def main():
train, test, _ = imdb.load_data(path='imdb.pkl',
n_words=10000,
valid_portion=0.1)
trainX, trainY = train
testX, testY = test
trainX = pad_sequences(trainX, maxlen=100, value=0.)
testX = pad_sequences(testX, maxlen=100, value=0.)
trainY = np.asarray(trainY)
testY = np.asarray(testY)
data_set = DataSet(trainX, trainY, testX, testY)
training_cnf = {
'classification': True,
'validation_scores': [('validation accuracy', util.accuracy_tf)],
'num_epochs': 50,
'input_size': (100, ),
'lr_policy': StepDecayPolicy(schedule={
0: 0.01,
30: 0.001,
})
}
util.init_logging('train.log',
file_log_level=logging.INFO,
console_log_level=logging.INFO)
learner = SupervisedLearner(model,
training_cnf,
classification=training_cnf['classification'],
is_summary=False)
learner.fit(data_set, weights_from=None, start_epoch=1)
def train():
mnist = input_data.read_data_sets("MNIST_data/", one_hot=False)
width = 28
height = 28
train_images = mnist[0].images.reshape(-1, height, width, 1)
train_labels = mnist[0].labels
validation_images = mnist[1].images.reshape(-1, height, width, 1)
validation_labels = mnist[1].labels
data_set = DataSet(train_images, train_labels,
validation_images, validation_labels)
training_cnf = {
'classification': True,
'validation_scores': [('validation accuracy', util.accuracy_wrapper), ('validation kappa', util.kappa_wrapper)],
'num_epochs': 50,
'lr_policy': StepDecayPolicy(
schedule={
0: 0.01,
30: 0.001,
}
)
}
util.init_logging('train.log', file_log_level=logging.INFO,
console_log_level=logging.INFO)
trainer = SupervisedTrainer(model, training_cnf, classification=training_cnf[
'classification'], is_summary=True)
trainer.fit(data_set, weights_from=None,
start_epoch=1, verbose=1, summary_every=10)
def train():
mnist = input_data.read_data_sets("MNIST_data/", one_hot=False)
width = 28
height = 28
train_images = mnist[0].images.reshape(-1, height, width, 1)
train_labels = mnist[0].labels
validation_images = mnist[1].images.reshape(-1, height, width, 1)
validation_labels = mnist[1].labels
data_set = DataSet(train_images, train_labels, validation_images, validation_labels)
training_cnf = {
'classification':
True,
'validation_scores': [('accuracy', tf.metrics.accuracy),
('kappa', tf.contrib.metrics.cohen_kappa)],
'num_epochs':
50,
'batch_size_train':
32,
'batch_size_test':
32,
'input_size': (28, 28, 1),
'lr_policy':
StepDecayPolicy(schedule={
0: 0.01,
30: 0.001,
})
}
learner = SupervisedLearner(
model,
training_cnf,
classification=training_cnf['classification'],
is_summary=True,
num_classes=10)
learner.fit(data_set, weights_from=None, start_epoch=1, summary_every=10)
'iterator_type':
'queued', # parallel or queued
'batch_size_train':
16,
'batch_size_test':
16,
'l2_reg':
0.002,
'aug_params': {
'zoom_range': (1 / 1.05, 1.05),
'rotation_range': (-5, 5),
'shear_range': (0, 0),
'translation_range': (-20, 20),
'do_flip': True,
'allow_stretch': True,
},
'num_epochs':
100,
'summary_every':
5,
'lr_policy':
StepDecayPolicy(schedule={
0: 0.001,
15: 0.0001,
}),
'optimizer':
tf.train.AdamOptimizer(),
'validation_scores': [('validation accuracy', util.accuracy_wrapper),
('validation kappa', util.kappa_wrapper)],
}
logits = fully_connected(x, n_output=10, name="logits", **logit_args)
predictions = softmax(logits, name='predictions', **common_args)
return end_points(is_training)
training_cnf = {
'classification':
True,
'validation_scores': [('validation accuracy', util.accuracy_wrapper),
('validation kappa', util.kappa_wrapper)],
'num_epochs':
50,
'lr_policy':
StepDecayPolicy(schedule={
0: 0.01,
30: 0.001,
})
}
util.init_logging('train.log',
file_log_level=logging.INFO,
console_log_level=logging.INFO)
trainer = SupervisedTrainer(model,
training_cnf,
classification=training_cnf['classification'])
trainer.fit(data_set,
weights_from=None,
start_epoch=1,
verbose=1,
summary_every=10)
},
'standardizer':
AggregateStandardizer(mean=np.array(
[108.64628601, 75.86886597, 54.34005737], dtype=np.float32),
std=np.array([70.53946096, 51.71475228, 43.03428563],
dtype=np.float32),
u=np.array([[-0.56543481, 0.71983482, 0.40240142],
[-0.5989477, -0.02304967, -0.80036049],
[-0.56694071, -0.6935729, 0.44423429]],
dtype=np.float32),
ev=np.array([1.65513492, 0.48450358, 0.1565086],
dtype=np.float32),
sigma=0.5),
'num_epochs':
555,
# 'lr_policy': PolyDecayPolicy(0.00005),
'lr_policy':
StepDecayPolicy({
0: 0.0002,
100: 0.0002,
200: 0.0002,
400: 0.0002,
500: 0.0001
}),
'classification':
True,
'validation_scores': [('validation accuracy', util.accuracy_wrapper),
('validation kappa', util.kappa_wrapper)],
# 'validation_scores': [('validation kappa', util.kappa_wrapper)],
}
from tefla.core.lr_policy import StepDecayPolicy
from tefla.utils import util
cnf = {
'name': __name__.split('.')[-1],
'classification': True,
'iterator_type': 'queued', # parallel or queued
'batch_size_train': 16,
'batch_size_test': 16,
'l2_reg': 0.005,
'aug_params': {
'zoom_range': (1 / 1.1, 1.1),
'rotation_range': (-5, 5),
'shear_range': (1, 1),
'translation_range': (-20, 20),
'do_flip': True,
'allow_stretch': True,
},
'num_epochs': 15,
'summary_every': 5,
'lr_policy': StepDecayPolicy(
schedule={
0: 0.001,
10:0.0001,
}
),
'optimizer': tf.train.AdamOptimizer(),
'validation_scores': [('validation accuracy', util.accuracy_wrapper), ('validation kappa', util.kappa_wrapper)],
}