Skip to content
项目
群组
代码片段
帮助
当前项目
正在载入...
登录 / 注册
切换导航面板
P
pytensor
项目
项目
详情
活动
周期分析
仓库
仓库
文件
提交
分支
标签
贡献者
图表
比较
统计图
议题
0
议题
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
CI / CD
CI / CD
流水线
作业
日程
统计图
Wiki
Wiki
代码片段
代码片段
成员
成员
折叠边栏
关闭边栏
活动
图像
聊天
创建新问题
作业
提交
问题看板
Open sidebar
testgroup
pytensor
Commits
8c745911
提交
8c745911
authored
9月 24, 2008
作者:
projects@lgcm
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
docstrings, hiding inplace ops in tensor
上级
d3613582
全部展开
隐藏空白字符变更
内嵌
并排
正在显示
5 个修改的文件
包含
85 行增加
和
84 行删除
+85
-84
_test_compile.py
_test_compile.py
+2
-2
_test_tensor.py
_test_tensor.py
+0
-0
_test_tensor_opt.py
_test_tensor_opt.py
+12
-12
tensor.py
tensor.py
+70
-69
tensor_opt.py
tensor_opt.py
+1
-1
没有找到文件。
_test_compile.py
浏览文件 @
8c745911
...
@@ -77,7 +77,7 @@ class T_Function(unittest.TestCase):
...
@@ -77,7 +77,7 @@ class T_Function(unittest.TestCase):
def
test_closure
(
self
):
def
test_closure
(
self
):
x
,
y
,
z
=
tensor
.
scalars
(
'xyz'
)
x
,
y
,
z
=
tensor
.
scalars
(
'xyz'
)
v
=
tensor
.
value
(
numpy
.
zeros
(()))
v
=
tensor
.
value
(
numpy
.
zeros
(()))
e
=
x
+
tensor
.
add_inplace
(
v
,
1
)
e
=
x
+
tensor
.
_
add_inplace
(
v
,
1
)
f
=
function
([
x
],
[
e
])
f
=
function
([
x
],
[
e
])
assert
f
(
1.
)
==
2.
assert
f
(
1.
)
==
2.
assert
f
(
1.
)
==
3.
assert
f
(
1.
)
==
3.
...
@@ -109,7 +109,7 @@ class T_Function(unittest.TestCase):
...
@@ -109,7 +109,7 @@ class T_Function(unittest.TestCase):
def
test_borrow_false_through_inplace
(
self
):
def
test_borrow_false_through_inplace
(
self
):
x
,
y
,
z
=
tensor
.
scalars
(
'xyz'
)
x
,
y
,
z
=
tensor
.
scalars
(
'xyz'
)
# if borrow_outputs is False, we must not reuse the temporary created for x+y
# if borrow_outputs is False, we must not reuse the temporary created for x+y
e
=
tensor
.
add_inplace
(
x
+
y
,
z
)
e
=
tensor
.
_
add_inplace
(
x
+
y
,
z
)
for
linker
in
'py c c|py c&py'
.
split
():
for
linker
in
'py c c|py c&py'
.
split
():
f
=
function
([
x
,
y
,
z
],
[
e
],
borrow_outputs
=
False
,
linker
=
linker
)
f
=
function
([
x
,
y
,
z
],
[
e
],
borrow_outputs
=
False
,
linker
=
linker
)
res1
=
f
(
1.0
,
2.0
,
3.0
)
res1
=
f
(
1.0
,
2.0
,
3.0
)
...
...
_test_tensor.py
浏览文件 @
8c745911
差异被折叠。
点击展开。
_test_tensor_opt.py
浏览文件 @
8c745911
...
@@ -3,12 +3,12 @@
...
@@ -3,12 +3,12 @@
import
unittest
import
unittest
from
theano
import
gof
import
gof
from
t
heano.t
ensor_opt
import
*
from
tensor_opt
import
*
from
theano
import
tensor
import
tensor
from
t
heano.t
ensor
import
Tensor
from
tensor
import
Tensor
from
theano.
gof
import
Env
from
gof
import
Env
from
theano.
elemwise
import
DimShuffle
from
elemwise
import
DimShuffle
import
numpy
import
numpy
#import scalar_opt
#import scalar_opt
...
@@ -43,7 +43,7 @@ def inputs(xbc = (0, 0), ybc = (0, 0), zbc = (0, 0)):
...
@@ -43,7 +43,7 @@ def inputs(xbc = (0, 0), ybc = (0, 0), zbc = (0, 0)):
# def test_user_inplace(self):
# def test_user_inplace(self):
# x, y, z = inputs()
# x, y, z = inputs()
# e0 = x + y
# e0 = x + y
# e1 = tensor.mul_inplace(x, y)
# e1 = tensor.
_
mul_inplace(x, y)
# g = Env([x, y], [e0, e1])
# g = Env([x, y], [e0, e1])
# self.failUnless(str(g) == "[Broadcast{Add}(x, y), Broadcast{Mul}{0: 0}(x, y)]")
# self.failUnless(str(g) == "[Broadcast{Add}(x, y), Broadcast{Mul}{0: 0}(x, y)]")
# inplace_optimizer.optimize(g)
# inplace_optimizer.optimize(g)
...
@@ -52,7 +52,7 @@ def inputs(xbc = (0, 0), ybc = (0, 0), zbc = (0, 0)):
...
@@ -52,7 +52,7 @@ def inputs(xbc = (0, 0), ybc = (0, 0), zbc = (0, 0)):
# def test_inplace_on_second_argument(self):
# def test_inplace_on_second_argument(self):
# x, y, z = inputs()
# x, y, z = inputs()
# e0 = x + y
# e0 = x + y
# e1 = tensor.mul_inplace(x, z)
# e1 = tensor.
_
mul_inplace(x, z)
# g = Env([x, y], [e0, e1])
# g = Env([x, y], [e0, e1])
# self.failUnless(str(g) == "[Broadcast{Add}(x, y), Broadcast{Mul}{0: 0}(x, z)]")
# self.failUnless(str(g) == "[Broadcast{Add}(x, y), Broadcast{Mul}{0: 0}(x, z)]")
# inplace_optimizer.optimize(g)
# inplace_optimizer.optimize(g)
...
@@ -98,9 +98,9 @@ class _test_dimshuffle_lift(unittest.TestCase):
...
@@ -98,9 +98,9 @@ class _test_dimshuffle_lift(unittest.TestCase):
from
t
heano.t
ensor
import
*
from
tensor
import
*
from
theano.
sandbox
import
pprint
from
sandbox
import
pprint
class
_test_greedy_distribute
(
unittest
.
TestCase
):
class
_test_greedy_distribute
(
unittest
.
TestCase
):
def
test_main
(
self
):
def
test_main
(
self
):
...
@@ -279,8 +279,8 @@ class _test_canonize(unittest.TestCase):
...
@@ -279,8 +279,8 @@ class _test_canonize(unittest.TestCase):
# # def test_inplace(self):
# # def test_inplace(self):
# # x, y, z = inputs()
# # x, y, z = inputs()
# # #e = tensor.add_inplace(x, y + z)
# # #e = tensor.
_
add_inplace(x, y + z)
# # e = x + tensor.add_inplace(y, z)
# # e = x + tensor.
_
add_inplace(y, z)
# # g = Env([x, y, z], [e])
# # g = Env([x, y, z], [e])
# # opt = CliqueOptimizer(through_broadcast = False,
# # opt = CliqueOptimizer(through_broadcast = False,
# # scalar_optimizer = None,
# # scalar_optimizer = None,
...
...
tensor.py
浏览文件 @
8c745911
...
@@ -415,9 +415,9 @@ class _tensor_py_operators:
...
@@ -415,9 +415,9 @@ class _tensor_py_operators:
def
__rand__
(
self
,
other
):
return
and_
(
other
,
self
)
def
__rand__
(
self
,
other
):
return
and_
(
other
,
self
)
def
__ror__
(
self
,
other
):
return
or_
(
other
,
self
)
def
__ror__
(
self
,
other
):
return
or_
(
other
,
self
)
def
__rxor__
(
self
,
other
):
return
xor
(
other
,
self
)
def
__rxor__
(
self
,
other
):
return
xor
(
other
,
self
)
def
__iand__
(
self
,
other
):
return
and_inplace
(
self
,
other
)
def
__iand__
(
self
,
other
):
return
_
and_inplace
(
self
,
other
)
def
__ior__
(
self
,
other
):
return
or_inplace
(
self
,
other
)
def
__ior__
(
self
,
other
):
return
_
or_inplace
(
self
,
other
)
def
__ixor__
(
self
,
other
):
return
xor_inplace
(
self
,
other
)
def
__ixor__
(
self
,
other
):
return
_
xor_inplace
(
self
,
other
)
#ARITHMETIC - NORMAL
#ARITHMETIC - NORMAL
def
__add__
(
self
,
other
):
return
add
(
self
,
other
)
def
__add__
(
self
,
other
):
return
add
(
self
,
other
)
...
@@ -428,11 +428,11 @@ class _tensor_py_operators:
...
@@ -428,11 +428,11 @@ class _tensor_py_operators:
def
__mod__
(
self
,
other
):
return
mod
(
self
,
other
)
def
__mod__
(
self
,
other
):
return
mod
(
self
,
other
)
#ARITHMETIC - INPLACE
#ARITHMETIC - INPLACE
def
__iadd__
(
self
,
other
):
return
add_inplace
(
self
,
other
)
def
__iadd__
(
self
,
other
):
return
_
add_inplace
(
self
,
other
)
def
__isub__
(
self
,
other
):
return
sub_inplace
(
self
,
other
)
def
__isub__
(
self
,
other
):
return
_
sub_inplace
(
self
,
other
)
def
__imul__
(
self
,
other
):
return
mul_inplace
(
self
,
other
)
def
__imul__
(
self
,
other
):
return
_
mul_inplace
(
self
,
other
)
def
__idiv__
(
self
,
other
):
return
div_inplace
(
self
,
other
)
def
__idiv__
(
self
,
other
):
return
_
div_inplace
(
self
,
other
)
def
__ipow__
(
self
,
other
):
return
pow_inplace
(
self
,
other
)
def
__ipow__
(
self
,
other
):
return
_
pow_inplace
(
self
,
other
)
#ARITHMETIC - RIGHT-OPERAND
#ARITHMETIC - RIGHT-OPERAND
def
__radd__
(
self
,
other
):
return
add
(
other
,
self
)
def
__radd__
(
self
,
other
):
return
add
(
other
,
self
)
...
@@ -493,7 +493,7 @@ elemwise.TensorValue = TensorValue
...
@@ -493,7 +493,7 @@ elemwise.TensorValue = TensorValue
def
_elemwise
(
scalar_op
,
name
,
doc_prefix
=
''
):
def
_elemwise
(
scalar_op
,
name
,
doc_prefix
=
''
):
straight
=
elemwise
.
Elemwise
(
scalar_op
,
name
=
name
)
straight
=
elemwise
.
Elemwise
(
scalar_op
,
name
=
name
)
inplace_scalar_op
=
scalar_op
.
__class__
(
scal
.
transfer_type
(
0
))
inplace_scalar_op
=
scalar_op
.
__class__
(
scal
.
transfer_type
(
0
))
inplace
=
elemwise
.
Elemwise
(
inplace_scalar_op
,
{
0
:
0
},
name
=
name
+
"_inplace"
)
inplace
=
elemwise
.
Elemwise
(
inplace_scalar_op
,
{
0
:
0
},
name
=
'_'
+
name
+
"_inplace"
)
# don't add the inplace versions, they aren't supposed to be part of the user interface
# don't add the inplace versions, they aren't supposed to be part of the user interface
_constructor_list
.
append
(
straight
)
_constructor_list
.
append
(
straight
)
...
@@ -664,97 +664,97 @@ def argmax(x, axis=None):
...
@@ -664,97 +664,97 @@ def argmax(x, axis=None):
# Comparison
# Comparison
##########################
##########################
lt
,
lt_inplace
=
_elemwise
(
scal
.
lt
,
'lt'
)
lt
,
_lt_inplace
=
_elemwise
(
scal
.
lt
,
'lt'
,
"""less than (elemwise)"""
"""less than (elemwise)"""
)
gt
,
gt_inplace
=
_elemwise
(
scal
.
gt
,
'gt'
)
gt
,
_gt_inplace
=
_elemwise
(
scal
.
gt
,
'gt'
,
"""greater than (elemwise)"""
"""greater than (elemwise)"""
)
le
,
le_inplace
=
_elemwise
(
scal
.
le
,
'le'
)
le
,
_le_inplace
=
_elemwise
(
scal
.
le
,
'le'
,
"""less than, or equal to (elemwise)"""
"""less than, or equal to (elemwise)"""
)
ge
,
ge_inplace
=
_elemwise
(
scal
.
ge
,
'ge'
)
ge
,
_ge_inplace
=
_elemwise
(
scal
.
ge
,
'ge'
,
"""greater than, or equal to (elemwise)"""
"""greater than, or equal to (elemwise)"""
)
eq
,
eq_inplace
=
_elemwise
(
scal
.
eq
,
'eq'
)
eq
,
_eq_inplace
=
_elemwise
(
scal
.
eq
,
'eq'
,
"""equal to (elemwise)"""
"""equal to (elemwise)"""
)
neq
,
neq_inplace
=
_elemwise
(
scal
.
neq
,
'neq'
)
neq
,
_neq_inplace
=
_elemwise
(
scal
.
neq
,
'neq'
,
"""not equal to (elemwise)"""
"""not equal to (elemwise)"""
)
##########################
##########################
# Bit-wise
# Bit-wise
##########################
##########################
and_
,
and_inplace
=
_elemwise
(
scal
.
and_
,
'and_'
)
and_
,
_and_inplace
=
_elemwise
(
scal
.
and_
,
'and_'
,
"""bitwise AND (elemwise)"""
"""bitwise AND (elemwise)"""
)
or_
,
or_inplace
=
_elemwise
(
scal
.
or_
,
'or_'
)
or_
,
_or_inplace
=
_elemwise
(
scal
.
or_
,
'or_'
,
"""bitwise OR (elemwise)"""
"""bitwise OR (elemwise)"""
)
xor
,
xor_inplace
=
_elemwise
(
scal
.
xor
,
'xor'
)
xor
,
_xor_inplace
=
_elemwise
(
scal
.
xor
,
'xor'
,
"""bitwise XOR (elemwise)"""
"""bitwise XOR (elemwise)"""
)
invert
,
invert_inplace
=
_elemwise
(
scal
.
invert
,
'invert'
)
invert
,
_invert_inplace
=
_elemwise
(
scal
.
invert
,
'invert'
,
"""bitwise NOT (elemwise)"""
"""bitwise NOT (elemwise)"""
)
##########################
##########################
# Math
# Math
##########################
##########################
_abs
,
abs_inplace
=
_elemwise
(
scal
.
abs
,
'abs'
,
_abs
,
_abs_inplace
=
_elemwise
(
scal
.
abs
,
'abs'
,
"""absolute value (elemwise)"""
)
"""absolute value (elemwise)"""
)
exp
,
exp_inplace
=
_elemwise
(
scal
.
exp
,
'exp'
,
exp
,
_
exp_inplace
=
_elemwise
(
scal
.
exp
,
'exp'
,
"""exponential (elemwise)"""
)
"""exponential (elemwise)"""
)
neg
,
neg_inplace
=
_elemwise
(
scal
.
neg
,
'neg'
,
neg
,
_
neg_inplace
=
_elemwise
(
scal
.
neg
,
'neg'
,
"""negative (elemwise)"""
)
"""negative (elemwise)"""
)
inv
,
inv_inplace
=
_elemwise
(
scal
.
inv
,
'inv'
,
inv
,
_
inv_inplace
=
_elemwise
(
scal
.
inv
,
'inv'
,
"""multiplicative inverse (elemwise)"""
)
"""multiplicative inverse (elemwise)"""
)
log
,
log_inplace
=
_elemwise
(
scal
.
log
,
'log'
,
log
,
_
log_inplace
=
_elemwise
(
scal
.
log
,
'log'
,
"""logarithm base-e (elemwise)"""
)
"""logarithm base-e (elemwise)"""
)
log2
,
log2_inplace
=
_elemwise
(
scal
.
log2
,
'log2'
)
log2
,
_log2_inplace
=
_elemwise
(
scal
.
log2
,
'log2'
,
"""logarithm base-2 (elemwise)"""
"""logarithm base-2 (elemwise)"""
)
sgn
,
sgn_inplace
=
_elemwise
(
scal
.
sgn
,
'sgn'
)
sgn
,
_sgn_inplace
=
_elemwise
(
scal
.
sgn
,
'sgn'
,
"""sign (elemwise)"""
"""sign (elemwise)"""
)
sqr
,
sqr_inplace
=
_elemwise
(
scal
.
sqr
,
'sqr'
)
sqr
,
_sqr_inplace
=
_elemwise
(
scal
.
sqr
,
'sqr'
,
"""square (elemwise)"""
"""square (elemwise)"""
)
sqrt
,
sqrt_inplace
=
_elemwise
(
scal
.
sqrt
,
'sqrt'
)
sqrt
,
_sqrt_inplace
=
_elemwise
(
scal
.
sqrt
,
'sqrt'
,
"""square root (elemwise)"""
"""square root (elemwise)"""
)
cos
,
cos_inplace
=
_elemwise
(
scal
.
cos
,
'cos'
)
cos
,
_cos_inplace
=
_elemwise
(
scal
.
cos
,
'cos'
,
"""cosine (elemwise)"""
"""cosine (elemwise)"""
)
sin
,
sin_inplace
=
_elemwise
(
scal
.
sin
,
'sin'
)
sin
,
_sin_inplace
=
_elemwise
(
scal
.
sin
,
'sin'
,
"""sine (elemwise)"""
"""sine (elemwise)"""
)
tan
,
tan_inplace
=
_elemwise
(
scal
.
tan
,
'tan'
)
tan
,
_tan_inplace
=
_elemwise
(
scal
.
tan
,
'tan'
,
"""tan = sin/cos (elemwise)"""
"""tan = sin/cos (elemwise)"""
)
cosh
,
cosh_inplace
=
_elemwise
(
scal
.
cosh
,
'cosh'
)
cosh
,
_cosh_inplace
=
_elemwise
(
scal
.
cosh
,
'cosh'
,
"""hyperbolic cosine (elemwise)"""
"""hyperbolic cosine (elemwise)"""
)
sinh
,
sinh_inplace
=
_elemwise
(
scal
.
sinh
,
'sinh'
)
sinh
,
_sinh_inplace
=
_elemwise
(
scal
.
sinh
,
'sinh'
,
"""hyperbolic sine (elemwise)"""
"""hyperbolic sine (elemwise)"""
)
tanh
,
tanh_inplace
=
_elemwise
(
scal
.
tanh
,
'tanh'
)
tanh
,
_tanh_inplace
=
_elemwise
(
scal
.
tanh
,
'tanh'
,
"""hyperbolic tangent (elemwise)"""
"""hyperbolic tangent (elemwise)"""
)
##########################
##########################
# Misc
# Misc
##########################
##########################
fill
,
fill_inplace
=
_elemwise
(
scal
.
second
,
'fill'
)
fill
,
_fill_inplace
=
_elemwise
(
scal
.
second
,
'fill'
,
"""fill WRITEME (elemwise)"""
"""fill WRITEME (elemwise)"""
)
@constructor
@constructor
def
ones_like
(
model
):
def
ones_like
(
model
):
...
@@ -870,22 +870,23 @@ repeat = Repeat()
...
@@ -870,22 +870,23 @@ repeat = Repeat()
# Arithmetics
# Arithmetics
##########################
##########################
add
,
add_inplace
=
_elemwise
(
scal
.
add
,
'add'
,
'addition (element-wise)'
)
add
,
_add_inplace
=
_elemwise
(
scal
.
add
,
'add'
,
"""addition (element-wise)"""
)
sub
,
sub_inplace
=
_elemwise
(
scal
.
sub
,
'sub'
)
sub
,
_sub_inplace
=
_elemwise
(
scal
.
sub
,
'sub'
,
"""subtraction (elemwise)"""
"""subtraction (elemwise)"""
)
mul
,
mul_inplace
=
_elemwise
(
scal
.
mul
,
'mul'
)
mul
,
_mul_inplace
=
_elemwise
(
scal
.
mul
,
'mul'
,
"""multiplication (elemwise)"""
"""multiplication (elemwise)"""
)
div
,
div_inplace
=
_elemwise
(
scal
.
div
,
'div'
)
div
,
_div_inplace
=
_elemwise
(
scal
.
div
,
'div'
,
"""division (elemwise)"""
"""division (elemwise)"""
)
mod
,
mod_inplace
=
_elemwise
(
scal
.
mod
,
'mod'
)
mod
,
_mod_inplace
=
_elemwise
(
scal
.
mod
,
'mod'
,
"""modulo (elemwise)"""
"""modulo (elemwise)"""
)
pow
,
pow_inplace
=
_elemwise
(
scal
.
pow
,
'pow'
)
pow
,
_pow_inplace
=
_elemwise
(
scal
.
pow
,
'pow'
,
"""raise to given exponent (elemwise)"""
"""raise to given exponent (elemwise)"""
)
##########################
##########################
...
@@ -917,12 +918,12 @@ class TransposeInplace(Op):
...
@@ -917,12 +918,12 @@ class TransposeInplace(Op):
def
__str__
(
self
):
def
__str__
(
self
):
return
"TransposeView"
return
"TransposeView"
transpose_inplace
=
TransposeInplace
()
_
transpose_inplace
=
TransposeInplace
()
"""WRITEME"""
"""WRITEME"""
def
transpose
(
x
,
**
kwargs
):
def
transpose
(
x
,
**
kwargs
):
"""WRITEME"""
"""WRITEME"""
return
transpose_inplace
(
tensor_copy
(
x
),
**
kwargs
)
return
_
transpose_inplace
(
tensor_copy
(
x
),
**
kwargs
)
...
...
tensor_opt.py
浏览文件 @
8c745911
...
@@ -20,7 +20,7 @@ def in2out(*local_opts):
...
@@ -20,7 +20,7 @@ def in2out(*local_opts):
# gemm: (d,a,b,c,s) -> d = d*s + a*dot(b,c)
# gemm: (d,a,b,c,s) -> d = d*s + a*dot(b,c)
# Transforms d -= a * dot(b, c) into gemm(d, -a, b, c, 1.0)
# Transforms d -= a * dot(b, c) into gemm(d, -a, b, c, 1.0)
gemm_pattern_1
=
gof
.
PatternSub
((
T
.
sub_inplace
,
gemm_pattern_1
=
gof
.
PatternSub
((
T
.
_
sub_inplace
,
'd'
,
'd'
,
(
T
.
mul
,
(
T
.
mul
,
dict
(
pattern
=
(
T
.
DimShuffle
((),
[
'x'
,
'x'
],
inplace
=
True
),
'a'
),
dict
(
pattern
=
(
T
.
DimShuffle
((),
[
'x'
,
'x'
],
inplace
=
True
),
'a'
),
...
...
编写
预览
Markdown
格式
0%
重试
或
添加新文件
添加附件
取消
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!
取消
请
注册
或者
登录
后发表评论