def cifar_resnet():
dev = 0
cntk_dev = cntk_device(dev)
epoch_size = sys.maxsize
mbs = create_mb_source(epoch_size)
stream_infos = mbs.stream_infos()
for si in stream_infos:
if si.m_name == 'features':
features_si = si
elif si.m_name == 'labels':
labels_si = si
image_shape = features_si.m_sample_layout.dimensions()
image_shape = (image_shape[2], image_shape[0], image_shape[1])
num_classes = labels_si.m_sample_layout.dimensions()[0]
image_input = variable(image_shape,
features_si.m_element_type,
needs_gradient=False,
name="Images")
classifier_output = resnet_classifer(image_input, num_classes, dev,
"classifierOutput")
label_var = variable((num_classes),
features_si.m_element_type,
needs_gradient=False,
name="Labels")
ce = cross_entropy_with_softmax(classifier_output, label_var)
pe = classification_error(classifier_output, label_var)
#TODO: add save and load module code
image_classifier = combine([ce, pe, classifier_output], "ImageClassifier")
lr = learning_rates_per_sample(0.0078125)
mb_size = 32
num_mbs = 1000
trainer = Trainer(classifier_output, ce, pe,
[sgdlearner(classifier_output.owner.parameters(), lr)])
for i in range(0, num_mbs):
mb = mbs.get_next_minibatch(mb_size, cntk_dev)
arguments = dict()
arguments[image_input] = mb[features_si].m_data
arguments[label_var] = mb[labels_si].m_data
trainer.train_minibatch(arguments, cntk_dev)
freq = 20
if i % freq == 0:
training_loss = get_train_loss(trainer)
eval_crit = get_train_eval_criterion(trainer)
print(
"Minibatch: {}, Train Loss: {}, Train Evaluation Criterion: {}"
.format(i, training_loss, eval_crit))
def _test_cifar_resnet():
dev = 0
cntk_dev = cntk_device(dev)
epoch_size = sys.maxsize
mbs = create_mb_source(epoch_size)
stream_infos = mbs.stream_infos()
for si in stream_infos:
if si.m_name == 'features':
features_si = si
elif si.m_name == 'labels':
labels_si = si
image_shape = features_si.m_sample_layout.dimensions()
image_shape = (image_shape[2], image_shape[0], image_shape[1])
num_classes = labels_si.m_sample_layout.dimensions()[0]
image_input = variable(image_shape,
features_si.m_element_type,
needs_gradient=False,
name="Images")
classifer_output = resnet_classifer(image_input, num_classes, dev,
"classifierOutput")
label_var = variable((num_classes, ),
features_si.m_element_type,
needs_gradient=False,
name="Labels")
ce = cross_entropy_with_softmax(classifer_output.output(), label_var)
pe = classification_error(classifer_output.output(), label_var)
image_classifier = combine([ce, pe, classifer_output], "ImageClassifier")
learning_rate_per_sample = cntk_py.learning_rates_per_sample(0.0078125)
trainer = cntk_py.Trainer(image_classifier, ce.output(), [
cntk_py.sgdlearner(image_classifier.parameters(),
learning_rate_per_sample)
])
mb_size = 32
num_mbs = 100
minibatch_size_limits = dict()
minibatch_size_limits[features_si] = (0, mb_size)
minibatch_size_limits[labels_si] = (0, mb_size)
for i in range(0, num_mbs):
mb = mbs.get_next_minibatch(minibatch_size_limits, cntk_dev)
arguments = dict()
arguments[image_input] = mb[features_si].m_data
arguments[label_var] = mb[labels_si].m_data
trainer.train_minibatch(arguments, cntk_dev)
freq = 20
if i % freq == 0:
print(str(i + freq) + ": " + str(get_train_loss(trainer)))
def test_simple_mnist():
input_dim = 784
num_output_classes = 10
num_hidden_layers = 1
hidden_layers_dim = 200
epoch_size = sys.maxsize
minibatch_size = 32
num_samples_per_sweep = 60000
num_sweeps_to_train_with = 3
num_minibatches_to_train = (num_samples_per_sweep * num_sweeps_to_train_with) / minibatch_size
lr = cntk_py.learning_rates_per_sample(0.003125)
input = variable((input_dim,), np.float32, needs_gradient=False, name="features")
scaled_input = element_times(constant((), 0.00390625), input)
label = variable((num_output_classes,), np.float32, needs_gradient=False, name="labels")
dev = cntk_py.DeviceDescriptor.cpudevice()
netout = fully_connected_classifier_net(scaled_input.output(), num_output_classes, hidden_layers_dim, num_hidden_layers, dev, sigmoid)
ce = cross_entropy_with_softmax(netout.output(), label)
pe = classification_error(netout.output(), label)
ffnet = combine([ce, pe, netout], "classifier_model")
cm = create_mb_source(input_dim, num_output_classes, epoch_size)
stream_infos = cm.stream_infos()
for si in stream_infos:
if si.m_name == 'features':
features_si = si
elif si.m_name == 'labels':
labels_si = si
minibatch_size_limits = dict()
minibatch_size_limits[features_si] = (0,minibatch_size)
minibatch_size_limits[labels_si] = (0,minibatch_size)
trainer = cntk_py.Trainer(ffnet, ce.output(), [cntk_py.sgdlearner(ffnet.parameters(), lr)])
for i in range(0,int(num_minibatches_to_train)):
mb=cm.get_next_minibatch(minibatch_size_limits, dev)
arguments = dict()
arguments[input] = mb[features_si].m_data
arguments[label] = mb[labels_si].m_data
trainer.train_minibatch(arguments, dev)
freq = 20
if i % freq == 0:
training_loss = get_train_loss(trainer)
print(str(i+freq) + ": " + str(training_loss))
#TODO: move the testing code into a separate test module ?
assert np.allclose(training_loss, 0.6142425537109375, atol=TOLERANCE_ABSOLUTE)
def ffnet():
input_dim = 2
num_output_classes = 2
num_hidden_layers = 2
hidden_layers_dim = 50
epoch_size = sys.maxsize
minibatch_size = 25
num_samples_per_sweep = 10000
num_sweeps_to_train_with = 2
num_minibatches_to_train = (num_samples_per_sweep * num_sweeps_to_train_with) / minibatch_size
lr = learning_rates_per_sample(0.02)
input = variable((input_dim,), np.float32, needs_gradient=False, name="features")
label = variable((num_output_classes,), np.float32, needs_gradient=False, name="labels")
dev = -1
cntk_dev = cntk_device(dev)
netout = fully_connected_classifier_net(input, num_output_classes, hidden_layers_dim, num_hidden_layers, dev, sigmoid)
ce = cross_entropy_with_softmax(netout, label)
pe = classification_error(netout, label)
#TODO: add save and load module code
ffnet = combine([ce, pe, netout], "classifier_model")
rel_path = r"../../../../Examples/Other/Simple2d/Data/SimpleDataTrain_cntk_text.txt"
path = os.path.join(os.path.dirname(os.path.abspath(__file__)), rel_path)
cm = create_text_mb_source(path, input_dim, num_output_classes, epoch_size)
stream_infos = cm.stream_infos()
for si in stream_infos:
if si.m_name == 'features':
features_si = si
elif si.m_name == 'labels':
labels_si = si
trainer = Trainer(netout, ce, pe, [sgdlearner(netout.owner.parameters(), lr)])
for i in range(0,int(num_minibatches_to_train)):
mb=cm.get_next_minibatch(minibatch_size, cntk_dev)
arguments = dict()
arguments[input] = mb[features_si].m_data
arguments[label] = mb[labels_si].m_data
trainer.train_minibatch(arguments, cntk_dev)
freq = 20
if i % freq == 0:
training_loss = get_train_loss(trainer)
eval_crit = get_train_eval_criterion(trainer)
print ("Minibatch: {}, Train Loss: {}, Train Evaluation Criterion: {}".format(i, training_loss, eval_crit))
def train_sequence_classifier(device):
input_dim = 2000
cell_dim = 25
hidden_dim = 25
embedding_dim = 50
num_output_classes = 5
features = variable(shape=input_dim, is_sparse=True, name="features")
classifier_output = LSTM_sequence_classifer_net(features,
num_output_classes,
embedding_dim, hidden_dim,
cell_dim, device)
label = variable(num_output_classes,
dynamic_axes=[Axis.default_batch_axis()],
name="labels")
ce = cross_entropy_with_softmax(classifier_output, label)
pe = classification_error(classifier_output, label)
#TODO: add save and load module code
lstm_net = combine([ce, pe, classifier_output], "classifier_model")
rel_path = r"../../../../Tests/EndToEndTests/Text/SequenceClassification/Data/Train.ctf"
path = os.path.join(os.path.dirname(os.path.abspath(__file__)), rel_path)
cm = create_text_mb_source(path, input_dim, num_output_classes, 0, True,
False, "x", "y")
stream_infos = cm.stream_infos()
for si in stream_infos:
if si.m_name == 'features':
features_si = si
elif si.m_name == 'labels':
labels_si = si
minibatch_size = 200
lr = lr = learning_rates_per_sample(0.0005)
trainer = Trainer(classifier_output, ce, pe,
[sgdlearner(classifier_output.owner.parameters(), lr)])
freq = 1
i = 0
cntk_dev = cntk_device(device)
while True:
mb = cm.get_next_minibatch(minibatch_size, cntk_dev)
if len(mb) == 0:
break
arguments = dict()
arguments[features] = mb[features_si].m_data
arguments[label] = mb[labels_si].m_data
trainer.train_minibatch(arguments, cntk_dev)
if i % freq == 0:
training_loss = get_train_loss(trainer)
eval_crit = get_train_eval_criterion(trainer)
print(
"Minibatch: {}, Train Loss: {}, Train Evaluation Criterion: {}"
.format(i, training_loss, eval_crit))
i += 1
minibatch_config["deserializers"] = [deserializer_config]
return create_minibatch_source(minibatch_config)
if __name__=='__main__':
input_dim = 2
num_output_classes = 2
num_hidden_layers = 2
hidden_layers_dim = 50
minibatch_size = 25
num_samples_per_sweep = 10000
num_sweeps_to_train_with = 2
num_minibatches_to_train = (num_samples_per_sweep * num_sweeps_to_train_with) / minibatch_size
lr = 0.02
input = variable((input_dim,), np.float32, needs_gradient=False, name="features")
label = variable((num_output_classes,), np.float32, needs_gradient=False, name="labels")
dev = cntk_py.DeviceDescriptor.cpudevice()
netout = fully_connected_classifier_net(input, num_output_classes, hidden_layers_dim, num_hidden_layers, dev, sigmoid)
ce = cross_entropy_with_softmax(netout.output(), label)
pe = classification_error(netout.output(), label)
ffnet = combine([ce, pe, netout], "classifier_model")
cm = create_mb_source()
streamInfos = cm.stream_infos();
minibatchSizeLimits = dict()
minibatchSizeLimits[streamInfos[0]] = (0,minibatch_size)
minibatchSizeLimits[streamInfos[1]] = (0,minibatch_size)