def TruncatedNormal(self, tf_node, inputs):
"""
Outputs random values from a truncated normal distribution.
`tf.truncated_normal()` call generates several ops, the
The `TruncatedNormal` op is what we implement here.
shape --> TruncatedNormal
|
V
stddev -----> Mul
|
V
mean -------> Add
|
V
(output)
"""
# get inputs
shape = tuple(inputs[0].const.astype(int))
# generate truncated standard normal
mu, sigma, lo, up = 0., 1., -2., 2
generator = scipy.stats.truncnorm(
(lo - mu) / sigma, (up - mu) / sigma, loc=mu, scale=sigma)
np_val = generator.rvs(shape)
return ns.constant(np_val, name=tf_node.name)
def Const(self, tf_node, inputs):
# convert to numpy value
np_val = tensor_util.MakeNdarray(tf_node.attr['value'].tensor)
if np_val.dtype == np.dtype('O'):
return None
else:
return ns.constant(np_val, name=tf_node.name)
def Fill(self, tf_node, inputs):
# get inputs
shape_op, const_val_op = inputs
# get shape, const_val
shape = tuple(shape_op.const.astype(int))
const_val = const_val_op.const
# convert to numpy value
np_val = np.zeros(shape)
np_val.fill(const_val)
# create op
return ns.constant(np_val, name=tf_node.name)
def RandomStandardNormal(self, tf_node, inputs):
"""
Outputs random values from a normal distribution. `tf.random_normal()`
call generates several ops. The `RandomStandardNormal` op is what we
implement here.
Inputs to tf_node:
shape, mean, dtype, seed, name
"""
# get inputs
shape = tuple(inputs[0].const.astype(int))
# generate standard normal
np_val = np.random.standard_normal(size=shape)
return ns.constant(np_val, name=tf_node.name)
def ZerosLike(self, tf_node, inputs):
shape = inputs[0].axes.lengths
np_val = np.zeros(shape)
return ns.constant(np_val, name=tf_node.name)