TensorFlow 1.13.2
更新时间:2023-01-18
TensorFlow
基于tensorflow框架的MNIST图像分类任务示例代码,训练数据集点击这里下载
单机训练(计算节点数为1),示例代码如下:
import os
import tensorflow as tf
import numpy as np
from tensorflow import keras
layers = tf.layers
tf.logging.set_verbosity(tf.logging.INFO)
def conv_model(feature, target, mode):
"""2-layer convolution model."""
# Convert the target to a one-hot tensor of shape (batch_size, 10) and
# with a on-value of 1 for each one-hot vector of length 10.
target = tf.one_hot(tf.cast(target, tf.int32), 10, 1, 0)
# Reshape feature to 4d tensor with 2nd and 3rd dimensions being
# image width and height final dimension being the number of color channels.
feature = tf.reshape(feature, [-1, 28, 28, 1])
# First conv layer will compute 32 features for each 5x5 patch
with tf.variable_scope('conv_layer1'):
h_conv1 = layers.conv2d(feature, 32, kernel_size=[5, 5],
activation=tf.nn.relu, padding="SAME")
h_pool1 = tf.nn.max_pool(
h_conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
# Second conv layer will compute 64 features for each 5x5 patch.
with tf.variable_scope('conv_layer2'):
h_conv2 = layers.conv2d(h_pool1, 64, kernel_size=[5, 5],
activation=tf.nn.relu, padding="SAME")
h_pool2 = tf.nn.max_pool(
h_conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
# reshape tensor into a batch of vectors
h_pool2_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 64])
# Densely connected layer with 1024 neurons.
h_fc1 = layers.dropout(
layers.dense(h_pool2_flat, 1024, activation=tf.nn.relu),
rate=0.5, training=mode == tf.estimator.ModeKeys.TRAIN)
# Compute logits (1 per class) and compute loss.
logits = layers.dense(h_fc1, 10, activation=None)
loss = tf.losses.softmax_cross_entropy(target, logits)
return tf.argmax(logits, 1), loss
def train_input_generator(x_train, y_train, batch_size=64):
assert len(x_train) == len(y_train)
while True:
p = np.random.permutation(len(x_train))
x_train, y_train = x_train[p], y_train[p]
index = 0
while index <= len(x_train) - batch_size:
yield x_train[index:index + batch_size], \
y_train[index:index + batch_size],
index += batch_size
def main(_):
work_path = os.getcwd()
# Download and load MNIST dataset.
(x_train, y_train), (x_test, y_test) = \
keras.datasets.mnist.load_data('%s/train_data/mnist.npz' % work_path)
# The shape of downloaded data is (-1, 28, 28), hence we need to reshape it
# into (-1, 784) to feed into our network. Also, need to normalize the
# features between 0 and 1.
x_train = np.reshape(x_train, (-1, 784)) / 255.0
x_test = np.reshape(x_test, (-1, 784)) / 255.0
# Build model...
with tf.name_scope('input'):
image = tf.placeholder(tf.float32, [None, 784], name='image')
label = tf.placeholder(tf.float32, [None], name='label')
predict, loss = conv_model(image, label, tf.estimator.ModeKeys.TRAIN)
opt = tf.train.RMSPropOptimizer(0.001)
global_step = tf.train.get_or_create_global_step()
train_op = opt.minimize(loss, global_step=global_step)
hooks = [
tf.train.StopAtStepHook(last_step=20000),
tf.train.LoggingTensorHook(tensors={'step': global_step, 'loss': loss},
every_n_iter=10),
]
# Horovod: pin GPU to be used to process local rank (one GPU per process)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
config.gpu_options.visible_device_list = '0'
# Horovod: save checkpoints only on worker 0 to prevent other workers from
# corrupting them.
checkpoint_dir = './checkpoints'
training_batch_generator = train_input_generator(x_train,
y_train, batch_size=100)
# The MonitoredTrainingSession takes care of session initialization,
# restoring from a checkpoint, saving to a checkpoint, and closing when done
# or an error occurs.
with tf.train.MonitoredTrainingSession(checkpoint_dir=checkpoint_dir,
hooks=hooks,
config=config) as mon_sess:
while not mon_sess.should_stop():
# Run a training step synchronously.
image_, label_ = next(training_batch_generator)
mon_sess.run(train_op, feed_dict={image: image_, label: label_})
checkpoint_file = tf.train.latest_checkpoint(checkpoint_dir)
saver = tf.train.Saver()
inputs_classes = tf.saved_model.utils.build_tensor_info(image)
outputs_classes = tf.saved_model.utils.build_tensor_info(predict)
signature = (tf.saved_model.signature_def_utils.build_signature_def(
inputs={tf.saved_model.signature_constants.CLASSIFY_INPUTS: inputs_classes},
outputs={tf.saved_model.signature_constants.CLASSIFY_OUTPUT_CLASSES: outputs_classes},
method_name=tf.saved_model.signature_constants.CLASSIFY_METHOD_NAME))
os.system("rm -rf ./output")
with tf.Session() as sess:
sess.run([tf.local_variables_initializer(), tf.tables_initializer()])
saver.restore(sess, checkpoint_file)
builder = tf.saved_model.builder.SavedModelBuilder('./output')
legacy_init_op = tf.group(tf.tables_initializer(), name='legacy_init_op')
builder.add_meta_graph_and_variables(sess, [tf.saved_model.tag_constants.SERVING], signature_def_map={'predict_images': signature}, legacy_init_op=legacy_init_op)
builder.save()
if __name__ == "__main__":
tf.app.run()
分布式训练(计算节点数大于1),示例代码如下:
说明:demo分布式程序没有做数据的分片操作,仅供参考
import os
import tensorflow as tf
import horovod.tensorflow as hvd
import numpy as np
from tensorflow import keras
layers = tf.layers
tf.logging.set_verbosity(tf.logging.INFO)
def conv_model(feature, target, mode):
"""2-layer convolution model."""
# Convert the target to a one-hot tensor of shape (batch_size, 10) and
# with a on-value of 1 for each one-hot vector of length 10.
target = tf.one_hot(tf.cast(target, tf.int32), 10, 1, 0)
# Reshape feature to 4d tensor with 2nd and 3rd dimensions being
# image width and height final dimension being the number of color channels.
feature = tf.reshape(feature, [-1, 28, 28, 1])
# First conv layer will compute 32 features for each 5x5 patch
with tf.variable_scope('conv_layer1'):
h_conv1 = layers.conv2d(feature, 32, kernel_size=[5, 5],
activation=tf.nn.relu, padding="SAME")
h_pool1 = tf.nn.max_pool(
h_conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
# Second conv layer will compute 64 features for each 5x5 patch.
with tf.variable_scope('conv_layer2'):
h_conv2 = layers.conv2d(h_pool1, 64, kernel_size=[5, 5],
activation=tf.nn.relu, padding="SAME")
h_pool2 = tf.nn.max_pool(
h_conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
# reshape tensor into a batch of vectors
h_pool2_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 64])
# Densely connected layer with 1024 neurons.
h_fc1 = layers.dropout(
layers.dense(h_pool2_flat, 1024, activation=tf.nn.relu),
rate=0.5, training=mode == tf.estimator.ModeKeys.TRAIN)
# Compute logits (1 per class) and compute loss.
logits = layers.dense(h_fc1, 10, activation=None)
loss = tf.losses.softmax_cross_entropy(target, logits)
return tf.argmax(logits, 1), loss
def train_input_generator(x_train, y_train, batch_size=64):
assert len(x_train) == len(y_train)
while True:
p = np.random.permutation(len(x_train))
x_train, y_train = x_train[p], y_train[p]
index = 0
while index <= len(x_train) - batch_size:
yield x_train[index:index + batch_size], \
y_train[index:index + batch_size],
index += batch_size
def main(_):
# Horovod: initialize Horovod.
hvd.init()
work_path = os.getcwd()
# Download and load MNIST dataset.
(x_train, y_train), (x_test, y_test) = \
keras.datasets.mnist.load_data('%s/train_data/mnist.npz' % work_path)
# The shape of downloaded data is (-1, 28, 28), hence we need to reshape it
# into (-1, 784) to feed into our network. Also, need to normalize the
# features between 0 and 1.
x_train = np.reshape(x_train, (-1, 784)) / 255.0
x_test = np.reshape(x_test, (-1, 784)) / 255.0
# Build model...
with tf.name_scope('input'):
image = tf.placeholder(tf.float32, [None, 784], name='image')
label = tf.placeholder(tf.float32, [None], name='label')
predict, loss = conv_model(image, label, tf.estimator.ModeKeys.TRAIN)
serve_graph_file = "./serve_graph.meta"
tf.train.export_meta_graph(serve_graph_file, as_text=True)
# Horovod: adjust learning rate based on number of GPUs.
opt = tf.train.RMSPropOptimizer(0.001 * hvd.size())
# Horovod: add Horovod Distributed Optimizer.
opt = hvd.DistributedOptimizer(opt)
global_step = tf.train.get_or_create_global_step()
train_op = opt.minimize(loss, global_step=global_step)
hooks = [
# Horovod: BroadcastGlobalVariablesHook broadcasts initial variable states
# from rank 0 to all other processes. This is necessary to ensure consistent
# initialization of all workers when training is started with random weights
# or restored from a checkpoint.
hvd.BroadcastGlobalVariablesHook(0),
# Horovod: adjust number of steps based on number of GPUs.
tf.train.StopAtStepHook(last_step=10000 // hvd.size()),
tf.train.LoggingTensorHook(tensors={'step': global_step, 'loss': loss},
every_n_iter=10),
]
# Horovod: pin GPU to be used to process local rank (one GPU per process)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
config.gpu_options.visible_device_list = str(hvd.local_rank())
# Horovod: save checkpoints only on worker 0 to prevent other workers from
# corrupting them.
checkpoint_dir = './checkpoints' if hvd.rank() == 0 else None
training_batch_generator = train_input_generator(x_train,
y_train, batch_size=100)
# The MonitoredTrainingSession takes care of session initialization,
# restoring from a checkpoint, saving to a checkpoint, and closing when done
# or an error occurs.
with tf.train.MonitoredTrainingSession(checkpoint_dir=checkpoint_dir,
hooks=hooks,
config=config) as mon_sess:
while not mon_sess.should_stop():
# Run a training step synchronously.
image_, label_ = next(training_batch_generator)
mon_sess.run(train_op, feed_dict={image: image_, label: label_})
if hvd.rank() != 0:
return
checkpoint_file = tf.train.latest_checkpoint(checkpoint_dir)
tf.reset_default_graph()
saver = tf.train.import_meta_graph(serve_graph_file)
inputs_classes = tf.saved_model.utils.build_tensor_info(image)
outputs_classes = tf.saved_model.utils.build_tensor_info(predict)
signature = (tf.saved_model.signature_def_utils.build_signature_def(
inputs={tf.saved_model.signature_constants.CLASSIFY_INPUTS: inputs_classes},
outputs={tf.saved_model.signature_constants.CLASSIFY_OUTPUT_CLASSES: outputs_classes},
method_name=tf.saved_model.signature_constants.CLASSIFY_METHOD_NAME))
os.system("rm -rf ./output")
with tf.Session() as sess:
sess.run([tf.local_variables_initializer(), tf.tables_initializer()])
saver.restore(sess, checkpoint_file)
builder = tf.saved_model.builder.SavedModelBuilder('./output')
legacy_init_op = tf.group(tf.tables_initializer(), name='legacy_init_op')
builder.add_meta_graph_and_variables(sess, [tf.saved_model.tag_constants.SERVING], signature_def_map={'predict_images': signature}, legacy_init_op=legacy_init_op)
builder.save()
if __name__ == "__main__":
tf.app.run()