def _truncated_normal(self, shape, dtype):
key, counter = self._prepare_key_counter(shape)
return gen_stateless_random_ops_v2.stateless_truncated_normal_v2(
shape=shape,
key=key,
counter=counter,
dtype=dtype,
alg=self.algorithm)
def _truncated_normal(self, shape, dtype):
if compat.forward_compatible(2020, 10, 25):
key, counter = self._prepare_key_counter(shape)
return gen_stateless_random_ops_v2.stateless_truncated_normal_v2(
shape=shape,
key=key,
counter=counter,
dtype=dtype,
alg=self.algorithm)
return gen_stateful_random_ops.stateful_truncated_normal(
self.state.handle, self.algorithm, shape, dtype=dtype)
def stateless_truncated_normal(shape,
seed,
mean=0.0,
stddev=1.0,
dtype=dtypes.float32,
name=None):
"""Outputs deterministic pseudorandom values, truncated normally distributed.
This is a stateless version of `tf.random.truncated_normal`: if run twice with
the same seeds and shapes, it will produce the same pseudorandom numbers. The
output is consistent across multiple runs on the same hardware (and between
CPU and GPU), but may change between versions of TensorFlow or on non-CPU/GPU
hardware.
The generated values follow a normal distribution with specified mean and
standard deviation, except that values whose magnitude is more than 2 standard
deviations from the mean are dropped and re-picked.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
mean: A 0-D Tensor or Python value of type `dtype`. The mean of the
truncated normal distribution.
stddev: A 0-D Tensor or Python value of type `dtype`. The standard deviation
of the normal distribution, before truncation.
dtype: The type of the output.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random truncated normal values.
"""
with ops.name_scope(name, "stateless_truncated_normal",
[shape, seed, mean, stddev]) as name:
shape = tensor_util.shape_tensor(shape)
mean = ops.convert_to_tensor(mean, dtype=dtype, name="mean")
stddev = ops.convert_to_tensor(stddev, dtype=dtype, name="stddev")
if compat.forward_compatible(2020, 10, 25):
key, counter, alg = _get_key_counter_alg(seed)
rnd = gen_stateless_random_ops_v2.stateless_truncated_normal_v2(
shape, key=key, counter=counter, dtype=dtype, alg=alg)
else:
rnd = gen_stateless_random_ops.stateless_truncated_normal(
shape, seed, dtype)
result = math_ops.add(rnd * stddev, mean, name=name)
tensor_util.maybe_set_static_shape(result, shape)
return result
