Python tensorflow.python.ops.random_ops.multinomial() Examples

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.total_count, dtype=dtypes.int32)
    k = self.event_shape_tensor()[0]
    unnormalized_logits = array_ops.reshape(
        math_ops.log(random_ops.random_gamma(
            shape=[n],
            alpha=self.concentration,
            dtype=self.dtype,
            seed=seed)),
        shape=[-1, k])
    draws = random_ops.multinomial(
        logits=unnormalized_logits,
        num_samples=n_draws,
        seed=distribution_util.gen_new_seed(seed, salt="dirichlet_multinomial"))
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k), -2)
    final_shape = array_ops.concat([[n], self.batch_shape_tensor(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.n, dtype=dtypes.int32)
    if self.n.get_shape().ndims is not None:
      if self.n.get_shape().ndims != 0:
        raise NotImplementedError(
            "Sample only supported for scalar number of draws.")
    elif self.validate_args:
      is_scalar = check_ops.assert_rank(
          n_draws, 0,
          message="Sample only supported for scalar number of draws.")
      n_draws = control_flow_ops.with_dependencies([is_scalar], n_draws)
    k = self.event_shape()[0]
    # Flatten batch dims so logits has shape [B, k],
    # where B = reduce_prod(self.batch_shape()).
    logits = array_ops.reshape(self.logits, [-1, k])
    draws = random_ops.multinomial(logits=logits,
                                   num_samples=n * n_draws,
                                   seed=seed)
    draws = array_ops.reshape(draws, shape=[-1, n, n_draws])
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k),
                            reduction_indices=-2)  # shape: [B, n, k]
    x = array_ops.transpose(x, perm=[1, 0, 2])
    final_shape = array_ops.concat([[n], self.batch_shape(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.n, dtype=dtypes.int32)
    if self.n.get_shape().ndims is not None:
      if self.n.get_shape().ndims != 0:
        raise NotImplementedError(
            "Sample only supported for scalar number of draws.")
    elif self.validate_args:
      is_scalar = check_ops.assert_rank(
          n_draws, 0,
          message="Sample only supported for scalar number of draws.")
      n_draws = control_flow_ops.with_dependencies([is_scalar], n_draws)
    k = self.event_shape()[0]
    # Flatten batch dims so logits has shape [B, k],
    # where B = reduce_prod(self.batch_shape()).
    logits = array_ops.reshape(self.logits, [-1, k])
    draws = random_ops.multinomial(logits=logits,
                                   num_samples=n * n_draws,
                                   seed=seed)
    draws = array_ops.reshape(draws, shape=[-1, n, n_draws])
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k),
                            reduction_indices=-2)  # shape: [B, n, k]
    x = array_ops.transpose(x, perm=[1, 0, 2])
    final_shape = array_ops.concat([[n], self.batch_shape(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.total_count, dtype=dtypes.int32)
    k = self.event_shape_tensor()[0]
    unnormalized_logits = array_ops.reshape(
        math_ops.log(random_ops.random_gamma(
            shape=[n],
            alpha=self.concentration,
            dtype=self.dtype,
            seed=seed)),
        shape=[-1, k])
    draws = random_ops.multinomial(
        logits=unnormalized_logits,
        num_samples=n_draws,
        seed=distribution_util.gen_new_seed(seed, salt="dirichlet_multinomial"))
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k), -2)
    final_shape = array_ops.concat([[n], self.batch_shape_tensor(), [k]], 0)
    x = array_ops.reshape(x, final_shape)
    return math_ops.cast(x, self.dtype) 

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.total_count, dtype=dtypes.int32)
    if self.total_count.get_shape().ndims is not None:
      if self.total_count.get_shape().ndims != 0:
        raise NotImplementedError(
            "Sample only supported for scalar number of draws.")
    elif self.validate_args:
      is_scalar = check_ops.assert_rank(
          n_draws, 0,
          message="Sample only supported for scalar number of draws.")
      n_draws = control_flow_ops.with_dependencies([is_scalar], n_draws)
    k = self.event_shape_tensor()[0]
    # Flatten batch dims so logits has shape [B, k],
    # where B = reduce_prod(self.batch_shape_tensor()).
    x = random_ops.multinomial(
        logits=array_ops.reshape(self.logits, [-1, k]),
        num_samples=n * n_draws,
        seed=seed)
    x = array_ops.reshape(x, shape=[-1, n, n_draws])
    x = math_ops.reduce_sum(array_ops.one_hot(x, depth=k),
                            axis=-2)  # shape: [B, n, k]
    x = array_ops.transpose(x, perm=[1, 0, 2])
    final_shape = array_ops.concat([[n], self.batch_shape_tensor(), [k]], 0)
    x = array_ops.reshape(x, final_shape)
    return math_ops.cast(x, self.dtype) 

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.total_count, dtype=dtypes.int32)
    if self.total_count.get_shape().ndims is not None:
      if self.total_count.get_shape().ndims != 0:
        raise NotImplementedError(
            "Sample only supported for scalar number of draws.")
    elif self.validate_args:
      is_scalar = check_ops.assert_rank(
          n_draws, 0,
          message="Sample only supported for scalar number of draws.")
      n_draws = control_flow_ops.with_dependencies([is_scalar], n_draws)
    k = self.event_shape_tensor()[0]
    # Flatten batch dims so logits has shape [B, k],
    # where B = reduce_prod(self.batch_shape_tensor()).
    draws = random_ops.multinomial(
        logits=array_ops.reshape(self.logits, [-1, k]),
        num_samples=n * n_draws,
        seed=seed)
    draws = array_ops.reshape(draws, shape=[-1, n, n_draws])
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k),
                            axis=-2)  # shape: [B, n, k]
    x = array_ops.transpose(x, perm=[1, 0, 2])
    final_shape = array_ops.concat([[n], self.batch_shape_tensor(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

def _make_ops(self, num_samples, seed=None):
    prob_dist = tf.constant([[0.15, 0.5, 0.3, 0.05]])
    logits = tf.log(prob_dist)
    # Two independent sets of samples from the same distribution
    sample_op1 = random_ops.multinomial(logits, num_samples, seed)
    sample_op2 = random_ops.multinomial(logits, num_samples, seed)
    return (sample_op1, sample_op2) 

def _sample_n(self, n, seed=None):
    if self.logits.get_shape().ndims == 2:
      logits_2d = self.logits
    else:
      logits_2d = array_ops.reshape(self.logits, [-1, self.num_classes])
    samples = random_ops.multinomial(logits_2d, n, seed=seed)
    samples = math_ops.cast(samples, self.dtype)
    ret = array_ops.reshape(
        array_ops.transpose(samples),
        array_ops.concat(([n], self.batch_shape()), 0))
    return ret 

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.n, dtype=dtypes.int32)
    if self.n.get_shape().ndims is not None:
      if self.n.get_shape().ndims != 0:
        raise NotImplementedError(
            "Sample only supported for scalar number of draws.")
    elif self.validate_args:
      is_scalar = check_ops.assert_rank(
          n_draws, 0,
          message="Sample only supported for scalar number of draws.")
      n_draws = control_flow_ops.with_dependencies([is_scalar], n_draws)
    k = self.event_shape()[0]
    unnormalized_logits = array_ops.reshape(
        math_ops.log(random_ops.random_gamma(
            shape=[n],
            alpha=self.alpha,
            dtype=self.dtype,
            seed=seed)),
        shape=[-1, k])
    draws = random_ops.multinomial(
        logits=unnormalized_logits,
        num_samples=n_draws,
        seed=distribution_util.gen_new_seed(seed, salt="dirichlet_multinomial"))
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k),
                            reduction_indices=-2)
    final_shape = array_ops.concat([[n], self.batch_shape(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

def _sample_n(self, n, seed=None):
    sample_shape = array_ops.concat(([n], array_ops.shape(self.logits)), 0)
    logits = self.logits
    if logits.get_shape().ndims == 2:
      logits_2d = logits
    else:
      logits_2d = array_ops.reshape(logits, [-1, self.num_classes])
    samples = random_ops.multinomial(logits_2d, n, seed=seed)
    samples = array_ops.transpose(samples)
    samples = array_ops.one_hot(samples, self.num_classes, dtype=self.dtype)
    ret = array_ops.reshape(samples, sample_shape)
    return ret 

def _sample_n(self, n, seed=None):
    if self.logits.get_shape().ndims == 2:
      logits_2d = self.logits
    else:
      logits_2d = array_ops.reshape(self.logits, [-1, self.event_size])
    draws = random_ops.multinomial(logits_2d, n, seed=seed)
    draws = array_ops.reshape(
        array_ops.transpose(draws),
        array_ops.concat([[n], self.batch_shape_tensor()], 0))
    return math_ops.cast(draws, self.dtype) 

def _sample_n(self, n, seed=None):
    if self.logits.get_shape().ndims == 2:
      logits_2d = self.logits
    else:
      logits_2d = array_ops.reshape(self.logits, [-1, self.num_classes])
    samples = random_ops.multinomial(logits_2d, n, seed=seed)
    samples = math_ops.cast(samples, self.dtype)
    ret = array_ops.reshape(
        array_ops.transpose(samples),
        array_ops.concat(0, ([n], self.batch_shape())))
    return ret 

def testEmptyClasses(self):
    with self.test_session(use_gpu=self.use_gpu):
      x = tf.multinomial(tf.zeros([5, 0]), 7)
      with self.assertRaisesOpError("num_classes should be positive"):
        x.eval() 

def testEmpty(self):
    classes = 5
    with self.test_session(use_gpu=self.use_gpu):
      for batch in 0, 3:
        for samples in 0, 7:
          x = tf.multinomial(tf.zeros([batch, classes]), samples).eval()
          self.assertEqual(x.shape, (batch, samples)) 

def testLargeLogits(self):
    for neg in [True, False]:
      with self.test_session(use_gpu=self.use_gpu):
        logits = np.array([[1000.] * 5])
        if neg:
          logits *= -1
        samples = tf.multinomial(logits, 10).eval()
      # Sampled classes should be in-range.
      self.assertTrue((samples >= 0).all())
      self.assertTrue((samples < 5).all()) 

def testSmallEntropy(self):
    tf.set_random_seed(1618)
    with self.test_session(use_gpu=self.use_gpu):
      # A logit value of -10 corresponds to a probability of ~5e-5.
      logits = tf.constant([[-10., 10., -10.], [-10., -10., 10.]])
      num_samples = 1000
      samples = tf.multinomial(logits, num_samples).eval()
      self.assertAllEqual([[1] * num_samples, [2] * num_samples], samples) 

def _sample_n(self, n, seed=None):
    if self.logits.get_shape().ndims == 2:
      logits_2d = self.logits
    else:
      logits_2d = array_ops.reshape(self.logits, [-1, self.num_classes])
    samples = random_ops.multinomial(logits_2d, n, seed=seed)
    samples = math_ops.cast(samples, self.dtype)
    ret = array_ops.reshape(
        array_ops.transpose(samples),
        array_ops.concat(([n], self.batch_shape()), 0))
    return ret 

def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.n, dtype=dtypes.int32)
    if self.n.get_shape().ndims is not None:
      if self.n.get_shape().ndims != 0:
        raise NotImplementedError(
            "Sample only supported for scalar number of draws.")
    elif self.validate_args:
      is_scalar = check_ops.assert_rank(
          n_draws, 0,
          message="Sample only supported for scalar number of draws.")
      n_draws = control_flow_ops.with_dependencies([is_scalar], n_draws)
    k = self.event_shape()[0]
    unnormalized_logits = array_ops.reshape(
        math_ops.log(random_ops.random_gamma(
            shape=[n],
            alpha=self.alpha,
            dtype=self.dtype,
            seed=seed)),
        shape=[-1, k])
    draws = random_ops.multinomial(
        logits=unnormalized_logits,
        num_samples=n_draws,
        seed=distribution_util.gen_new_seed(seed, salt="dirichlet_multinomial"))
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k),
                            reduction_indices=-2)
    final_shape = array_ops.concat([[n], self.batch_shape(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

def _sample_n(self, n, seed=None):
    sample_shape = array_ops.concat(([n], array_ops.shape(self.logits)), 0)
    logits = self.logits
    if logits.get_shape().ndims == 2:
      logits_2d = logits
    else:
      logits_2d = array_ops.reshape(logits, [-1, self.num_classes])
    samples = random_ops.multinomial(logits_2d, n, seed=seed)
    samples = array_ops.transpose(samples)
    samples = array_ops.one_hot(samples, self.num_classes, dtype=self.dtype)
    ret = array_ops.reshape(samples, sample_shape)
    return ret 

def _sample_n(self, n, seed=None):
    sample_shape = array_ops.concat([[n], array_ops.shape(self.logits)], 0)
    logits = self.logits
    if logits.get_shape().ndims == 2:
      logits_2d = logits
    else:
      logits_2d = array_ops.reshape(logits, [-1, self.event_size])
    samples = random_ops.multinomial(logits_2d, n, seed=seed)
    samples = array_ops.transpose(samples)
    samples = array_ops.one_hot(samples, self.event_size, dtype=self.dtype)
    ret = array_ops.reshape(samples, sample_shape)
    return ret 

def _sample_n(self, n, seed=None):
    if self.logits.get_shape().ndims == 2:
      logits_2d = self.logits
    else:
      logits_2d = array_ops.reshape(self.logits, [-1, self.event_size])
    samples = random_ops.multinomial(logits_2d, n, seed=seed)
    samples = math_ops.cast(samples, self.dtype)
    ret = array_ops.reshape(
        array_ops.transpose(samples),
        array_ops.concat([[n], self.batch_shape_tensor()], 0))
    return ret 

def __init__(self,
               total_count,
               concentration,
               validate_args=False,
               allow_nan_stats=True,
               name="DirichletMultinomial"):
    """Initialize a batch of DirichletMultinomial distributions.

    Args:
      total_count:  Non-negative floating point tensor, whose dtype is the same
        as `concentration`. The shape is broadcastable to `[N1,..., Nm]` with
        `m >= 0`. Defines this as a batch of `N1 x ... x Nm` different
        Dirichlet multinomial distributions. Its components should be equal to
        integer values.
      concentration: Positive floating point tensor, whose dtype is the
        same as `n` with shape broadcastable to `[N1,..., Nm, K]` `m >= 0`.
        Defines this as a batch of `N1 x ... x Nm` different `K` class Dirichlet
        multinomial distributions.
      validate_args: Python `bool`, default `False`. When `True` distribution
        parameters are checked for validity despite possibly degrading runtime
        performance. When `False` invalid inputs may silently render incorrect
        outputs.
      allow_nan_stats: Python `bool`, default `True`. When `True`, statistics
        (e.g., mean, mode, variance) use the value "`NaN`" to indicate the
        result is undefined. When `False`, an exception is raised if one or
        more of the statistic's batch members are undefined.
      name: Python `str` name prefixed to Ops created by this class.
    """
    parameters = locals()
    with ops.name_scope(name, values=[total_count, concentration]):
      # Broadcasting works because:
      # * The broadcasting convention is to prepend dimensions of size [1], and
      #   we use the last dimension for the distribution, whereas
      #   the batch dimensions are the leading dimensions, which forces the
      #   distribution dimension to be defined explicitly (i.e. it cannot be
      #   created automatically by prepending). This forces enough explicitness.
      # * All calls involving `counts` eventually require a broadcast between
      #  `counts` and concentration.
      self._total_count = ops.convert_to_tensor(total_count, name="total_count")
      if validate_args:
        self._total_count = (
            distribution_util.embed_check_nonnegative_integer_form(
                self._total_count))
      self._concentration = self._maybe_assert_valid_concentration(
          ops.convert_to_tensor(concentration,
                                name="concentration"),
          validate_args)
      self._total_concentration = math_ops.reduce_sum(self._concentration, -1)
    super(DirichletMultinomial, self).__init__(
        dtype=self._concentration.dtype,
        validate_args=validate_args,
        allow_nan_stats=allow_nan_stats,
        reparameterization_type=distribution.NOT_REPARAMETERIZED,
        parameters=parameters,
        graph_parents=[self._total_count,
                       self._concentration],
        name=name) 

def __init__(self,
               total_count,
               concentration,
               validate_args=False,
               allow_nan_stats=True,
               name="DirichletMultinomial"):
    """Initialize a batch of DirichletMultinomial distributions.

    Args:
      total_count:  Non-negative floating point tensor, whose dtype is the same
        as `concentration`. The shape is broadcastable to `[N1,..., Nm]` with
        `m >= 0`. Defines this as a batch of `N1 x ... x Nm` different
        Dirichlet multinomial distributions. Its components should be equal to
        integer values.
      concentration: Positive floating point tensor, whose dtype is the
        same as `n` with shape broadcastable to `[N1,..., Nm, k]` `m >= 0`.
        Defines this as a batch of `N1 x ... x Nm` different `k` class Dirichlet
        multinomial distributions.
      validate_args: Python `bool`, default `False`. When `True` distribution
        parameters are checked for validity despite possibly degrading runtime
        performance. When `False` invalid inputs may silently render incorrect
        outputs.
      allow_nan_stats: Python `bool`, default `True`. When `True`, statistics
        (e.g., mean, mode, variance) use the value "`NaN`" to indicate the
        result is undefined. When `False`, an exception is raised if one or
        more of the statistic's batch members are undefined.
      name: Python `str` name prefixed to Ops created by this class.
    """
    parameters = locals()
    with ops.name_scope(name, values=[total_count, concentration]):
      # Broadcasting works because:
      # * The broadcasting convention is to prepend dimensions of size [1], and
      #   we use the last dimension for the distribution, whereas
      #   the batch dimensions are the leading dimensions, which forces the
      #   distribution dimension to be defined explicitly (i.e. it cannot be
      #   created automatically by prepending). This forces enough explicitness.
      # * All calls involving `counts` eventually require a broadcast between
      #  `counts` and concentration.
      self._total_count = self._maybe_assert_valid_total_count(
          ops.convert_to_tensor(total_count, name="total_count"),
          validate_args)
      self._concentration = self._maybe_assert_valid_concentration(
          ops.convert_to_tensor(concentration,
                                name="concentration"),
          validate_args)
      self._total_concentration = math_ops.reduce_sum(self._concentration, -1)
    super(DirichletMultinomial, self).__init__(
        dtype=self._concentration.dtype,
        validate_args=validate_args,
        allow_nan_stats=allow_nan_stats,
        reparameterization_type=distribution.NOT_REPARAMETERIZED,
        parameters=parameters,
        graph_parents=[self._total_count,
                       self._concentration],
        name=name)