LabeledTensor is a library for adding semantically meaningful dimension and coordinate labels to tensors in Tensorflow.
LabeledTensor was inspired by xarray and pandas, projects that adds labels to NumPy array.
LabeledTensor is an immutable object consisting of two components:
tensor: the tf.Tensor object containing the labeled tensor's data.axes: an OrderedDict-like object with keys given by axis names (e.g., "channel") and values given by Axis objects.Axis objects keep track of the size of a dimension and, optionally, coordinate labels along that axis (e.g., ("red", "green", "blue")) in the form of a tuple stored in Axis.labels.
Operations on LabeledTensors use, preserve and transform axis names and labels.
Try out the following snippet in a script or Jupyter notebook:
import tensorflow as tf
lt = tf.contrib.labeled_tensor
# Create two LabeledTensors:
raw_image = tf.ones((299, 299, 3))
axes = ['row', 'column', ('channel', ['red', 'green', 'blue'])]
image = lt.LabeledTensor(raw_image, axes)
assert image.tensor is raw_image
weights = lt.LabeledTensor(tf.constant([0.1, 0.3, 0.6]),
[image.axes['channel']])
# Examples of valid operations:
lt.transpose(image, ['column', 'row', 'channel'])
lt.reshape(image, ['row', 'column'], ['pixel'])
lt.concat([image, image], 'row')
lt.reduce_sum(image, ['channel'])
lt.select(image, {'channel': 'red'})
lt.cast(image / 256.0, tf.uint8)
image * weights
lt.matmul(image[0, :, :], weights)
tf.cos(image) # automatically converts to tf.Tensor
LabeledTensor has wrappers for quite a few TensorFlow ops.
To easily add your own, you can use the define_unary_op, define_binary_op and define_reduce_op functions, e.g.,
log = lt.define_unary_op('log', tf.log)
Please reach out to the authors: