Python math_ops.cos方法代码示例

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def cosine_decay(learning_rate, global_step, max_step, name=None):
    from tensorflow.python.framework import ops
    from tensorflow.python.ops import math_ops
    from tensorflow.python.framework import constant_op

    with ops.name_scope(name, "CosineDecay",
                        [learning_rate, global_step, max_step]) as name:
        learning_rate = ops.convert_to_tensor(0.5 * learning_rate, name="learning_rate")
        dtype = learning_rate.dtype
        global_step = math_ops.cast(global_step, dtype)

        const = math_ops.cast(constant_op.constant(1), learning_rate.dtype)

        freq = math_ops.cast(constant_op.constant(np.pi / max_step), learning_rate.dtype)
        osc = math_ops.cos(math_ops.multiply(freq, global_step))
        osc = math_ops.add(osc, const)

        return math_ops.multiply(osc, learning_rate, name=name) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def test_all_unary_elemwise():
    _test_forward_unary_elemwise(_test_abs)
    _test_forward_unary_elemwise(_test_floor)
    _test_forward_unary_elemwise(_test_exp)
    _test_forward_unary_elemwise(_test_log)
    _test_forward_unary_elemwise(_test_sin)
    _test_forward_unary_elemwise(_test_sqrt)
    _test_forward_unary_elemwise(_test_rsqrt)
    _test_forward_unary_elemwise(_test_neg)
    _test_forward_unary_elemwise(_test_square)
    # ceil and cos come with TFLite 1.14.0.post1 fbs schema
    if package_version.parse(tf.VERSION) >= package_version.parse('1.14.0'):
        _test_forward_unary_elemwise(_test_ceil)
        _test_forward_unary_elemwise(_test_cos)
        _test_forward_unary_elemwise(_test_round)
        # This fails with TF and Tflite 1.15.2, this could not have been tested
        # in CI or anywhere else. The failure mode is that we see a backtrace
        # from the converter that we need to provide a custom Tan operator
        # implementation.
        #_test_forward_unary_elemwise(_test_tan)
        _test_forward_unary_elemwise(_test_elu)

#######################################################################
# Element-wise
# ------------ 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _compute_eta_t(self):
    PI = 3.141592653589793
    t_frac = math_ops.cast(self.t_cur / (self.total_iterations - 1), 'float32')
    eta_t = self.eta_min + 0.5 * (self.eta_max - self.eta_min) * \
        (1 + math_ops.cos(PI * t_frac))
    return eta_t 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _SinGrad(op, grad):
  """Returns grad * cos(x)."""
  x = op.inputs[0]
  with ops.control_dependencies([grad.op]):
    x = math_ops.conj(x)
    return grad * math_ops.cos(x) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _TanGrad(op, grad):
  """Returns grad * 1/sec^2(x)."""
  x = op.inputs[0]
  with ops.control_dependencies([grad.op]):
    x = math_ops.conj(x)
    secx = math_ops.reciprocal(math_ops.cos(x))
    secx2 = math_ops.square(secx)
    return grad * secx2 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def cos(x):
  """Computes cos of x element-wise.

  Arguments:
      x: Tensor or variable.

  Returns:
      A tensor.
  """
  return math_ops.cos(x) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def angles_to_projective_transforms(angles, image_height, image_width):
  """Returns projective transform(s) for the given angle(s).

  Args:
    angles: A scalar angle to rotate all images by, or (for batches of images)
      a vector with an angle to rotate each image in the batch.
    image_height: Height of the image(s) to be transformed.
    image_width: Width of the image(s) to be transformed.

  Returns:
    A tensor of shape (num_images, 8). Projective transforms which can be given
      to `tf.contrib.image.transform`.
  """
  angle_or_angles = ops.convert_to_tensor(
      angles, name="angles", dtype=dtypes.float32)
  if len(angle_or_angles.get_shape()) == 0:  # pylint: disable=g-explicit-length-test
    angles = angle_or_angles[None]
  elif len(angle_or_angles.get_shape()) == 1:
    angles = angle_or_angles
  else:
    raise TypeError("Angles should have rank 0 or 1.")
  x_offset = ((image_width - 1) - (math_ops.cos(angles) *
                                   (image_width - 1) - math_ops.sin(angles) *
                                   (image_height - 1))) / 2.0
  y_offset = ((image_height - 1) - (math_ops.sin(angles) *
                                    (image_width - 1) + math_ops.cos(angles) *
                                    (image_height - 1))) / 2.0
  num_angles = array_ops.shape(angles)[0]
  return array_ops.concat(
      values=[
          math_ops.cos(angles)[:, None],
          -math_ops.sin(angles)[:, None],
          x_offset[:, None],
          math_ops.sin(angles)[:, None],
          math_ops.cos(angles)[:, None],
          y_offset[:, None],
          array_ops.zeros((num_angles, 2), dtypes.float32),
      ],
      axis=1) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def setUp(self):
    super(CoreUnaryOpsTest, self).setUp()

    self.ops = [
        ('abs', operator.abs, math_ops.abs, core.abs_function),
        ('neg', operator.neg, math_ops.negative, core.neg),
        # TODO(shoyer): add unary + to core TensorFlow
        ('pos', None, None, None),
        ('sign', None, math_ops.sign, core.sign),
        ('reciprocal', None, math_ops.reciprocal, core.reciprocal),
        ('square', None, math_ops.square, core.square),
        ('round', None, math_ops.round, core.round_function),
        ('sqrt', None, math_ops.sqrt, core.sqrt),
        ('rsqrt', None, math_ops.rsqrt, core.rsqrt),
        ('log', None, math_ops.log, core.log),
        ('exp', None, math_ops.exp, core.exp),
        ('log', None, math_ops.log, core.log),
        ('ceil', None, math_ops.ceil, core.ceil),
        ('floor', None, math_ops.floor, core.floor),
        ('cos', None, math_ops.cos, core.cos),
        ('sin', None, math_ops.sin, core.sin),
        ('tan', None, math_ops.tan, core.tan),
        ('acos', None, math_ops.acos, core.acos),
        ('asin', None, math_ops.asin, core.asin),
        ('atan', None, math_ops.atan, core.atan),
        ('lgamma', None, math_ops.lgamma, core.lgamma),
        ('digamma', None, math_ops.digamma, core.digamma),
        ('erf', None, math_ops.erf, core.erf),
        ('erfc', None, math_ops.erfc, core.erfc),
        ('lgamma', None, math_ops.lgamma, core.lgamma),
    ]
    total_size = np.prod([v.size for v in self.original_lt.axes.values()])
    self.test_lt = core.LabeledTensor(
        math_ops.cast(self.original_lt, dtypes.float32) / total_size,
        self.original_lt.axes) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _TanGrad(op, grad):
  """Returns grad * 1/sec^2(x)."""
  x = op.inputs[0]
  with ops.control_dependencies([grad.op]):
    x = math_ops.conj(x)
    secx = math_ops.inv(math_ops.cos(x))
    secx2 = math_ops.square(secx)
    return grad * secx2 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _test_cos(data):
    """ One iteration of cos """
    return _test_unary_elemwise(math_ops.cos, data)
#######################################################################
# Tan
# --- 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _SinGrad(op, grad):
  """Returns grad * cos(x)."""
  x = op.inputs[0]
  with ops.control_dependencies([grad]):
    x = math_ops.conj(x)
    return grad * math_ops.cos(x) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _TanGrad(op, grad):
  """Returns grad * 1/sec^2(x)."""
  x = op.inputs[0]
  with ops.control_dependencies([grad]):
    x = math_ops.conj(x)
    secx = math_ops.reciprocal(math_ops.cos(x))
    secx2 = math_ops.square(secx)
    return grad * secx2 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def map(self, input_tensor):
    """Maps each row of input_tensor using random Fourier features.

    Args:
      input_tensor: a `Tensor` containing input features. It's shape is
      [batch_size, self._input_dim].

    Returns:
      A `Tensor` of shape [batch_size, self._output_dim] containing RFFM-mapped
      features.

    Raises:
      InvalidShapeError: if the shape of the `input_tensor` is inconsistent with
        expected input dimension.
    """
    input_tensor_shape = input_tensor.get_shape()
    if len(input_tensor_shape) != 2:
      raise dkm.InvalidShapeError(
          'The shape of the tensor should be 2. Got %d instead.' %
          len(input_tensor_shape))

    features_dim = input_tensor_shape[1]
    if features_dim != self._input_dim:
      raise dkm.InvalidShapeError(
          'Invalid dimension: expected %d input features, got %d instead.' %
          (self._input_dim, features_dim))

    # Add ops that compute (deterministically) omega_matrix and bias based on
    # the provided seed.
    # TODO(sibyl-vie3Poto): Storing the mapper's parameters (omega_matrix and bias) as
    # constants incurs no RPC calls to the parameter server during distributed
    # training. However, if the parameters grow too large (for instance if they
    # don't fit into memory or if they blow up the size of the GraphDef proto),
    # stroring them as constants is no longer an option. In this case, we should
    # have a heuristic to choose out of one of the following alternatives:
    # a) store them as variables (in the parameter server)
    # b) store them as worker local variables
    # c) generating on the fly the omega matrix at each step
    np.random.seed(self._seed)
    omega_matrix_shape = [self._input_dim, self._output_dim]
    bias_shape = [self._output_dim]

    omega_matrix = constant_op.constant(
        np.random.normal(
            scale=1.0 / self._stddev, size=omega_matrix_shape),
        dtype=dtypes.float32)
    bias = constant_op.constant(
        np.random.uniform(
            low=0.0, high=2 * np.pi, size=bias_shape),
        dtype=dtypes.float32)

    x_omega_plus_bias = math_ops.add(
        math_ops.matmul(input_tensor, omega_matrix), bias)
    return math.sqrt(2.0 / self._output_dim) * math_ops.cos(x_omega_plus_bias) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import cos [as 别名]
def _add_sinusoids_signal(x, time, min_timescale=1.0, max_timescale=1.0e4):
        """Adds a bunch of sinusoids of different frequencies to a Tensor.

        Each channel of the input Tensor is incremented by a sinusoid of a different
        frequency and phase.

        This allows attention to learn to use absolute and relative positions.
        Timing signals should be added to some precursors of both the query and the
        memory inputs to attention.

        The use of relative position is possible because sin(x+y) and cos(x+y) can be
        experessed in terms of y, sin(x) and cos(x).

        In particular, we use a geometric sequence of timescales starting with
        min_timescale and ending with max_timescale.  The number of different
        timescales is equal to channels / 2. For each timescale, we
        generate the two sinusoidal signals sin(timestep/timescale) and
        cos(timestep/timescale).  All of these sinusoids are concatenated in
        the channels dimension.

        Args:
          x: a Tensor with shape [batch, length, channels]
          min_timescale: a float
          max_timescale: a float

        Returns:
          a Tensor the same shape as x.
        """
        channels = x.get_shape().as_list()[-1]
        if x.get_shape().ndims == 3:  # [batch_size, timesteps, dim]
            length = array_ops.shape(x)[1]
            position = math_ops.to_float(math_ops.range(length))
        elif x.get_shape().ndims == 2:  # [batch_size, dim]
            length = 1
            position = math_ops.to_float(math_ops.range(time, time + 1))
        else:
            raise ValueError("need a Tensor with rank 2 or 3")
        num_timescales = channels // 2
        log_timescale_increment = (
            math.log(float(max_timescale) / float(min_timescale)) /
            (math_ops.to_float(num_timescales) - 1))
        inv_timescales = min_timescale * math_ops.exp(
            math_ops.to_float(math_ops.range(num_timescales)) * -log_timescale_increment)
        scaled_time = array_ops.expand_dims(position, 1) * array_ops.expand_dims(inv_timescales, 0)
        signal = array_ops.concat([math_ops.sin(scaled_time), math_ops.cos(scaled_time)], axis=1)
        signal = array_ops.pad(signal, [[0, 0], [0, math_ops.mod(channels, 2)]])
        if x.get_shape().ndims == 3:
            signal = array_ops.reshape(signal, [1, length, channels])
        else:
            signal = array_ops.reshape(signal, [1, channels])
        return x + signal