def load_dataset(config):
"""Load dataset following instruction in `config`."""
if dataset_is_mnist_family(config['dataset']):
crop_width = config.get('crop_width', None) # unused
img_width = config.get('img_width', None) # unused
scratch = config.get('scratch', get_default_scratch())
basepath = os.path.join(scratch, config['dataset'].lower())
data_path = os.path.join(basepath, 'data')
save_path = os.path.join(basepath, 'ckpts')
tf.gfile.MakeDirs(data_path)
tf.gfile.MakeDirs(save_path)
# black-on-white MNIST (harder to learn than white-on-black MNIST)
# Running locally (pre-download data locally)
mnist_train, mnist_eval, mnist_test = local_mnist.read_data_sets(
data_path, one_hot=True)
train_data = np.concatenate([mnist_train.images, mnist_eval.images],
axis=0)
attr_train = np.concatenate([mnist_train.labels, mnist_eval.labels],
axis=0)
eval_data = mnist_test.images
attr_eval = mnist_test.labels
attribute_names = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
elif config['dataset'] == 'CELEBA':
crop_width = config['crop_width']
img_width = config['img_width']
postfix = '_crop_%d_res_%d.npy' % (crop_width, img_width)
# Load Data
scratch = config.get('scratch', get_default_scratch())
basepath = os.path.join(scratch, 'celeba')
data_path = os.path.join(basepath, 'data')
save_path = os.path.join(basepath, 'ckpts')
(train_data, eval_data, _, attr_train, attr_eval, _,
attribute_names) = _load_celeba(data_path, postfix)
else:
raise NotImplementedError
return ObjectBlob(
crop_width=crop_width,
img_width=img_width,
basepath=basepath,
data_path=data_path,
save_path=save_path,
train_data=train_data,
attr_train=attr_train,
eval_data=eval_data,
attr_eval=attr_eval,
attribute_names=attribute_names,
)
def run_training():
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)
with tf.Graph().as_default():
x = tf.placeholder(tf.float32, shape=[None, 784])
y_ = tf.placeholder(tf.float32, shape=[None, 10])
keep_prob = tf.placeholder(tf.float32)
y_conv = inference(x, keep_prob)
cross_entropy = loss(y_conv, y_)
train_step = training(cross_entropy)
accuracy = validation(y_conv, y_)
# Build the summary Tensor based on the TF collection of Summaries.
summary = tf.summary.merge_all()
sess = tf.InteractiveSession()
summary_writer = tf.summary.FileWriter(FLAGS.train_dir, sess.graph)
sess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
training_util.write_graph(sess.graph.as_graph_def(add_shapes=True),
FLAGS.train_dir, "graph.pbtxt")
for step in xrange(2000):
batch = mnist.train.next_batch(50)
feed_dict = {x: batch[0], y_: batch[1], keep_prob: 0.5}
_, loss_value = sess.run([train_step, cross_entropy],
feed_dict=feed_dict)
if step % 500 == 0:
saver.save(sess, os.path.join(FLAGS.train_dir, "model.ckpt"),
step)
train_accuracy = accuracy.eval(feed_dict=feed_dict)
print("step %d, loss_value %.2f,training accuracy %g" %
(step, loss_value, train_accuracy))
# Update the events file.
summary_str = sess.run(summary, feed_dict=feed_dict)
summary_writer.add_summary(summary_str, step)
summary_writer.flush()
# Test
print("test accuracy %g" % accuracy.eval(feed_dict={
x: mnist.test.images,
y_: mnist.test.labels,
keep_prob: 1.0
}))
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)
keep_prob = tf.placeholder(tf.float32)
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])
y_conv = tf.matmul(h_fc1_drop, W_fc2) + b_fc2
# train and evaluate
cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=y_conv, labels=y_))
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
correct_prediction = tf.equal(tf.argmax(y_conv,1), tf.argmax(y_,1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)
sess = tf.InteractiveSession()
sess.run(tf.global_variables_initializer())
for i in range(20000):
batch = mnist.train.next_batch(50)
if i%100 == 0:
train_accuracy = accuracy.eval(feed_dict={
x:batch[0], y_: batch[1], keep_prob: 1.0})
print("step %d, training accuracy %g"%(i, train_accuracy))
train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
print("test accuracy %g"%accuracy.eval(feed_dict={
x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0}))