496582, -0.6530766487121582, 0.32231855392456055, 0.6294634342193604, 0.5507853031158447, -0.4867420196533203, 0.4329197406768799, 0.6168341636657715, -0.8720219135284424, 0.8639121055603027], [0.02407360076904297, -0.11185193061828613, -0.38637852668762207, -0.8244953155517578, -0.648916482925415, 0.44907593727111816, -0.368192195892334, 0.0190126895904541, -0.9450500011444092, 0.41033458709716797, -0.7877917289733887, 0.617938756942749, 0.551692008972168, -0.48288512229919434, 0.019921541213989258, -0.8765170574188232, 0.5651748180389404, 0.850874662399292, -0.5792787075042725, 0.1748213768005371, -0.6905481815338135, -0.521310567855835, 0.062479496002197266, 0.17763280868530273, -0.4628307819366455, 0.8870463371276855, -0.8685822486877441, -0.29169774055480957], [-0.14687561988830566, 0.8801963329315186, 0.11353135108947754, -0.6009430885314941, 0.8818719387054443, -0.8621203899383545, -0.48749589920043945, 0.5224916934967041, 0.7050364017486572, -0.9968757629394531, 0.7235188484191895, 0.662771463394165, 0.588390588760376, -0.9624209403991699, 0.39203453063964844, -0.2210233211517334, 0.9266352653503418, -0.9132544994354248, -0.5175333023071289, 0.7251780033111572, 0.3030557632446289, 0.5743863582611084, 0.14350247383117676, -0.3735086917877197, -0.6299927234649658, 0.5088682174682617, -0.6953752040863037, 0.23583650588989258], [0.25864362716674805, 0.5736973285675049, -0.3975222110748291, -0.6369199752807617, 0.5376672744750977, -0.19951462745666504, 0.6843924522399902, 0.6296660900115967, 0.36865997314453125, -0.7243289947509766, 0.5768749713897705, -0.5493001937866211, 0.31238412857055664, 0.21019506454467773, -0.368206262588501, -0.33148622512817383, -0.3421964645385742, -0.15616083145141602, 0.6617193222045898, 0.4400820732116699, 0.7893157005310059, -0.2935945987701416, 0.6241741180419922, -0.26036930084228516, -0.6958446502685547, 0.27047157287597656, -0.9095940589904785, 0.9525108337402344], [-0.5233585834503174, -0.45003342628479004, 0.15099048614501953, 0.6257956027984619, 0.9017877578735352, -0.18155455589294434, -0.20237135887145996, -0.014468908309936523, 0.01797318458557129, -0.5453977584838867, 0.21428155899047852, -0.9678947925567627, 0.5137600898742676, -0.1094369888305664, -0.13572359085083008, -0.1704423427581787, -0.9122319221496582, 0.8274900913238525, -0.11746454238891602, -0.8701446056365967, -0.9545385837554932, -0.6735866069793701, 0.9445557594299316, -0.3842940330505371, -0.6240942478179932, -0.1673595905303955, -0.3959221839904785, -0.05602693557739258], [0.8833198547363281, 0.14288711547851562, -0.9623878002166748, -0.26968836784362793, 0.9689288139343262, -0.3792128562927246, 0.2520296573638916, 0.1477947235107422, -0.24453139305114746, 0.94329833984375, 0.8014910221099854, 0.5443501472473145, 0.8486857414245605, -0.0795745849609375, -0.30250096321105957, -0.909733772277832, 0.8387842178344727, -0.41989898681640625, -0.8364224433898926, 0.04792976379394531, -0.38036274909973145, 0.12747883796691895, -0.5356688499450684, -0.04269552230834961, -0.2070469856262207, -0.6911153793334961, 0.33954668045043945, 0.25260138511657715], [0.3128373622894287, 0.36142778396606445, -0.2512378692626953, -0.6497259140014648, -0.21787405014038086, -0.9972476959228516, -0.026904821395874023, -0.4214746952056885, -0.3354766368865967, -0.6112637519836426, -0.7594058513641357, 0.09093379974365234, -0.7331845760345459, 0.5222046375274658, -0.8997514247894287, -0.3749384880065918, -0.0775001049041748, -0.26296448707580566, 0.404325008392334, -0.25776195526123047, 0.9136955738067627, 0.2623283863067627, -0.6411356925964355, 0.6646602153778076, -0.12833356857299805, -0.4184732437133789, -0.3663449287414551, -0.5468103885650635], [0.3770618438720703, 0.5572817325592041, -0.3657073974609375, -0.5056321620941162, 0.6555137634277344, 0.9557509422302246, -0.6900768280029297, -0.4980638027191162, 0.05510354042053223, -0.610318660736084, 0.9753992557525635, 0.7569930553436279, 0.4011664390563965, 0.8439173698425293, -0.5921270847320557, -0.2775266170501709, -0.061129093170166016, -0.49707984924316406, 0.5820951461791992, 0.008175849914550781, 0.06372833251953125, 0.3061811923980713, -0.5091361999511719, 0.9751057624816895, 0.4571402072906494, 0.6769094467163086, -0.46695923805236816, 0.44080281257629395], [-0.6510493755340576, -0.032715559005737305, 0.3482983112335205, -0.8135421276092529, 0.1506943702697754, 0.5220685005187988, -0.8834004402160645, -0.908900260925293, 0.3211519718170166, 0.896381139755249, -0.9448244571685791, -0.6193962097167969, -0.009401559829711914, 0.38227057456970215, -0.9219558238983154, -0.029483318328857422, -0.3889012336730957, 0.5242419242858887, 0.7338912487030029, -0.8713808059692383, -0.04948568344116211, -0.797940731048584, -0.9933724403381348, -0.262890100479126, -0.7165846824645996, -0.9763388633728027, -0.4105076789855957, 0.5907857418060303], [-0.6091890335083008, 0.7921168804168701, -0.033307790756225586, -0.9177074432373047, 0.4553513526916504, 0.5754055976867676, 0.6747269630432129, 0.0015664100646972656, -0.36865878105163574, -0.7999486923217773, 0.993431568145752, -0.7310445308685303, -0.49965715408325195, -0.028263330459594727, -0.20190834999084473, -0.7398116588592529, 0.10513901710510254, 0.22136950492858887, 0.42579007148742676, -0.4703383445739746, -0.8729751110076904, 0.28951215744018555, -0.3110074996948242, 0.9935362339019775, -0.29533815383911133, -0.3384673595428467, -0.07292437553405762, 0.8471579551696777], [-0.04648447036743164, -0.6876633167266846, -0.4921104907989502, -0.1925184726715088, -0.5843420028686523, -0.2852492332458496, 0.1251826286315918, -0.5709969997406006, 0.6744673252105713, 0.17812824249267578, 0.675086259841919, 0.1219022274017334, 0.6496286392211914, 0.05892682075500488, 0.33709263801574707, -0.9087743759155273, -0.4785141944885254, 0.2689247131347656, 0.3600578308105469, -0.6822943687438965, -0.0801839828491211, 0.9234893321990967, 0.5037875175476074, -0.8148105144500732, 0.6820621490478516, -0.8345451354980469, 0.9400079250335693, -0.752265453338623], [0.4599113464355469, 0.0292510986328125, -0.7475054264068604, -0.4074263572692871, 0.8800950050354004, 0.41760849952697754, 0.3225588798522949, 0.8359887599945068, -0.8655965328216553, 0.17258906364440918, 0.752063512802124, -0.48968982696533203, -0.9149389266967773, -0.46224403381347656, -0.26379919052124023, -0.9342350959777832, -0.5444085597991943, -0.4576752185821533, -0.12544608116149902, -0.3810274600982666, -0.5277583599090576, 0.3267025947570801, 0.4762594699859619, 0.09010529518127441, -0.2434844970703125, -0.8013439178466797, -0.23540687561035156, 0.8844473361968994], [-0.8922028541564941, -0.7912418842315674, -0.14429497718811035, -0.3925011157989502, 0.1870102882385254, -0.28865933418273926, 0.11481523513793945, -0.9174098968505859, -0.5264348983764648, -0.7296302318572998, 0.44644737243652344, 0.11140275001525879, 0.08105874061584473, 0.3871574401855469, -0.5030152797698975, -0.9560322761535645, -0.07085466384887695, 0.7949709892272949, 0.37600183486938477, -0.8664641380310059, -0.08428549766540527, 0.9169652462005615, 0.9479010105133057, 0.3576698303222656, 0.6677501201629639, -0.34919166564941406, -0.11635351181030273, -0.05966901779174805], [-0.9716870784759521, -0.7901370525360107, 0.9152195453643799, -0.42171406745910645, 0.20472931861877441, -0.9847748279571533, 0.9935972690582275, -0.5975158214569092, 0.9557573795318604, 0.41234254837036133, -0.26670074462890625, 0.869682788848877, 0.4443938732147217, -0.13968205451965332, -0.7745835781097412, -0.5692052841186523, 0.321591854095459, -0.3146944046020508, -0.3921670913696289, -0.36029601097106934, 0.33528995513916016, 0.029220104217529297, -0.8788590431213379, 0.6883881092071533, -0.8260040283203125, -0.43041515350341797, 0.42810845375061035, 0.40050792694091797], [0.6319584846496582, 0.32369279861450195, -0.6308813095092773, 0.07861590385437012, 0.5387494564056396, -0.902024507522583, -0.5346510410308838, 0.10787153244018555, 0.36048197746276855, 0.7801573276519775, -0.39187049865722656, 0.409869909286499, 0.5972449779510498, -0.7578165531158447, 0.30403685569763184, -0.7885205745697021, 0.01341390609741211, -0.023469209671020508, -0.11588907241821289, -0.941727876663208, -0.09618496894836426, 0.8042230606079102, 0.4683551788330078, -0.6497313976287842, 0.7443571090698242, 0.7150907516479492, -0.5759801864624023, -0.6523513793945312], [0.3150167465209961, -0.1853046417236328, 0.1839759349822998, -0.9234504699707031, -0.666614294052124, -0.740748405456543, 0.5700008869171143, 0.4091987609863281, -0.8912513256072998, 0.027419567108154297, 0.07219696044921875, -0.3128829002380371, 0.9465951919555664, 0.8715424537658691, -0.4559173583984375, -0.5862705707550049, -0.6734106540679932, 0.8419013023376465, -0.5523068904876709, -0.5019669532775879, -0.4969966411590576, -0.030994892120361328, 0.7572557926177979, 0.3824281692504883, 0.6366133689880371, -0.13271117210388184, 0.632627010345459, -0.9462625980377197], [-0.3923935890197754, 0.5466675758361816, -0.9815363883972168, -0.12867975234985352, 0.9932572841644287, -0.712486743927002, -0.17107725143432617, -0.41582536697387695, 0.10990118980407715, 0.11867499351501465, -0.7774863243103027, -0.9382553100585938, 0.4290587902069092, -0.1412363052368164, -0.498490571975708, 0.7793624401092529, 0.9015710353851318, -0.5107312202453613, 0.12394595146179199, 0.4988982677459717, -0.5731511116027832, -0.8105146884918213, 0.19758391380310059, 0.4081540107727051, -0.20432400703430176, 0.9924988746643066, -0.16952204704284668, 0.45574188232421875], [-0.023319005966186523, -0.3514130115509033, 0.4652390480041504, 0.7690563201904297, 0.7906622886657715, 0.9922001361846924, -0.6670932769775391, -0.5720524787902832, 0.3491201400756836, -0.18706512451171875, 0.4899415969848633, -0.5637645721435547, 0.8986928462982178, -0.7359066009521484, -0.6342356204986572, 0.6608924865722656, 0.6014213562011719, 0.7906208038330078, 0.19034814834594727, 0.16884255409240723, -0.8803558349609375, 0.4060337543487549, 0.9862797260284424, -0.48758840560913086, 0.3894319534301758, 0.4226539134979248, 0.00042724609375, -0.3623659610748291], [0.7065057754516602, -0.4103825092315674, 0.3343489170074463, -0.9341757297515869, -0.20749878883361816, -0.3589310646057129, -0.9470062255859375, 0.7851138114929199, -0.74678635597229, -0.05102086067199707, 0.16727972030639648, 0.12385678291320801, -0.41132497787475586, 0.5856695175170898, 0.06311273574829102, 0.8238284587860107, 0.2257852554321289, -0.6452250480651855, -0.6373245716094971, -0.1247873306274414, 0.0021851062774658203, -0.5648598670959473, 0.8261771202087402, 0.3713812828063965, -0.7185893058776855, 0.45911359786987305, 0.6722264289855957, -0.31804966926574707], [-0.2952606678009033, -0.12052011489868164, 0.9074821472167969, -0.9850583076477051, 0.6732273101806641, 0.7954013347625732, 0.938248872756958, -0.35915136337280273, -0.8291504383087158, 0.6020793914794922, -0.30096912384033203, 0.6136975288391113, 0.46443629264831543, -0.9057025909423828, 0.43993186950683594, -0.11653375625610352, 0.9514555931091309, 0.7985148429870605, -0.6911783218383789, -0.931948184967041, 0.06829452514648438, 0.711458683013916, 0.17953252792358398, 0.19076848030090332, 0.6339912414550781, -0.38822221755981445, 0.09893226623535156, -0.7663829326629639], [0.2640511989593506, 0.16821074485778809, 0.9845118522644043, 0.13789963722229004, -0.1097862720489502, 0.24889636039733887, 0.12135696411132812, 0.4419589042663574, -0.022361040115356445, 0.3793671131134033, 0.7053709030151367, 0.5814387798309326, 0.24962759017944336, -0.40136122703552246, -0.364804744720459, -0.2518625259399414, 0.25757312774658203, 0.6828348636627197, -0.08237361907958984, -0.07745933532714844, -0.9299273490905762, -0.3066704273223877, 0.22127842903137207, -0.6690163612365723, 0.7477891445159912, 0.5738682746887207, -0.9027633666992188, 0.33333301544189453], [0.44417595863342285, 0.5279359817504883, 0.36589932441711426, -0.049490928649902344, -0.3003063201904297, 0.8447706699371338, 0.8139019012451172, -0.4958505630493164, -0.3640005588531494, -0.7731854915618896, 0.4138674736022949, 0.3080577850341797, -0.5615301132202148, -0.4247474670410156, -0.25992918014526367, -0.7983508110046387, -0.4926283359527588, 0.08595061302185059, 0.9205574989318848, 0.6025278568267822, 0.37128734588623047, -0.1856670379638672, 0.9658422470092773, -0.6652145385742188, -0.2217710018157959, 0.8311929702758789, 0.9117603302001953, 0.4660191535949707], [-0.6164810657501221, -0.9509942531585693, 0.30228447914123535, 0.39792585372924805, -0.45194506645202637, -0.3315877914428711, 0.5393261909484863, 0.030223369598388672, 0.6013507843017578, 0.020318031311035156, 0.17352747917175293, -0.25470709800720215, -0.7904071807861328, 0.5383470058441162, -0.43855953216552734, -0.350492000579834, -0.5038156509399414, 0.0511777400970459, 0.9628784656524658, -0.520085334777832, 0.33083176612854004, 0.3698580265045166, 0.9121549129486084, -0.664802074432373, -0.45629024505615234, -0.44208550453186035, 0.6502475738525391, -0.597346305847168], [-0.5405080318450928, 0.2640984058380127, -0.2119138240814209, 0.04504036903381348, -0.10604047775268555, -0.8093295097351074, 0.20915007591247559, 0.08994936943054199, 0.5347979068756104, -0.550915002822876, 0.3008608818054199, 0.2416989803314209, 0.025292158126831055, -0.2790646553039551, 0.5850257873535156, 0.7020430564880371, -0.9527721405029297, 0.8273317813873291, -0.8028199672698975, 0.3686351776123047, -0.9391682147979736, -0.4261181354522705, 0.1674947738647461, 0.27993154525756836, 0.7567532062530518, 0.9245085716247559, 0.9245762825012207, -0.296356201171875], [-0.7747671604156494, 0.7632861137390137, 0.10236740112304688, 0.14044547080993652, 0.9318621158599854, -0.5622165203094482, -0.6605725288391113, -0.8286299705505371, 0.8717818260192871, -0.22177672386169434, -0.6030778884887695, 0.20917797088623047, 0.31551361083984375, 0.7741527557373047, -0.3320643901824951, -0.9014863967895508, 0.44268250465393066, 0.25649309158325195, -0.5621528625488281, -0.6077632904052734, 0.21485304832458496, -0.658627986907959, -0.9116294384002686, -0.294114351272583, 0.0452420711517334, 0.8542745113372803, 0.7148771286010742, 0.3244490623474121]]
# pylint: enable=line-too-long
def real_digit():
return tf.expand_dims(tf.expand_dims(REAL_DIGIT, 0), -1)
def fake_digit():
return tf.expand_dims(tf.expand_dims(FAKE_DIGIT, 0), -1)
def one_hot_real():
return tf.constant(ONE_HOT)
def one_hot1():
return tf.constant([[1.0] + [0.0] * 9])
class MnistScoreTest(tf.test.TestCase):
def test_any_batch_size(self):
inputs = tf.placeholder(tf.float32, shape=[None, 28, 28, 1])
mscore = util.mnist_score(inputs)
for batch_size in [4, 16, 30]:
with self.test_session() as sess:
sess.run(mscore, feed_dict={inputs: np.zeros([batch_size, 28, 28, 1])})
def test_deterministic(self):
m_score = util.mnist_score(real_digit())
with self.test_session():
m_score1 = m_score.eval()
m_score2 = m_score.eval()
self.assertEqual(m_score1, m_score2)
with self.test_session():
m_score3 = m_score.eval()
self.assertEqual(m_score1, m_score3)
def test_single_example_correct(self):
real_score = util.mnist_score(real_digit())
fake_score = util.mnist_score(fake_digit())
with self.test_session():
self.assertNear(1.0, real_score.eval(), 1e-6)
self.assertNear(1.0, fake_score.eval(), 1e-6)
def test_minibatch_correct(self):
mscore = util.mnist_score(
tf.concat([real_digit(), real_digit(), fake_digit()], 0))
with self.test_session():
self.assertNear(1.612828, mscore.eval(), 1e-6)
def test_batch_splitting_doesnt_change_value(self):
for num_batches in [1, 2, 4, 8]:
mscore = util.mnist_score(
tf.concat([real_digit()] * 4 + [fake_digit()] * 4, 0),
num_batches=num_batches)
with self.test_session():
self.assertNear(1.649209, mscore.eval(), 1e-6)
class MnistFrechetDistanceTest(tf.test.TestCase):
def test_any_batch_size(self):
inputs = tf.placeholder(tf.float32, shape=[None, 28, 28, 1])
fdistance = util.mnist_frechet_distance(inputs, inputs)
for batch_size in [4, 16, 30]:
with self.test_session() as sess:
sess.run(fdistance,
feed_dict={inputs: np.zeros([batch_size, 28, 28, 1])})
def test_deterministic(self):
fdistance = util.mnist_frechet_distance(
tf.concat([real_digit()] * 2, 0),
tf.concat([fake_digit()] * 2, 0))
with self.test_session():
fdistance1 = fdistance.eval()
fdistance2 = fdistance.eval()
self.assertNear(fdistance1, fdistance2, 2e-1)
with self.test_session():
fdistance3 = fdistance.eval()
self.assertNear(fdistance1, fdistance3, 2e-1)
def test_single_example_correct(self):
fdistance = util.mnist_frechet_distance(
tf.concat([real_digit()] * 2, 0),
tf.concat([real_digit()] * 2, 0))
with self.test_session():
self.assertNear(0.0, fdistance.eval(), 2e-1)
def test_minibatch_correct(self):
fdistance = util.mnist_frechet_distance(
tf.concat([real_digit(), real_digit(), fake_digit()], 0),
tf.concat([real_digit(), fake_digit(), fake_digit()], 0))
with self.test_session():
self.assertNear(43.5, fdistance.eval(), 2e-1)
def test_batch_splitting_doesnt_change_value(self):
for num_batches in [1, 2, 4, 8]:
fdistance = util.mnist_frechet_distance(
tf.concat([real_digit()] * 6 + [fake_digit()] * 2, 0),
tf.concat([real_digit()] * 2 + [fake_digit()] * 6, 0),
num_batches=num_batches)
with self.test_session():
self.assertNear(97.8, fdistance.eval(), 2e-1)
class MnistCrossEntropyTest(tf.test.TestCase):
def test_any_batch_size(self):
num_classes = 10
one_label = np.array([[1] + [0] * (num_classes - 1)])
inputs = tf.placeholder(tf.float32, shape=[None, 28, 28, 1])
one_hot_label = tf.placeholder(tf.int32, shape=[None, num_classes])
entropy = util.mnist_cross_entropy(inputs, one_hot_label)
for batch_size in [4, 16, 30]:
with self.test_session() as sess:
sess.run(entropy, feed_dict={
inputs: np.zeros([batch_size, 28, 28, 1]),
one_hot_label: np.concatenate([one_label] * batch_size)})
def test_deterministic(self):
xent = util.mnist_cross_entropy(real_digit(), one_hot_real())
with self.test_session():
ent1 = xent.eval()
ent2 = xent.eval()
self.assertEqual(ent1, ent2)
with self.test_session():
ent3 = xent.eval()
self.assertEqual(ent1, ent3)
def test_single_example_correct(self):
# The correct label should have low cross entropy.
correct_xent = util.mnist_cross_entropy(real_digit(), one_hot_real())
# The incorrect label should have high cross entropy.
wrong_xent = util.mnist_cross_entropy(real_digit(), one_hot1())
# A random digit should have medium cross entropy for any label.
fake_xent1 = util.mnist_cross_entropy(fake_digit(), one_hot_real())
fake_xent6 = util.mnist_cross_entropy(fake_digit(), one_hot1())
with self.test_session():
self.assertNear(0.00996, correct_xent.eval(), 1e-5)
self.assertNear(18.63073, wrong_xent.eval(), 1e-5)
self.assertNear(2.2, fake_xent1.eval(), 1e-1)
self.assertNear(2.2, fake_xent6.eval(), 1e-1)
def test_minibatch_correct(self):
# Reorded minibatches should have the same value.
xent1 = util.mnist_cross_entropy(
tf.concat([real_digit(), real_digit(), fake_digit()], 0),
tf.concat([one_hot_real(), one_hot1(), one_hot1()], 0))
xent2 = util.mnist_cross_entropy(
tf.concat([real_digit(), fake_digit(), real_digit()], 0),
tf.concat([one_hot_real(), one_hot1(), one_hot1()], 0))
with self.test_session():
self.assertNear(6.972539, xent1.eval(), 1e-5)
self.assertNear(xent1.eval(), xent2.eval(), 1e-5)
class GetNoiseTest(tf.test.TestCase):
def test_get_noise_categorical_syntax(self):
util.get_eval_noise_categorical(
noise_samples=4,
categorical_sample_points=np.arange(0, 10),
continuous_sample_points=np.linspace(-2.0, 2.0, 10),
unstructured_noise_dims=62,
continuous_noise_dims=2)
def test_get_noise_continuous_dim1_syntax(self):
util.get_eval_noise_continuous_dim1(
noise_samples=4,
categorical_sample_points=np.arange(0, 10),
continuous_sample_points=np.linspace(-2.0, 2.0, 10),
unstructured_noise_dims=62,
continuous_noise_dims=2)
def test_get_noise_continuous_dim2_syntax(self):
util.get_eval_noise_continuous_dim2(
noise_samples=4,
categorical_sample_points=np.arange(0, 10),
continuous_sample_points=np.linspace(-2.0, 2.0, 10),
unstructured_noise_dims=62,
continuous_noise_dims=2)
def test_get_infogan_syntax(self):
util.get_infogan_noise(
batch_size=4,
categorical_dim=10,
structured_continuous_dim=3,
total_continuous_noise_dims=62)
if __name__ == '__main__':
tf.test.main()
research/gan/mnist_estimator/launch_jobs.sh deleted 100755 → 0
View file @ f673f7a8
# Copyright 2016 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
#!/bin/bash
#
# This script performs the following operations:
# 1. Downloads the MNIST dataset.
# 2. Trains an unconditional model on the MNIST training set using a
# tf.Estimator.
# 3. Evaluates the models and writes sample images to disk.
#
#
# Usage:
# cd models/research/gan/mnist_estimator
# ./launch_jobs.sh ${git_repo}
set -e
# Location of the git repository.
git_repo=$1
if [[ "$git_repo" == "" ]]; then
echo "'git_repo' must not be empty."
exit
fi
# Base name for where the evaluation images will be saved to.
EVAL_DIR=/tmp/mnist-estimator
# Where the dataset is saved to.
DATASET_DIR=/tmp/mnist-data
export PYTHONPATH=$PYTHONPATH:$git_repo:$git_repo/research:$git_repo/research/gan:$git_repo/research/slim
# A helper function for printing pretty output.
Banner () {
local text=$1
local green='\033[0;32m'
local nc='\033[0m' # No color.
echo -e "${green}${text}${nc}"
}
# Download the dataset.
python "${git_repo}/research/slim/download_and_convert_data.py" \
--dataset_name=mnist \
--dataset_dir=${DATASET_DIR}
# Run unconditional GAN.
NUM_STEPS=1600
Banner "Starting training GANEstimator ${NUM_STEPS} steps..."
python "${git_repo}/research/gan/mnist_estimator/train.py" \
--max_number_of_steps=${NUM_STEPS} \
--eval_dir=${EVAL_DIR} \
--dataset_dir=${DATASET_DIR} \
--alsologtostderr
Banner "Finished training GANEstimator ${NUM_STEPS} steps. See ${EVAL_DIR} for output images."
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment