Skip to content

P
pytensor
提交 dc83dda9 authored 作者: amrithasuresh's avatar amrithasuresh
浏览文件
操作

Updated numpy as np

上级 5715084e
隐藏空白字符变更
内嵌 并排
正在显示 包含 43 行增加 和 43 行删除
theano/tensor/opt.py
浏览文件 @ dc83dda9
......@@ -14,7 +14,7 @@ import time
import traceback
import warnings
import numpy
import numpy as np
from six import integer_types, iteritems
from six.moves import reduce, xrange
......@@ -786,7 +786,7 @@ class MakeVector(T.Op):
# So there will be (1 * nb_dtype) + ((nb len(inp) - 1 ))
# different c code with the following algo
out_shape = len(inp)
out_num = numpy.dtype(node.outputs[0].dtype).num
out_num = np.dtype(node.outputs[0].dtype).num
# don't use dtype_%(out)s as when check_input=False, it isn't defined.
out_dtype = node.outputs[0].type.dtype_specs()[1]
if len(inp) > 0:
......@@ -1064,8 +1064,8 @@ class ShapeFeature(object):
if type(s_i) is float and int(s_i) == s_i:
s_i = int(s_i)
if (type(s_i) in integer_types or
isinstance(s_i, numpy.integer) or
(isinstance(s_i, numpy.ndarray) and s_i.ndim == 0)):
isinstance(s_i, np.integer) or
(isinstance(s_i, np.ndarray) and s_i.ndim == 0)):
# this shape is a constant
if s_i < 0:
msg = "There is a negative shape in the graph!"
......@@ -1975,7 +1975,7 @@ def local_subtensor_remove_broadcastable_index(node):
elif isinstance(elem, slice):
if elem != slice(None):
return
elif isinstance(elem, (integer_types, numpy.integer)):
elif isinstance(elem, (integer_types, np.integer)):
if elem in [0, -1] and node.inputs[0].broadcastable[dim]:
remove_dim.append(dim)
else:
......@@ -2027,7 +2027,7 @@ def local_subtensor_make_vector(node):
else:
return
if isinstance(idx, (integer_types, numpy.integer)):
if isinstance(idx, (integer_types, np.integer)):
# We don't need to copy over any stack traces here
return [x.owner.inputs[idx]]
elif isinstance(idx, Variable):
......@@ -2035,7 +2035,7 @@ def local_subtensor_make_vector(node):
# if it is a constant we can do something with it
try:
v = get_scalar_constant_value(idx, only_process_constants=True)
if isinstance(v, numpy.integer):
if isinstance(v, np.integer):
# Python 2.4 wants to index only with Python integers
v = int(v)
# We don't need to copy over any stack traces here
......@@ -2359,14 +2359,14 @@ class Assert(T.Op):
if not isinstance(value, Variable):
value = T.as_tensor_variable(value)
cond = [T.as_tensor_variable(c) for c in conds]
assert numpy.all([c.type.ndim == 0 for c in cond])
assert np.all([c.type.ndim == 0 for c in cond])
return gof.Apply(self, [value] + cond, [value.type()])
def perform(self, node, inputs, out_):
out, = out_
v = inputs[0]
out[0] = v
assert numpy.all(inputs[1:]), self.msg
assert np.all(inputs[1:]), self.msg
def grad(self, input, output_gradients):
return output_gradients + [DisconnectedType()()] * (len(input) - 1)
......@@ -2708,7 +2708,7 @@ def local_useless_subtensor(node):
length_pos = shape_of[node.inputs[0]][pos]
if isinstance(idx.stop, (integer_types, numpy.integer)):
if isinstance(idx.stop, (integer_types, np.integer)):
length_pos_data = sys.maxsize
try:
length_pos_data = get_scalar_constant_value(length_pos,
......@@ -2766,7 +2766,7 @@ def local_useless_subtensor(node):
idx = idx.value
if len(idx) != length:
return False
if numpy.any(idx != numpy.arange(length)):
if np.any(idx != np.arange(length)):
return False
elif idx.owner is not None and isinstance(idx.owner.op, T.ARange):
try:
......@@ -3625,7 +3625,7 @@ def local_useless_rebroadcast(node):
"""
if isinstance(node.op, T.Rebroadcast):
x = node.inputs[0]
if numpy.all(x.broadcastable == node.outputs[0].broadcastable):
if np.all(x.broadcastable == node.outputs[0].broadcastable):
# No broadcastable flag was modified
# No need to copy over stack trace,
# because x should already have a stack trace.
......@@ -3938,8 +3938,8 @@ def local_useless_switch(node):
isinstance(node.op.scalar_op, scalar.basic.Switch)):
cond = T.extract_constant(node.inputs[0],
only_process_constants=True)
if ((type(cond) is numpy.ndarray and cond.ndim == 0) or
isinstance(cond, numpy.number)):
if ((type(cond) is np.ndarray and cond.ndim == 0) or
isinstance(cond, np.number)):
if cond == 0:
correct_out = node.inputs[2]
else:
......@@ -4985,7 +4985,7 @@ class Canonizer(gof.LocalOptimizer):
num, denum = self.simplify(list(orig_num), list(orig_denum), out.type)
def same(x, y):
return len(x) == len(y) and all(numpy.all(xe == ye) for xe, ye in
return len(x) == len(y) and all(np.all(xe == ye) for xe, ye in
zip(x, y))
if same(orig_num, num) and same(orig_denum, denum):
......@@ -5029,7 +5029,7 @@ def mul_calculate(num, denum, aslist=False, out_type=None):
if aslist:
return []
else:
return numpy.int8(1)
return np.int8(1)
# Make sure we do not accidently upcast data types.
if out_type is None:
......@@ -5038,9 +5038,9 @@ def mul_calculate(num, denum, aslist=False, out_type=None):
out_dtype = out_type.dtype
one = theano._asarray(1, dtype=out_dtype)
v = reduce(numpy.multiply, num, one) / reduce(numpy.multiply, denum, one)
v = reduce(np.multiply, num, one) / reduce(np.multiply, denum, one)
if aslist:
if numpy.all(v == 1):
if np.all(v == 1):
return []
else:
return [v]
......@@ -5053,7 +5053,7 @@ register_canonicalize(local_mul_canonizer, name='local_mul_canonizer')
@gof.local_optimizer([T.neg])
def local_neg_to_mul(node):
if node.op == T.neg:
return [T.mul(numpy.array(-1, dtype=node.inputs[0].dtype),
return [T.mul(np.array(-1, dtype=node.inputs[0].dtype),
node.inputs[0])]
register_canonicalize(local_neg_to_mul)
......@@ -5078,13 +5078,13 @@ def local_sum_prod_mul_by_scalar(node):
if node_inps.owner and node_inps.owner.op == T.mul:
terms = node_inps.owner.inputs
scalars = [t.dimshuffle() for t in terms if
numpy.all(t.type.broadcastable)]
np.all(t.type.broadcastable)]
if len(scalars) == 0:
# Nothing to optimize here
return
non_scalars = [t for t in terms if not numpy.all(t.broadcastable)]
non_scalars = [t for t in terms if not np.all(t.broadcastable)]
# Perform the op only on the non-scalar inputs, if applicable
if len(non_scalars) == 0:
......@@ -5780,7 +5780,7 @@ def local_neg_div_neg(node):
# No other clients of the original division
new_num = num.owner.inputs[0]
return [T.true_div(new_num, denom)]
elif numpy.all(num.broadcastable) and isinstance(num, Constant):
elif np.all(num.broadcastable) and isinstance(num, Constant):
if len(frac.clients) == 1:
new_num = -num.data
return [T.true_div(new_num, denom)]
......@@ -5811,7 +5811,7 @@ register_canonicalize(local_mul_zero)
@gof.local_optimizer([T.true_div])
def local_div_to_inv(node):
if node.op == T.true_div and numpy.all(
if node.op == T.true_div and np.all(
local_mul_canonizer.get_constant(node.inputs[0]) == 1.0):
out = node.outputs[0]
new_out = T.inv(local_mul_canonizer.merge_num_denum(node.inputs[1:],
......@@ -5873,7 +5873,7 @@ def local_intdiv_by_one(node):
"""
if node.op in [T.int_div]:
if isinstance(node.inputs[1], T.TensorConstant) and \
numpy.all(node.inputs[1].value == 1):
np.all(node.inputs[1].value == 1):
return [node.inputs[0].astype(node.outputs[0].dtype)]
......@@ -5906,19 +5906,19 @@ def local_pow_specialize(node):
ysym.type.broadcastable):
rval = None
if numpy.all(y == 2):
if np.all(y == 2):
rval = [T.sqr(xsym)]
if numpy.all(y == 1):
if np.all(y == 1):
rval = [xsym]
if numpy.all(y == 0):
rval = [T.fill(xsym, numpy.asarray(1, dtype=odtype))]
if numpy.all(y == 0.5):
if np.all(y == 0):
rval = [T.fill(xsym, np.asarray(1, dtype=odtype))]
if np.all(y == 0.5):
rval = [T.sqrt(xsym)]
if numpy.all(y == -0.5):
if np.all(y == -0.5):
rval = [T.inv(T.sqrt(xsym))]
if numpy.all(y == -1):
if np.all(y == -1):
rval = [T.inv(xsym)]
if numpy.all(y == -2):
if np.all(y == -2):
rval = [T.inv(T.sqr(xsym))]
if rval:
rval[0] = T.cast(rval[0], odtype)
......@@ -5951,7 +5951,7 @@ def local_pow_specialize_device(node):
# taking the value outside ndarray solve the problem.
# it could be that in that case, numpy make the comparaison
# into the wrong type(do in int8 that overflow.)
if isinstance(y, numpy.ndarray):
if isinstance(y, np.ndarray):
assert y.size == 1
try:
y = y[0]
......@@ -5966,13 +5966,13 @@ def local_pow_specialize_device(node):
pow2 = [xsym]
pow2_scal = [theano.scalar.get_scalar_type(xsym.dtype)()]
y_to_do = abs(y)
for i in xrange(int(numpy.log2(y_to_do))):
for i in xrange(int(np.log2(y_to_do))):
pow2.append(T.sqr(pow2[i]))
pow2_scal.append(theano.scalar.sqr(pow2_scal[i]))
rval1 = None
rval1_scal = None
while y_to_do > 0:
log_to_do = int(numpy.log2(y_to_do))
log_to_do = int(np.log2(y_to_do))
if rval1:
rval1 *= pow2[log_to_do]
rval1_scal *= pow2_scal[log_to_do]
......@@ -6061,7 +6061,7 @@ def local_mul_specialize(node):
elif neg:
# Don't add an extra neg node as we can't
# fully replace this mul by a neg.
m1 = numpy.asarray(-1, dtype=node.outputs[0].dtype)
m1 = np.asarray(-1, dtype=node.outputs[0].dtype)
new_inputs = [m1] + new_inputs
rval = T.mul(*new_inputs)
......@@ -6092,7 +6092,7 @@ def local_add_specialize(node):
y = get_scalar_constant_value(input)
except NotScalarConstantError:
y = input
if numpy.all(y == 0.0):
if np.all(y == 0.0):
continue
new_inputs.append(input)
......@@ -6102,7 +6102,7 @@ def local_add_specialize(node):
# we got rid of the entire expression!
ndim = node.outputs[0].type.ndim
# Reuse call to constant for cache()
cst = T.constant(numpy.zeros((1,) * ndim, dtype=dtype))
cst = T.constant(np.zeros((1,) * ndim, dtype=dtype))
assert cst.type.broadcastable == (True,) * ndim
return fill_chain(cst)
......@@ -6209,7 +6209,7 @@ def local_log1p(node):
scalars, scalar_inputs, nonconsts = scalarconsts_rest(
log_arg.owner.inputs, only_process_constants=True)
# scalar_inputs are potentially dimshuffled and fill'd scalars
if scalars and numpy.allclose(numpy.sum(scalars), 1):
if scalars and np.allclose(np.sum(scalars), 1):
if nonconsts:
if len(nonconsts) > 1:
ninp = T.add(*nonconsts)
......@@ -6315,9 +6315,9 @@ def add_calculate(num, denum, aslist=False, out_type=None):
zero = theano._asarray(0, dtype=out_type.dtype)
# zero = 0.0 if out_type is None else theano._asarray(0,
# dtype=out_type.dtype)
v = reduce(numpy.add, num, zero) - reduce(numpy.add, denum, zero)
v = reduce(np.add, num, zero) - reduce(np.add, denum, zero)
if aslist:
if numpy.all(v == 0):
if np.all(v == 0):
return []
else:
return [v]
......@@ -6708,7 +6708,7 @@ def local_log_erfc(node):
node.tag.local_log_erfc_applied = True
x = node.inputs[0].owner.inputs[0]
stab_value = (-x ** 2 - T.log(x) - .5 * T.log(numpy.pi) +
stab_value = (-x ** 2 - T.log(x) - .5 * T.log(np.pi) +
T.log(1 - 1 / (2 * x ** 2) + 3 / (4 * x ** 4) -
15 / (8 * x ** 6)))
......@@ -6863,7 +6863,7 @@ def local_grad_log_erfc_neg(node):
# aaron value
stab_value = (x * T.pow(1 - 1 / (2 * (x ** 2)) +
3 / (4 * (x ** 4)) - 15 / (8 * (x ** 6)), -1) *
T.cast(T.sqrt(numpy.pi), dtype=x.dtype))
T.cast(T.sqrt(np.pi), dtype=x.dtype))
if x.dtype == 'float32' or x.dtype == 'float16':
threshold = 9.3
......
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论