TensorFlow 2.0 Beta is available Learn more

tf.contrib.factorization.GmmAlgorithm

View source on GitHub

Class GmmAlgorithm

Tensorflow Gaussian mixture model clustering class.

__init__

View source

__init__(
    data,
    num_classes,
    initial_means=None,
    params='wmc',
    covariance_type=FULL_COVARIANCE,
    random_seed=0
)

Constructor.

Args:

  • data: a list of Tensors with data, each row is a new example.
  • num_classes: number of clusters.
  • initial_means: a Tensor with a matrix of means. If None, means are computed by sampling randomly.
  • params: Controls which parameters are updated in the training process. Can contain any combination of "w" for weights, "m" for means, and "c" for covariances.
  • covariance_type: one of "full", "diag".
  • random_seed: Seed for PRNG used to initialize seeds.

Raises:

Exception if covariance type is unknown.

Methods

alphas

View source

alphas()

assignments

View source

assignments()

Returns a list of Tensors with the matrix of assignments per shard.

clusters

View source

clusters()

Returns the clusters with dimensions num_classes X 1 X num_dimensions.

covariances

View source

covariances()

Returns the covariances matrices.

init_ops

View source

init_ops()

Returns the initialization operation.

is_initialized

View source

is_initialized()

Returns a boolean operation for initialized variables.

log_likelihood_op

View source

log_likelihood_op()

Returns the log-likelihood operation.

scores

View source

scores()

Returns the per-sample likelihood fo the data.

Returns:

Log probabilities of each data point.

training_ops

View source

training_ops()

Returns the training operation.

Class Members

  • CLUSTERS_COVS_VARIABLE = 'clusters_covs'
  • CLUSTERS_VARIABLE = 'clusters'
  • CLUSTERS_WEIGHT = 'alphas'