def test_tf_optimizer_metrics(self):
features = np.random.randn(20, 10)
labels = np.random.randint(0, 10, size=[20])
with tf.Graph().as_default():
dataset = TFDataset.from_ndarrays((features, labels),
batch_size=4,
val_tensors=(features, labels))
feature_tensor, label_tensor = dataset.tensors
features = tf.layers.dense(feature_tensor, 8)
output = tf.layers.dense(features, 10)
loss = tf.reduce_mean(tf.losses.
sparse_softmax_cross_entropy(logits=output,
labels=label_tensor))
optimizer = TFOptimizer.from_loss(loss, {"dense/": Adam(1e-3), "dense_1/": SGD(0.0)},
val_outputs=[output],
val_labels=[label_tensor],
val_method=Accuracy(), metrics={"loss": loss})
initial_weights = optimizer.tf_model.training_helper_layer.get_weights()
optimizer.optimize(end_trigger=MaxEpoch(1))
updated_weights = optimizer.tf_model.training_helper_layer.get_weights()
for i in [0, 1]: # weights and bias combined with "dense/" should be updated
assert not np.allclose(initial_weights[i], updated_weights[i])
for i in [2, 3]: # weights and bias combined with "dense_1" should be unchanged
assert np.allclose(initial_weights[i], updated_weights[i])
optimizer.sess.close()
def test_control_inputs(self):
features = np.random.randn(20, 10)
labels = np.random.randint(0, 10, size=[20])
with tf.Graph().as_default():
dataset = TFDataset.from_ndarrays((features, labels),
batch_size=4,
val_tensors=(features, labels))
is_training = tf.placeholder(dtype=tf.bool, shape=())
feature_tensor, label_tensor = dataset.tensors
features = tf.layers.dense(feature_tensor, 8)
features = tf.layers.dropout(features, training=is_training)
output = tf.layers.dense(features, 10)
loss = tf.reduce_mean(
tf.losses.sparse_softmax_cross_entropy(logits=output,
labels=label_tensor))
optimizer = TFOptimizer.from_loss(
loss,
Adam(),
val_outputs=[output],
val_labels=[label_tensor],
val_method=Accuracy(),
tensor_with_value={is_training: (True, False)},
metrics={"loss": loss})
optimizer.optimize(end_trigger=MaxEpoch(1))
optimizer.sess.close()
def main():
sc = init_nncontext()
global_batch_size = 256
loss = create_model(creat_dataset(global_batch_size))
optimizer = TFOptimizer.from_loss(loss, SGD(1e-3), model_dir="/tmp/lenet/")
optimizer.optimize(end_trigger=MaxIteration(20))
def test_tfdataset_with_tfrecord(self):
train_path = os.path.join(resource_path,
"tfrecord/mnist_train.tfrecord")
test_path = os.path.join(resource_path, "tfrecord/mnist_test.tfrecord")
dataset = TFDataset.from_tfrecord_file(self.sc,
train_path,
batch_size=16,
validation_file_path=test_path)
raw_bytes = dataset.tensors[0]
images, labels = parse_fn(raw_bytes)
flat = tf.layers.flatten(images)
logits = tf.layers.dense(flat, 10)
loss = tf.reduce_mean(
tf.losses.sparse_softmax_cross_entropy(logits=logits,
labels=labels))
opt = TFOptimizer.from_loss(loss, Adam())
opt.optimize()
def main(max_epoch, data_num):
sc = init_nncontext()
# get data, pre-process and create TFDataset
def get_data_rdd(dataset):
(images_data,
labels_data) = mnist.read_data_sets("/tmp/mnist", dataset)
image_rdd = sc.parallelize(images_data[:data_num])
labels_rdd = sc.parallelize(labels_data[:data_num])
rdd = image_rdd.zip(labels_rdd) \
.map(lambda rec_tuple: [normalizer(rec_tuple[0], mnist.TRAIN_MEAN, mnist.TRAIN_STD),
np.array(rec_tuple[1])])
return rdd
training_rdd = get_data_rdd("train")
testing_rdd = get_data_rdd("test")
dataset = TFDataset.from_rdd(training_rdd,
names=["features", "labels"],
shapes=[[28, 28, 1], []],
types=[tf.float32, tf.int32],
batch_size=280,
val_rdd=testing_rdd)
# construct the model from TFDataset
images, labels = dataset.tensors
with slim.arg_scope(lenet.lenet_arg_scope()):
logits, end_points = lenet.lenet(images,
num_classes=10,
is_training=True)
loss = tf.reduce_mean(
tf.losses.sparse_softmax_cross_entropy(logits=logits, labels=labels))
# create a optimizer
optimizer = TFOptimizer.from_loss(loss,
Adam(1e-3),
val_outputs=[logits],
val_labels=[labels],
val_method=Top1Accuracy(),
model_dir="/tmp/lenet/")
# kick off training
optimizer.optimize(end_trigger=MaxEpoch(max_epoch))
saver = tf.train.Saver()
saver.save(optimizer.sess, "/tmp/lenet/model")
def test_checkpoint(self):
features = np.random.randn(20, 10)
labels = np.random.randint(0, 10, size=[20])
with tf.Graph().as_default():
dataset = TFDataset.from_ndarrays((features, labels),
batch_size=4,
val_tensors=(features, labels))
feature_tensor, label_tensor = dataset.tensors
features = tf.layers.dense(feature_tensor, 8)
output = tf.layers.dense(features, 10)
loss = tf.reduce_mean(tf.losses.
sparse_softmax_cross_entropy(logits=output,
labels=label_tensor))
model_dir = tempfile.mkdtemp()
try:
optimizer = TFOptimizer.from_loss(loss, Adam(),
val_outputs=[output],
val_labels=[label_tensor],
val_method=Accuracy(),
metrics={"loss": loss}, model_dir=model_dir)
optimizer.optimize(end_trigger=MaxEpoch(1))
import re
ckpt_path = None
versions = []
for (root, dirs, files) in os.walk(model_dir, topdown=True):
temp_versions = []
for file_name in files:
if re.match("^optimMethod-TFParkTraining\.[0-9]+$", file_name) is not None:
version = int(file_name.split(".")[1])
temp_versions.append(version)
if temp_versions:
ckpt_path = root
versions = temp_versions
break
assert ckpt_path is not None, "Cannot fine checkpoint file"
optimizer.load_checkpoint(ckpt_path, max(versions))
optimizer.optimize(end_trigger=MaxEpoch(1))
optimizer.sess.close()
finally:
import shutil
shutil.rmtree(model_dir)
def test_tfdataset_with_tf_data_dataset_which_contains_bool(self):
dataset = tf.data.Dataset.from_tensor_slices(
(np.random.randn(102, 28, 28,
1), np.random.randint(0, 10, size=(102, )),
np.ones(shape=(102, 28, 28, 1), dtype=np.bool)))
dataset = TFDataset.from_tf_data_dataset(dataset, batch_size=16)
feature, labels, mask = dataset.tensors
float_mask = tf.to_float(mask)
masked_feature = tf.to_float(feature) * float_mask
flatten = tf.layers.flatten(masked_feature)
logits = tf.layers.dense(flatten, 10)
loss = tf.reduce_mean(
tf.losses.sparse_softmax_cross_entropy(logits=logits,
labels=labels))
opt = TFOptimizer.from_loss(loss, Adam())
opt.optimize()
def main(max_epoch, data_num):
sc = init_nncontext()
# get data, pre-process and create TFDataset
(train_images_data,
train_labels_data) = mnist.read_data_sets("/tmp/mnist", "train")
(test_images_data,
test_labels_data) = mnist.read_data_sets("/tmp/mnist", "train")
train_images_data = (train_images_data[:data_num] -
mnist.TRAIN_MEAN) / mnist.TRAIN_STD
train_labels_data = train_labels_data[:data_num].astype(np.int)
test_images_data = (test_images_data[:data_num] -
mnist.TRAIN_MEAN) / mnist.TRAIN_STD
test_labels_data = (test_labels_data[:data_num]).astype(np.int)
dataset = TFDataset.from_ndarrays(
(train_images_data, train_labels_data),
batch_size=360,
val_tensors=(test_images_data, test_labels_data))
# construct the model from TFDataset
images, labels = dataset.tensors
with slim.arg_scope(lenet.lenet_arg_scope()):
logits, end_points = lenet.lenet(images,
num_classes=10,
is_training=True)
loss = tf.reduce_mean(
tf.losses.sparse_softmax_cross_entropy(logits=logits, labels=labels))
acc = accuracy(logits, labels)
# create a optimizer
optimizer = TFOptimizer.from_loss(loss,
Adam(1e-3),
metrics={"acc": acc},
model_dir="/tmp/lenet/")
# kick off training
optimizer.optimize(end_trigger=MaxEpoch(max_epoch))
saver = tf.train.Saver()
saver.save(optimizer.sess, "/tmp/lenet/model")
def test_tf_optimizer_with_sparse_gradient(self):
ids = np.random.randint(0, 10, size=[40])
labels = np.random.randint(0, 5, size=[40])
id_rdd = self.sc.parallelize(ids)
label_rdd = self.sc.parallelize(labels)
training_rdd = id_rdd.zip(label_rdd).map(lambda x: [x[0], x[1]])
with tf.Graph().as_default():
dataset = TFDataset.from_rdd(training_rdd,
names=["ids", "labels"],
shapes=[[], []],
types=[tf.int32, tf.int32],
batch_size=8)
id_tensor, label_tensor = dataset.tensors
embedding_table = tf.get_variable(name="word_embedding",
shape=[10, 5])
embedding = tf.nn.embedding_lookup(embedding_table, id_tensor)
loss = tf.reduce_mean(
tf.losses.sparse_softmax_cross_entropy(logits=embedding,
labels=label_tensor))
optimizer = TFOptimizer.from_loss(loss, Adam(1e-3))
optimizer.optimize(end_trigger=MaxEpoch(1))
optimizer.sess.close()
def fit(self,
data,
epochs=1,
batch_size=32,
feature_cols=None,
label_cols=None,
validation_data=None,
session_config=None,
checkpoint_trigger=None,
auto_shard_files=False,
feed_dict=None):
"""
Train this graph model with train data.
:param data: train data. It can be XShards, Spark DataFrame, tf.data.Dataset.
If data is XShards, each partition can be Pandas Dataframe or a dictionary of
{'x': feature, 'y': label}, where feature(label) is a numpy array or a tuple of
numpy arrays.
If data is tf.data.Dataset, each element is a tuple of input tensors.
:param epochs: number of epochs to train.
:param batch_size: total batch size for each iteration.
:param feature_cols: feature column names if train data is Spark DataFrame or XShards
of Pandas Dataframe.
:param label_cols: label column names if train data is Spark DataFrame or XShards of
Pandas Dataframe.
:param validation_data: validation data. Validation data type should be the same
as train data.
:param auto_shard_files: whether to automatically detect if the dataset is file-based and
and apply sharding on files, otherwise sharding on records. Default is False.
:param session_config: tensorflow session configuration for training.
Should be object of tf.ConfigProto
:param feed_dict: a dictionary. The key is TensorFlow tensor, usually a
placeholder, the value of the dictionary is a tuple of two elements. The first one of
the tuple is the value to feed to the tensor in training phase and the second one
is the value to feed to the tensor in validation phase.
:param checkpoint_trigger: when to trigger checkpoint during training.
Should be a zoo.orca.learn.trigger, like EveryEpoch(), SeveralIteration(
num_iterations),etc.
"""
assert self.labels is not None, \
"labels is None; it should not be None in training"
assert self.loss is not None, \
"loss is None; it should not be None in training"
assert self.optimizer is not None, \
"optimizer is None; it should not be None in training"
if isinstance(data, DataFrame):
assert feature_cols is not None, \
"feature columns is None; it should not be None in training"
assert label_cols is not None, \
"label columns is None; it should not be None in training"
if isinstance(data, SparkXShards):
if data._get_class_name() == 'pandas.core.frame.DataFrame':
assert feature_cols is not None, \
"feature columns is None; it should not be None in training"
assert label_cols is not None, \
"label columns is None; it should not be None in training"
data, validation_data = process_xshards_of_pandas_dataframe(
data, feature_cols, label_cols, validation_data, "fit")
if checkpoint_trigger is not None:
checkpoint_trigger = Trigger.convert_trigger(checkpoint_trigger)
memory_type = OrcaContext.train_data_store
dataset = to_dataset(data,
batch_size=batch_size,
batch_per_thread=-1,
validation_data=validation_data,
feature_cols=feature_cols,
label_cols=label_cols,
hard_code_batch_size=False,
sequential_order=False,
shuffle=True,
auto_shard_files=auto_shard_files,
memory_type=memory_type)
if feed_dict is not None:
tensor_with_value = {
key: (value[0], value[1])
for key, value in feed_dict.items()
}
else:
tensor_with_value = None
if self.use_bigdl_optim:
self.tf_optimizer = TFOptimizer.from_loss(
self.loss,
self.optimizer,
session=self.sess,
inputs=(self.inputs, self.labels),
dataset=dataset,
clip_norm=self.clip_norm,
clip_value=self.clip_value,
metrics=self.metrics,
tensor_with_value=tensor_with_value,
session_config=session_config,
model_dir=self.model_dir,
updates=self.updates)
else:
self.tf_optimizer = TFOptimizer.from_train_op(
train_op=self.train_op,
loss=self.loss,
inputs=self.inputs,
labels=self.labels,
dataset=dataset,
metrics=self.metrics,
updates=self.updates,
sess=self.sess,
tensor_with_value=tensor_with_value,
session_config=session_config,
model_dir=self.model_dir)
if self.load_checkpoint:
self.tf_optimizer.load_checkpoint(self.checkpoint_path,
self.checkpoint_version)
if self.log_dir and self.app_name:
self.tf_optimizer.estimator.set_tensorboard(
self.log_dir, self.app_name)
self.tf_optimizer.optimize(end_trigger=MaxEpoch(epochs),
checkpoint_trigger=checkpoint_trigger)
return self
lrSchedule.add(Poly(0.5, maxIteration), polyIteration)
optim = SGD(learningrate=options.learningRate, learningrate_decay=0.0,
weightdecay=options.weightDecay, momentum=0.9, dampening=0.0,
nesterov=False,
leaningrate_schedule=lrSchedule)
if options.maxEpoch:
checkpoint_trigger = EveryEpoch()
end_trigger = MaxEpoch(options.maxEpoch)
else:
checkpoint_trigger = SeveralIteration(options.checkpointIteration)
end_trigger = MaxIteration(options.maxIteration)
optimizer = TFOptimizer.from_loss(loss, optim,
val_outputs=[logits],
val_labels=[zero_based_label],
val_method=[Accuracy(), Top5Accuracy(), Loss()],
tensor_with_value={is_training: [True, False]},
model_dir="/tmp/logs")
if options.resumeTrainingCheckpoint is not None:
assert options.resumeTrainingVersion is not None,\
"--resumeTrainingVersion must be specified when --resumeTrainingCheckpoint is."
optimizer.load_checkpoint(options.resumeTrainingCheckpoint,
options.resumeTrainingVersion)
optimizer.optimize(end_trigger=end_trigger, checkpoint_trigger=checkpoint_trigger)
if options.checkpoint:
saver = tf.train.Saver()
saver.save(optimizer.sess, options.checkpoint)
