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
4e36f43f
提交
4e36f43f
authored
9月 30, 2011
作者:
David Warde-Farley
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
STY: various blas/test_blas fixes
上级
a8646fdc
隐藏空白字符变更
内嵌
并排
正在显示
2 个修改的文件
包含
52 行增加
和
34 行删除
+52
-34
blas.py
theano/tensor/blas.py
+28
-13
test_blas.py
theano/tensor/tests/test_blas.py
+24
-21
没有找到文件。
theano/tensor/blas.py
浏览文件 @
4e36f43f
...
@@ -1206,13 +1206,16 @@ class GemmOptimizer(Optimizer):
...
@@ -1206,13 +1206,16 @@ class GemmOptimizer(Optimizer):
try
:
try
:
env
.
replace_all_validate
(
env
.
replace_all_validate
(
zip
(
node
.
outputs
,
new_outputs
),
zip
(
node
.
outputs
,
new_outputs
),
reason
=
'GemmOptimizer'
)
reason
=
'GemmOptimizer'
)
did_something
=
True
did_something
=
True
break
break
except
InconsistencyError
,
e
:
except
InconsistencyError
,
e
:
#TODO: retry other applications of gemm (see comment in _gemm_from_node
# TODO: retry other applications of gemm (see comment
# in _gemm_from_node
pass
pass
class
Dot22
(
GemmRelated
):
class
Dot22
(
GemmRelated
):
"""Compute a matrix-matrix product.
"""Compute a matrix-matrix product.
This is a specialization of the more general Dot()
This is a specialization of the more general Dot()
...
@@ -1227,7 +1230,7 @@ class Dot22(GemmRelated):
...
@@ -1227,7 +1230,7 @@ class Dot22(GemmRelated):
raise
TypeError
(
'dtype mismatch to Dot22'
)
raise
TypeError
(
'dtype mismatch to Dot22'
)
bz
=
(
x
.
type
.
broadcastable
[
0
],
y
.
type
.
broadcastable
[
1
])
bz
=
(
x
.
type
.
broadcastable
[
0
],
y
.
type
.
broadcastable
[
1
])
outputs
=
[
T
.
tensor
(
x
.
type
.
dtype
,
bz
)]
outputs
=
[
T
.
tensor
(
x
.
type
.
dtype
,
bz
)]
return
Apply
(
self
,
[
x
,
y
],
outputs
)
return
Apply
(
self
,
[
x
,
y
],
outputs
)
def
perform
(
self
,
node
,
inp
,
out
):
def
perform
(
self
,
node
,
inp
,
out
):
x
,
y
=
inp
x
,
y
=
inp
...
@@ -1235,9 +1238,11 @@ class Dot22(GemmRelated):
...
@@ -1235,9 +1238,11 @@ class Dot22(GemmRelated):
try
:
try
:
z
[
0
]
=
numpy
.
asarray
(
numpy
.
dot
(
x
,
y
))
z
[
0
]
=
numpy
.
asarray
(
numpy
.
dot
(
x
,
y
))
except
ValueError
,
e
:
except
ValueError
,
e
:
# The error raised by numpy has no shape information, we mean to add that
# The error raised by numpy has no shape information, we mean to
# add that
e
.
args
=
e
.
args
+
(
x
.
shape
,
y
.
shape
)
e
.
args
=
e
.
args
+
(
x
.
shape
,
y
.
shape
)
raise
raise
def
__str__
(
self
):
def
__str__
(
self
):
return
"_dot22"
return
"_dot22"
...
@@ -1250,10 +1255,12 @@ class Dot22(GemmRelated):
...
@@ -1250,10 +1255,12 @@ class Dot22(GemmRelated):
npy_intp dims[2];
npy_intp dims[2];
dims[0] =
%(_x)
s->dimensions[0];
dims[0] =
%(_x)
s->dimensions[0];
dims[1] =
%(_y)
s->dimensions[1];
dims[1] =
%(_y)
s->dimensions[1];
%(_zout)
s = (PyArrayObject*)PyArray_SimpleNew(2, dims, type_num_
%(_x)
s);
%(_zout)
s = (PyArrayObject*)PyArray_SimpleNew(2, dims,
type_num_
%(_x)
s);
//fprintf(stderr, "Dot Allocating
%%
i
%%
i
\\
n", dims[0], dims[1]);
//fprintf(stderr, "Dot Allocating
%%
i
%%
i
\\
n", dims[0], dims[1]);
if(!
%(_zout)
s) {
if(!
%(_zout)
s) {
PyErr_SetString(PyExc_MemoryError, "failed to alloc dot22 output");
PyErr_SetString(PyExc_MemoryError,
"failed to alloc dot22 output");
%(fail)
s
%(fail)
s
}
}
}
}
...
@@ -1270,16 +1277,19 @@ class Dot22(GemmRelated):
...
@@ -1270,16 +1277,19 @@ class Dot22(GemmRelated):
double a = 1.0;
double a = 1.0;
double b = 0.0;
double b = 0.0;
"""
"""
def
c_code
(
self
,
node
,
name
,
inp
,
out
,
sub
):
#DEBUG
def
c_code
(
self
,
node
,
name
,
inp
,
out
,
sub
):
# DEBUG
_x
,
_y
=
inp
_x
,
_y
=
inp
_zout
,
=
out
_zout
,
=
out
if
node
.
inputs
[
0
]
.
type
.
dtype
.
startswith
(
'complex'
):
if
node
.
inputs
[
0
]
.
type
.
dtype
.
startswith
(
'complex'
):
raise
utils
.
MethodNotDefined
(
'
%
s.c_code'
\
raise
utils
.
MethodNotDefined
(
'
%
s.c_code'
\
%
self
.
__class__
.
__name__
)
%
self
.
__class__
.
__name__
)
if
len
(
self
.
c_libraries
())
<=
0
:
if
len
(
self
.
c_libraries
())
<=
0
:
return
super
(
Dot22
,
self
)
.
c_code
(
node
,
name
,
(
_x
,
_y
),
(
_zout
,
),
sub
)
return
super
(
Dot22
,
self
)
.
c_code
(
node
,
name
,
(
_x
,
_y
),
(
_zout
,
),
sub
)
full_code
=
self
.
build_gemm_call
()
%
dict
(
locals
(),
**
sub
)
full_code
=
self
.
build_gemm_call
()
%
dict
(
locals
(),
**
sub
)
return
full_code
return
full_code
def
c_code_cache_version
(
self
):
def
c_code_cache_version
(
self
):
gv
=
self
.
build_gemm_version
()
gv
=
self
.
build_gemm_version
()
if
gv
:
if
gv
:
...
@@ -1436,6 +1446,7 @@ optdb.register('InplaceBlasOpt',
...
@@ -1436,6 +1446,7 @@ optdb.register('InplaceBlasOpt',
blas_opt_inplace
,
blas_opt_inplace
,
70.0
,
'fast_run'
,
'inplace'
)
70.0
,
'fast_run'
,
'inplace'
)
class
Dot22Scalar
(
GemmRelated
):
class
Dot22Scalar
(
GemmRelated
):
"""Compute a matrix-matrix product.
"""Compute a matrix-matrix product.
This is a specialization of the more general Dot()
This is a specialization of the more general Dot()
...
@@ -1473,6 +1484,7 @@ class Dot22Scalar(GemmRelated):
...
@@ -1473,6 +1484,7 @@ class Dot22Scalar(GemmRelated):
# The error raised by numpy has no shape information, we mean to add that
# The error raised by numpy has no shape information, we mean to add that
e
.
args
=
e
.
args
+
(
x
.
shape
,
y
.
shape
)
e
.
args
=
e
.
args
+
(
x
.
shape
,
y
.
shape
)
raise
raise
def
__str__
(
self
):
def
__str__
(
self
):
return
"_dot22scalar"
return
"_dot22scalar"
...
@@ -1492,6 +1504,7 @@ class Dot22Scalar(GemmRelated):
...
@@ -1492,6 +1504,7 @@ class Dot22Scalar(GemmRelated):
#undef REAL
#undef REAL
float b = 0.0;
float b = 0.0;
"""
"""
case_double_ab_constants
=
"""
case_double_ab_constants
=
"""
#define REAL double
#define REAL double
double a = (
%(_a)
s->descr->type_num == PyArray_FLOAT)
double a = (
%(_a)
s->descr->type_num == PyArray_FLOAT)
...
@@ -1500,16 +1513,18 @@ class Dot22Scalar(GemmRelated):
...
@@ -1500,16 +1513,18 @@ class Dot22Scalar(GemmRelated):
#undef REAL
#undef REAL
double b = 0.0;
double b = 0.0;
"""
"""
def
c_code
(
self
,
node
,
name
,
inp
,
out
,
sub
):
#DEBUG
def
c_code
(
self
,
node
,
name
,
inp
,
out
,
sub
):
#DEBUG
_x
,
_y
,
_a
=
inp
_x
,
_y
,
_a
=
inp
_zout
,
=
out
_zout
,
=
out
if
node
.
inputs
[
0
]
.
type
.
dtype
.
startswith
(
'complex'
):
if
node
.
inputs
[
0
]
.
type
.
dtype
.
startswith
(
'complex'
):
raise
utils
.
MethodNotDefined
(
'
%
s.c_code'
\
raise
utils
.
MethodNotDefined
(
'
%
s.c_code'
\
%
self
.
__class__
.
__name__
)
%
self
.
__class__
.
__name__
)
if
len
(
self
.
c_libraries
())
<=
0
:
if
len
(
self
.
c_libraries
())
<=
0
:
return
super
(
Dot22Scalar
,
self
)
.
c_code
(
node
,
name
,
(
_x
,
_y
),
(
_zout
,
),
sub
)
return
super
(
Dot22Scalar
,
self
)
.
c_code
(
node
,
name
,
(
_x
,
_y
),
(
_zout
,
),
sub
)
full_code
=
self
.
build_gemm_call
()
%
dict
(
locals
(),
**
sub
)
full_code
=
self
.
build_gemm_call
()
%
dict
(
locals
(),
**
sub
)
return
full_code
return
full_code
def
c_code_cache_version
(
self
):
def
c_code_cache_version
(
self
):
gv
=
self
.
build_gemm_version
()
gv
=
self
.
build_gemm_version
()
if
gv
:
if
gv
:
...
@@ -1558,10 +1573,10 @@ def local_dot22_to_dot22scalar(node):
...
@@ -1558,10 +1573,10 @@ def local_dot22_to_dot22scalar(node):
for
i
,
x
in
enumerate
(
m
.
owner
.
inputs
):
for
i
,
x
in
enumerate
(
m
.
owner
.
inputs
):
if
_as_scalar
(
x
)
and
(
theano
.
scalar
.
upcast
(
x
.
type
.
dtype
,
d
.
type
.
dtype
)
if
_as_scalar
(
x
)
and
(
theano
.
scalar
.
upcast
(
x
.
type
.
dtype
,
d
.
type
.
dtype
)
==
d
.
type
.
dtype
):
==
d
.
type
.
dtype
):
scalar_idx
=
i
scalar_idx
=
i
break
break
if
scalar_idx
<
0
:
if
scalar_idx
<
0
:
_logger
.
info
(
'Not optimizing dot22 with inputs
%
s
%
s, as the type '
_logger
.
info
(
'Not optimizing dot22 with inputs
%
s
%
s, as the type '
'of the scalar cannot be upcasted to the matrix type'
,
'of the scalar cannot be upcasted to the matrix type'
,
node
.
inputs
,
[
x
.
type
for
x
in
node
.
inputs
])
node
.
inputs
,
[
x
.
type
for
x
in
node
.
inputs
])
...
@@ -1593,7 +1608,7 @@ def local_dot22_to_dot22scalar(node):
...
@@ -1593,7 +1608,7 @@ def local_dot22_to_dot22scalar(node):
==
d
.
type
.
dtype
)):
==
d
.
type
.
dtype
)):
scalar_idx
=
i
scalar_idx
=
i
break
break
if
scalar_idx
<
0
:
if
scalar_idx
<
0
:
_logger
.
info
(
'Not optimizing dot22 with inputs
%
s
%
s, as the type '
_logger
.
info
(
'Not optimizing dot22 with inputs
%
s
%
s, as the type '
'of the scalar cannot be upcasted to the matrix type'
,
'of the scalar cannot be upcasted to the matrix type'
,
node
.
inputs
,
[
x
.
type
for
x
in
node
.
inputs
])
node
.
inputs
,
[
x
.
type
for
x
in
node
.
inputs
])
...
...
theano/tensor/tests/test_blas.py
浏览文件 @
4e36f43f
...
@@ -5,7 +5,8 @@ import theano.tensor as T
...
@@ -5,7 +5,8 @@ import theano.tensor as T
#from theano.gof import Env
#from theano.gof import Env
from
theano.printing
import
pp
from
theano.printing
import
pp
import
numpy
,
theano
import
numpy
import
theano
from
numpy
import
(
arange
,
array
,
common_type
,
complex64
,
complex128
,
float32
,
from
numpy
import
(
arange
,
array
,
common_type
,
complex64
,
complex128
,
float32
,
float64
,
newaxis
,
shape
,
transpose
,
zeros
)
float64
,
newaxis
,
shape
,
transpose
,
zeros
)
from
numpy.testing
import
assert_
,
assert_array_almost_equal
from
numpy.testing
import
assert_
,
assert_array_almost_equal
...
@@ -13,10 +14,11 @@ from numpy.testing import assert_, assert_array_almost_equal
...
@@ -13,10 +14,11 @@ from numpy.testing import assert_, assert_array_almost_equal
#from numpy.testing.noseclasses import KnownFailureTest
#from numpy.testing.noseclasses import KnownFailureTest
#from theano.tensor.blas import *
#from theano.tensor.blas import *
from
theano.tensor.blas
import
(
_dot22
,
_dot22scalar
,
res_is_a
,
_as_scalar
,
_is_real_matrix
,
from
theano.tensor.blas
import
(
_dot22
,
_dot22scalar
,
res_is_a
,
_as_scalar
,
_gemm_canonicalize
,
_factor_canonicalized
,
Gemm
,
Gemv
,
gemm_inplace
,
gemm_no_inplace
,
_is_real_matrix
,
_gemm_canonicalize
,
InconsistencyError
,
_factor_canonicalized
,
Gemm
,
Gemv
,
Ger
,
ger
,
ger_destructive
)
gemm_inplace
,
gemm_no_inplace
,
InconsistencyError
,
Ger
,
ger
,
ger_destructive
)
from
unittest
import
TestCase
from
unittest
import
TestCase
from
theano.tests
import
unittest_tools
from
theano.tests
import
unittest_tools
from
copy
import
copy
,
deepcopy
from
copy
import
copy
,
deepcopy
...
@@ -29,7 +31,8 @@ import theano.tensor.blas_scipy
...
@@ -29,7 +31,8 @@ import theano.tensor.blas_scipy
if
config
.
mode
==
'FAST_COMPILE'
:
if
config
.
mode
==
'FAST_COMPILE'
:
mode_not_fast_compile
=
'FAST_RUN'
mode_not_fast_compile
=
'FAST_RUN'
else
:
mode_not_fast_compile
=
config
.
mode
else
:
mode_not_fast_compile
=
config
.
mode
mode_blas_opt
=
theano
.
compile
.
get_default_mode
()
.
including
(
'BlasOpt'
,
'specialize'
)
mode_blas_opt
=
theano
.
compile
.
get_default_mode
()
.
including
(
'BlasOpt'
,
'specialize'
)
...
@@ -675,9 +678,9 @@ def test_inplace1():
...
@@ -675,9 +678,9 @@ def test_inplace1():
def
test_dot22
():
def
test_dot22
():
for
dtype1
in
[
'float32'
,
'float64'
,
'complex64'
,
'complex128'
]:
for
dtype1
in
[
'float32'
,
'float64'
,
'complex64'
,
'complex128'
]:
a
=
T
.
matrix
(
dtype
=
dtype1
)
a
=
T
.
matrix
(
dtype
=
dtype1
)
for
dtype2
in
[
'float32'
,
'float64'
,
'complex64'
,
'complex128'
]:
for
dtype2
in
[
'float32'
,
'float64'
,
'complex64'
,
'complex128'
]:
b
=
T
.
matrix
(
dtype
=
dtype2
)
b
=
T
.
matrix
(
dtype
=
dtype2
)
f
=
theano
.
function
([
a
,
b
],
T
.
dot
(
a
,
b
),
mode
=
mode_blas_opt
)
f
=
theano
.
function
([
a
,
b
],
T
.
dot
(
a
,
b
),
mode
=
mode_blas_opt
)
topo
=
f
.
maker
.
env
.
toposort
()
topo
=
f
.
maker
.
env
.
toposort
()
if
dtype1
==
dtype2
:
if
dtype1
==
dtype2
:
...
@@ -691,12 +694,12 @@ def test_dot22():
...
@@ -691,12 +694,12 @@ def test_dot22():
bv
=
rng
.
uniform
(
size
=
b_shp
)
.
astype
(
dtype2
)
bv
=
rng
.
uniform
(
size
=
b_shp
)
.
astype
(
dtype2
)
f
(
av
,
bv
)
f
(
av
,
bv
)
cmp
((
3
,
4
),(
4
,
5
))
cmp
((
3
,
4
),
(
4
,
5
))
cmp
((
0
,
4
),(
4
,
5
))
cmp
((
0
,
4
),
(
4
,
5
))
cmp
((
3
,
0
),(
0
,
5
))
cmp
((
3
,
0
),
(
0
,
5
))
cmp
((
3
,
4
),(
4
,
0
))
cmp
((
3
,
4
),
(
4
,
0
))
cmp
((
0
,
4
),(
4
,
0
))
cmp
((
0
,
4
),
(
4
,
0
))
cmp
((
0
,
0
),(
0
,
0
))
cmp
((
0
,
0
),
(
0
,
0
))
def
test_dot22scalar
():
def
test_dot22scalar
():
## including does not seem to work for 'local_dot_to_dot22' and
## including does not seem to work for 'local_dot_to_dot22' and
...
@@ -706,11 +709,11 @@ def test_dot22scalar():
...
@@ -706,11 +709,11 @@ def test_dot22scalar():
#m = theano.compile.get_default_mode().including('BlasOpt', 'specialize')
#m = theano.compile.get_default_mode().including('BlasOpt', 'specialize')
rng
=
numpy
.
random
.
RandomState
(
unittest_tools
.
fetch_seed
())
rng
=
numpy
.
random
.
RandomState
(
unittest_tools
.
fetch_seed
())
for
dtype1
in
[
'complex64'
,
'complex128'
]:
for
dtype1
in
[
'complex64'
,
'complex128'
]:
a
=
T
.
matrix
(
'a'
,
dtype
=
dtype1
)
a
=
T
.
matrix
(
'a'
,
dtype
=
dtype1
)
for
dtype2
in
[
'complex64'
,
'complex128'
]:
for
dtype2
in
[
'complex64'
,
'complex128'
]:
b
=
T
.
matrix
(
'b'
,
dtype
=
dtype2
)
b
=
T
.
matrix
(
'b'
,
dtype
=
dtype2
)
for
dtype3
in
[
'complex64'
,
'complex128'
]:
for
dtype3
in
[
'complex64'
,
'complex128'
]:
c
=
T
.
matrix
(
'c'
,
dtype
=
dtype3
)
c
=
T
.
matrix
(
'c'
,
dtype
=
dtype3
)
for
dtype4
in
[
'complex64'
,
'complex128'
]:
for
dtype4
in
[
'complex64'
,
'complex128'
]:
cst
=
theano
.
tensor
.
basic
.
constant
(
.
2
,
dtype
=
dtype4
)
cst
=
theano
.
tensor
.
basic
.
constant
(
.
2
,
dtype
=
dtype4
)
cst2
=
theano
.
tensor
.
basic
.
constant
(
.
1
,
dtype
=
dtype4
)
cst2
=
theano
.
tensor
.
basic
.
constant
(
.
1
,
dtype
=
dtype4
)
...
@@ -978,11 +981,11 @@ def matrixmultiply(a, b):
...
@@ -978,11 +981,11 @@ def matrixmultiply(a, b):
class
BaseGemv
(
object
):
class
BaseGemv
(
object
):
def
get_data
(
self
,
x_stride
=
1
,
y_stride
=
1
):
def
get_data
(
self
,
x_stride
=
1
,
y_stride
=
1
):
rng
=
numpy
.
random
.
RandomState
(
unittest_tools
.
fetch_seed
())
rng
=
numpy
.
random
.
RandomState
(
unittest_tools
.
fetch_seed
())
mult
=
array
(
1
,
dtype
=
self
.
dtype
)
mult
=
array
(
1
,
dtype
=
self
.
dtype
)
if
self
.
dtype
in
[
complex64
,
complex128
]:
if
self
.
dtype
in
[
complex64
,
complex128
]:
mult
=
array
(
1
+
1
j
,
dtype
=
self
.
dtype
)
mult
=
array
(
1
+
1
j
,
dtype
=
self
.
dtype
)
alpha
=
array
(
1.
,
dtype
=
self
.
dtype
)
*
mult
alpha
=
array
(
1.
,
dtype
=
self
.
dtype
)
*
mult
beta
=
array
(
1.
,
dtype
=
self
.
dtype
)
*
mult
beta
=
array
(
1.
,
dtype
=
self
.
dtype
)
*
mult
a
=
rng
.
randn
(
3
,
3
)
.
astype
(
self
.
dtype
)
*
mult
a
=
rng
.
randn
(
3
,
3
)
.
astype
(
self
.
dtype
)
*
mult
x
=
arange
(
shape
(
a
)[
0
]
*
x_stride
,
dtype
=
self
.
dtype
)
*
mult
x
=
arange
(
shape
(
a
)[
0
]
*
x_stride
,
dtype
=
self
.
dtype
)
*
mult
y
=
arange
(
shape
(
a
)[
1
]
*
y_stride
,
dtype
=
self
.
dtype
)
*
mult
y
=
arange
(
shape
(
a
)[
1
]
*
y_stride
,
dtype
=
self
.
dtype
)
*
mult
...
...
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