def logistic_regression(X, y, class_weight=None):
"""Creates logistic regression TensorFlow subgraph.
Args:
X: tensor or placeholder for input features,
shape should be [batch_size, n_features].
y: tensor or placeholder for target,
shape should be [batch_size, n_classes].
class_weight: tensor, [n_classes], where for each class
it has weight of the class. If not provided
all ones are used.
Returns:
Predictions and loss tensors.
"""
with tf.variable_scope('logistic_regression'):
tf.histogram_summary('logistic_regression.X', X)
tf.histogram_summary('logistic_regression.y', y)
weights = tf.get_variable(
'weights', [X.get_shape()[1], y.get_shape()[-1]])
bias = tf.get_variable('bias', [y.get_shape()[-1]])
tf.histogram_summary('logistic_regression.weights', weights)
tf.histogram_summary('logistic_regression.bias', bias)
return softmax_classifier(X,
y,
weights,
bias,
class_weight=class_weight)
def logistic_regression(X, y, class_weight=None):
"""Creates logistic regression TensorFlow subgraph.
Args:
X: tensor or placeholder for input features,
shape should be [batch_size, n_features].
y: tensor or placeholder for target,
shape should be [batch_size, n_classes].
class_weight: tensor, [n_classes], where for each class
it has weight of the class. If not provided
all ones are used.
Returns:
Predictions and loss tensors.
"""
with tf.variable_scope('logistic_regression'):
tf.histogram_summary('logistic_regression.X', X)
tf.histogram_summary('logistic_regression.y', y)
weights = tf.get_variable('weights', [X.get_shape()[1],
y.get_shape()[-1]])
bias = tf.get_variable('bias', [y.get_shape()[-1]])
tf.histogram_summary('logistic_regression.weights', weights)
tf.histogram_summary('logistic_regression.bias', bias)
return softmax_classifier(X, y, weights, bias,
class_weight=class_weight)
def logistic_regression(X, y, class_weight=None):
"""Creates logistic regression TensorFlow subgraph.
Args:
X: tensor or placeholder for input features,
shape should be [batch_size, n_features].
y: tensor or placeholder for target,
shape should be [batch_size, n_classes].
class_weight: tensor, [n_classes], where for each class
it has weight of the class. If not provided
will check if graph contains tensor `class_weight:0`.
If that is not provided either all ones are used.
Returns:
Predictions and loss tensors.
"""
with tf.variable_scope('logistic_regression'):
tf.histogram_summary('logistic_regression.X', X)
tf.histogram_summary('logistic_regression.y', y)
weights = tf.get_variable('weights', [X.get_shape()[1],
y.get_shape()[-1]])
bias = tf.get_variable('bias', [y.get_shape()[-1]])
tf.histogram_summary('logistic_regression.weights', weights)
tf.histogram_summary('logistic_regression.bias', bias)
# If no class weight provided, try to retrieve one from pre-defined
# tensor name in the graph.
if not class_weight:
try:
class_weight = tf.get_default_graph().get_tensor_by_name('class_weight:0')
except KeyError:
pass
return softmax_classifier(X, y, weights, bias,
class_weight=class_weight)
def __init__(self, n_classes, input_shape, graph):
with graph.as_default():
self.global_step = tf.Variable(0, name="global_step", trainable=False)
self.inp = tf.placeholder(tf.float32, [None, input_shape], name="input")
self.out = tf.placeholder(tf.float32, [None, n_classes], name="output")
self.weights = tf.get_variable("weights", [input_shape, n_classes])
self.bias = tf.get_variable("bias", [n_classes])
self.predictions, self.loss = softmax_classifier(
self.inp, self.out, self.weights, self.bias)
def test_softmax_classifier(self):
with self.test_session() as session:
features = tf.placeholder(tf.float32, [None, 3])
labels = tf.placeholder(tf.float32, [None, 2])
weights = tf.constant([[0.1, 0.1], [0.1, 0.1], [0.1, 0.1]])
biases = tf.constant([0.2, 0.3])
class_weight = tf.constant([0.1, 0.9])
prediction, loss = ops.softmax_classifier(features, labels, weights, biases, class_weight)
self.assertEqual(prediction.get_shape()[1], 2)
self.assertEqual(loss.get_shape(), [])
value = session.run(loss, {features: [[0.2, 0.3, 0.2]], labels: [[0, 1]]})
self.assertAllClose(value, 0.55180627)
def logistic_regression(X, y):
"""Creates logistic regression TensorFlow subgraph.
Args:
X: tensor or placeholder for input features.
y: tensor or placeholder for target.
Returns:
Predictions and loss tensors.
"""
with tf.variable_scope('logistic_regression'):
weights = tf.get_variable('weights', [X.get_shape()[1], y.get_shape()[1]])
bias = tf.get_variable('bias', [y.get_shape()[1]])
return softmax_classifier(X, y, weights, bias)
def logistic_regression(X, y):
"""Creates logistic regression TensorFlow subgraph.
Args:
X: tensor or placeholder for input features.
y: tensor or placeholder for target.
Returns:
Predictions and loss tensors.
"""
with tf.variable_scope('logistic_regression'):
weights = tf.get_variable(
'weights', [X.get_shape()[1], y.get_shape()[1]])
bias = tf.get_variable('bias', [y.get_shape()[1]])
return softmax_classifier(X, y, weights, bias)
def test_softmax_classifier(self):
with self.test_session() as session:
features = tf.placeholder(tf.float32, [None, 3])
labels = tf.placeholder(tf.float32, [None, 2])
weights = tf.constant([[0.1, 0.1], [0.1, 0.1], [0.1, 0.1]])
biases = tf.constant([0.2, 0.3])
class_weight = tf.constant([0.1, 0.9])
prediction, loss = ops.softmax_classifier(features, labels,
weights, biases,
class_weight)
self.assertEqual(prediction.get_shape()[1], 2)
self.assertEqual(loss.get_shape(), [])
value = session.run(loss, {
features: [[0.2, 0.3, 0.2]],
labels: [[0, 1]]
})
self.assertAllClose(value, 0.55180627)
def logistic_regression(X, y):
"""Creates logistic regression TensorFlow subgraph.
Args:
X: tensor or placeholder for input features,
shape should be [batch_size, n_features].
y: tensor or placeholder for target,
shape should be [batch_size, n_classes].
Returns:
Predictions and loss tensors.
"""
with tf.variable_scope('logistic_regression'):
tf.histogram_summary('logistic_regression.X', X)
tf.histogram_summary('logistic_regression.y', y)
weights = tf.get_variable('weights', [X.get_shape()[1],
y.get_shape()[-1]])
bias = tf.get_variable('bias', [y.get_shape()[-1]])
tf.histogram_summary('logistic_regression.weights', weights)
tf.histogram_summary('logistic_regression.bias', bias)
return softmax_classifier(X, y, weights, bias)
def logistic_regression(X, y, class_weight=None):
"""Creates logistic regression TensorFlow subgraph.
Args:
X: tensor or placeholder for input features,
shape should be [batch_size, n_features].
y: tensor or placeholder for target,
shape should be [batch_size, n_classes].
class_weight: tensor, [n_classes], where for each class
it has weight of the class. If not provided
will check if graph contains tensor `class_weight:0`.
If that is not provided either all ones are used.
Returns:
Predictions and loss tensors.
"""
with tf.variable_scope('logistic_regression'):
tf.histogram_summary('logistic_regression.X', X)
tf.histogram_summary('logistic_regression.y', y)
weights = tf.get_variable(
'weights', [X.get_shape()[1], y.get_shape()[-1]])
bias = tf.get_variable('bias', [y.get_shape()[-1]])
tf.histogram_summary('logistic_regression.weights', weights)
tf.histogram_summary('logistic_regression.bias', bias)
# If no class weight provided, try to retrieve one from pre-defined
# tensor name in the graph.
if not class_weight:
try:
class_weight = tf.get_default_graph().get_tensor_by_name(
'class_weight:0')
except KeyError:
pass
return softmax_classifier(X,
y,
weights,
bias,
class_weight=class_weight)
