Make theano.tensor.* interface more similar to numpy.* #1916

Open
nouiz opened this Issue Jun 13, 2014 · 20 comments

Comments

Projects
None yet
9 participants
@nouiz
Member

nouiz commented Jun 13, 2014

Todo list, what need to be done.
Incomplete list of missing implementation.:

@dwf

This comment has been minimized.

Show comment
Hide comment
@dwf

dwf Nov 25, 2014

Member

The first stage of this ticket is to create the todo list, looking through NumPy's interface and Theano's interface and noting missing functions in the latter.

Member

dwf commented Nov 25, 2014

The first stage of this ticket is to create the todo list, looking through NumPy's interface and Theano's interface and noting missing functions in the latter.

@vdumoulin

This comment has been minimized.

Show comment
Hide comment
@vdumoulin

vdumoulin Nov 25, 2014

Contributor

@yingzha this is your new ticket.

Contributor

vdumoulin commented Nov 25, 2014

@yingzha this is your new ticket.

@yingzha

This comment has been minimized.

Show comment
Hide comment
@yingzha

yingzha Dec 2, 2014

Contributor
numpy.*                   theano.tensor.* interface compatible
ALLOW_THREADS             no              no
BUFSIZE                   no              no
CLIP                      no              no
ComplexWarning            no              no
DataSource                no              no
ERR_CALL                  no              no
ERR_DEFAULT               no              no
ERR_DEFAULT2              no              no
ERR_IGNORE                no              no
ERR_LOG                   no              no
ERR_PRINT                 no              no
ERR_RAISE                 no              no
ERR_WARN                  no              no
FLOATING_POINT_SUPPORT    no              no
FPE_DIVIDEBYZERO          no              no
FPE_INVALID               no              no
FPE_OVERFLOW              no              no
FPE_UNDERFLOW             no              no
False_                    no              no
Inf                       no              no
Infinity                  no              no
MAXDIMS                   no              no
MachAr                    no              no
ModuleDeprecationWarning  no              no
NAN                       no              no
NINF                      no              no
NZERO                     no              no
NaN                       no              no
PINF                      no              no
PZERO                     no              no
PackageLoader             no              no
RAISE                     no              no
RankWarning               no              no
SHIFT_DIVIDEBYZERO        no              no
SHIFT_INVALID             no              no
SHIFT_OVERFLOW            no              no
SHIFT_UNDERFLOW           no              no
ScalarType                no              no
Tester                    no              no
True_                     no              no
UFUNC_BUFSIZE_DEFAULT     no              no
UFUNC_PYVALS_NAME         no              no
WRAP                      no              no
__NUMPY_SETUP__           no              no
__all__                   no              no
__builtins__              yes
__config__                no              no
__doc__                   yes
__file__                  yes
__git_revision__          no              no
__name__                  yes
__package__               yes
__path__                  yes
__version__               no              no
_import_tools             no              no
_mat                      no              no
_old_rtld                 no              no
abs                       no              no
absolute                  no              no
absolute_import           no              no
add                       yes
add_docstring             no              no
add_newdoc                no              no
add_newdoc_ufunc          no              no
add_newdocs               no              no
alen                      no              no
all                       yes
allclose                  no              no
alltrue                   no              no
alterdot                  no              no
amax                      no              no
amin                      no              no
angle                     yes
any                       yes
append                    no              no
apply_along_axis          no              no
apply_over_axes           no              no
arange                    yes
arccos                    yes
arccosh                   yes
arcsin                    yes
arcsinh                   yes
arctan                    yes
arctan2                   yes
arctanh                   yes
argmax                    yes
argmin                    yes
argpartition              no              no
argsort                   yes
argwhere                  no              no
around                    no              no
array                     no              no
array2string              no              no
array_equal               no              no
array_equiv               no              no
array_repr                no              no
array_split               no              no
array_str                 no              no
asanyarray                no              no
asarray                   no              no
asarray_chkfinite         no              no
ascontiguousarray         no              no
asfarray                  no              no
asfortranarray            no              no
asmatrix                  no              no
asscalar                  no              no
atleast_1d                no              no
atleast_2d                no              no
atleast_3d                no              no
average                   no              no
bartlett                  yes
base_repr                 no              no
bench                     no              no
binary_repr               no              no
bincount                  yes
bitwise_and               yes
bitwise_not               yes
bitwise_or                yes
bitwise_xor               yes
blackman                  no              no
bmat                      no              no
bool                      no              no
bool8                     no              no
bool_                     no              no
broadcast                 no              no
broadcast_arrays          no              no
busday_count              no              no
busday_offset             no              no
busdaycalendar            no              no
byte                      no              no
byte_bounds               no              no
bytes_                    no              no
c_                        no              no
can_cast                  no              no
cast                      yes
cdouble                   no              no
ceil                      yes
cfloat                    no              no
char                      no              no
character                 no              no
chararray                 no              no
choose                    yes
clip                      yes
clongdouble               no              no
clongfloat                no              no
column_stack              no              no
common_type               no              no
compare_chararrays        no              no
compat                    no              no
complex                   yes
complex128                no              no
complex256                no              no
complex64                 no              no
complex_                  no              no
complexfloating           no              no
compress                  no              no
concatenate               yes
conj                      yes
conjugate                 no              no
convolve                  no              no
copy                      yes
copysign                  no              no
copyto                    no              no
core                      no              no
corrcoef                  no              no
correlate                 no              no
cos                       yes
cosh                      yes
count_nonzero             no              no
cov                       no              no
cross                     no              no
csingle                   no              no
ctypeslib                 no              no
cumprod                   yes
cumproduct                no              no
cumsum                    yes
datetime64                no              no
datetime_as_string        no              no
datetime_data             no              no
deg2rad                   yes
degrees                   no              no
delete                    no              no
deprecate                 no              no
deprecate_with_doc        no              no
diag                      yes
diag_indices              no              no
diag_indices_from         no              no
diagflat                  no              no
diagonal                  yes
diff                      no              no
digitize                  no              no
disable_rtld_global       no              no
disp                      no              no
distutils                 no              no
divide                    no              no
division                  no              no
dot                       yes
double                    no              no
dsplit                    no              no
dstack                    no              no
dtype                     no              no
dual                      no              no
e                         no              no
ediff1d                   no              no
einsum                    no              no
emath                     no              no
empty                     no              no
empty_like                no              no
enable_rtld_global        no              no
equal                     no              no
errstate                  no              no
euler_gamma               no              no
exp                       yes
exp2                      yes
expand_dims               no              no
expm1                     yes
extract                   no              no
eye                       yes
fabs                      no              no
fastCopyAndTranspose      no              no
fft                       no              no
fill_diagonal             yes
find_common_type          no              no
finfo                     no              no
fix                       no              no
flatiter                  no              no
flatnonzero               yes
flexible                  no              no
fliplr                    no              no
flipud                    no              no
float                     no              no
float128                  no              no
float16                   no              no
float32                   no              no
float64                   no              no
float_                    no              no
floating                  no              no
floor                     yes
floor_divide              no              no
fmax                      no              no
fmin                      no              no
fmod                      no              no
format_parser             no              no
frexp                     no              no
frombuffer                no              no
fromfile                  no              no
fromfunction              no              no
fromiter                  no              no
frompyfunc                no              no
fromregex                 no              no
fromstring                no              no
full                      no              no
full_like                 no              no
fv                        no              no
generic                   no              no
genfromtxt                no              no
get_array_wrap            no              no
get_include               no              no
get_numarray_include      no              no
get_printoptions          no              no
getbuffer                 no              no
getbufsize                no              no
geterr                    no              no
geterrcall                no              no
geterrobj                 no              no
gradient                  no              no
greater                   no              no
greater_equal             no              no
half                      no              no
hamming                   no              no
hanning                   no              no
histogram                 no              no
histogram2d               no              no
histogramdd               no              no
hsplit                    no              no
hstack                    no              no
hypot                     no              no
i0                        no              no
identity                  no              no
iinfo                     no              no
imag                      yes
in1d                      no              no
index_exp                 no              no
indices                   no              no
inexact                   no              no
inf                       no              no
info                      no              no
infty                     no              no
inner                     no              no
insert                    no              no
int                       no              no
int0                      no              no
int16                     no              no
int32                     no              no
int64                     no              no
int8                      no              no
int_                      no              no
int_asbuffer              no              no
intc                      no              no
integer                   no              no
interp                    no              no
intersect1d               no              no
intp                      no              no
invert                    yes
ipmt                      no              no
irr                       no              no
is_busday                 no              no
isclose                   no              no
iscomplex                 no              no
iscomplexobj              no              no
isfinite                  no              no
isfortran                 no              no
isinf                     yes
isnan                     yes
isneginf                  no              no
isposinf                  no              no
isreal                    no              no
isrealobj                 no              no
isscalar                  no              no
issctype                  no              no
issubclass_               no              no
issubdtype                no              no
issubsctype               no              no
iterable                  no              no
ix_                       no              no
kaiser                    no              no
kron                      no              no
ldexp                     no              no
left_shift                no              no
less                      no              no
less_equal                no              no
lexsort                   no              no
lib                       no              no
linalg                    no              no
linspace                  no              no
little_endian             no              no
load                      yes
loads                     no              no
loadtxt                   no              no
log                       yes
log10                     yes
log1p                     yes
log2                      yes
logaddexp                 no              no
logaddexp2                no              no
logical_and               no              no
logical_not               no              no
logical_or                no              no
logical_xor               no              no
logspace                  no              no
long                      no              no
longcomplex               no              no
longdouble                no              no
longfloat                 no              no
longlong                  no              no
lookfor                   no              no
ma                        no              no
mafromtxt                 no              no
mask_indices              no              no
mat                       no              no
math                      no              no
matrix                    yes
matrixlib                 no              no
max                       yes
maximum                   yes
maximum_sctype            no              no
may_share_memory          no              no
mean                      yes
median                    no              no
memmap                    no              no
meshgrid                  no              no
mgrid                     no              no
min                       yes
min_scalar_type           no              no
minimum                   yes
mintypecode               no              no
mirr                      no              no
mod                       yes
modf                      no              no
msort                     no              no
multiply                  no              no
nan                       no              no
nan_to_num                no              no
nanargmax                 no              no
nanargmin                 no              no
nanmax                    no              no
nanmean                   no              no
nanmin                    no              no
nanstd                    no              no
nansum                    no              no
nanvar                    no              no
nbytes                    no              no
ndarray                   no              no
ndenumerate               no              no
ndfromtxt                 no              no
ndim                      no              no
ndindex                   no              no
nditer                    no              no
negative                  no              no
nested_iters              no              no
newaxis                   no              no
newbuffer                 no              no
nextafter                 no              no
nonzero                   yes
not_equal                 no              no
nper                      no              no
npv                       no              no
number                    no              no
obj2sctype                no              no
object                    no              no
object0                   no              no
object_                   no              no
ogrid                     no              no
ones                      yes
ones_like                 yes
outer                     yes
packbits                  no              no
pad                       no              no
partition                 no              no
percentile                no              no
pi                        no              no
piecewise                 no              no
pkgload                   no              no
place                     no              no
pmt                       no              no
poly                      no              no
poly1d                    no              no
polyadd                   no              no
polyder                   no              no
polydiv                   no              no
polyfit                   no              no
polyint                   no              no
polymul                   no              no
polynomial                no              no
polysub                   no              no
polyval                   no              no
power                     yes
ppmt                      no              no
print_function            no              no
prod                      yes
product                   no              no
promote_types             no              no
ptp                       yes
put                       no              no
putmask                   no              no
pv                        no              no
r_                        no              no
rad2deg                   yes
radians                   no              no
random                    yes
rank                      no              no
rate                      no              no
ravel                     no              no
ravel_multi_index         no              no
real                      yes
real_if_close             no              no
rec                       no              no
recarray                  no              no
recfromcsv                no              no
recfromtxt                no              no
reciprocal                no              no
record                    no              no
remainder                 no              no
repeat                    yes
require                   no              no
reshape                   yes
resize                    no              no
restoredot                no              no
result_type               no              no
right_shift               no              no
rint                      no              no
roll                      yes
rollaxis                  no              no
roots                     no              no
rot90                     no              no
round                     yes
round_                    no              no
row_stack                 no              no
s_                        no              no
safe_eval                 no              no
save                      no              no
savetxt                   no              no
savez                     no              no
savez_compressed          no              no
sctype2char               no              no
sctypeDict                no              no
sctypeNA                  no              no
sctypes                   no              no
searchsorted              no              no
select                    no              no
set_numeric_ops           no              no
set_printoptions          no              no
set_string_function       no              no
setbufsize                no              no
setdiff1d                 no              no
seterr                    no              no
seterrcall                no              no
seterrobj                 no              no
setxor1d                  no              no
shape                     yes
short                     no              no
show_config               no              no
sign                      no              no
signbit                   no              no
signedinteger             no              no
sin                       yes
sinc                      no              no
single                    no              no
singlecomplex             no              no
sinh                      yes
size                      no              no
sometrue                  no              no
sort                      yes
sort_complex              no              no
source                    no              no
spacing                   no              no
split                     yes
sqrt                      yes
square                    yes
squeeze                   yes
std                       yes
str                       no              no
str_                      no              no
string0                   no              no
string_                   no              no
subtract                  no              no
sum                       yes
swapaxes                  yes
sys                       yes
take                      yes
tan                       yes
tanh                      yes
tensordot                 yes
test                      no              no
testing                   no              no
tile                      yes
timedelta64               no              no
trace                     no              no
transpose                 yes
trapz                     no              no
tri                       yes
tril                      yes
tril_indices              no              no
tril_indices_from         no              no
trim_zeros                no              no
triu                      yes
triu_indices              no              no
triu_indices_from         no              no
true_divide               no              no
trunc                     yes
typeDict                  no              no
typeNA                    no              no
typecodes                 no              no
typename                  no              no
ubyte                     no              no
ufunc                     no              no
uint                      no              no
uint0                     no              no
uint16                    no              no
uint32                    no              no
uint64                    no              no
uint8                     no              no
uintc                     no              no
uintp                     no              no
ulonglong                 no              no
unicode                   no              no
unicode0                  no              no
unicode_                  no              no
union1d                   no              no
unique                    no              no
unpackbits                no              no
unravel_index             no              no
unsignedinteger           no              no
unwrap                    no              no
use_fastnumpy             no              no
ushort                    no              no
vander                    no              no
var                       yes
vdot                      no              no
vectorize                 no              no
version                   no              no
void                      no              no
void0                     no              no
vsplit                    no              no
vstack                    no              no
warnings                  yes
where                     yes
who                       no              no
zeros                     yes
zeros_like                yes
Contributor

yingzha commented Dec 2, 2014

numpy.*                   theano.tensor.* interface compatible
ALLOW_THREADS             no              no
BUFSIZE                   no              no
CLIP                      no              no
ComplexWarning            no              no
DataSource                no              no
ERR_CALL                  no              no
ERR_DEFAULT               no              no
ERR_DEFAULT2              no              no
ERR_IGNORE                no              no
ERR_LOG                   no              no
ERR_PRINT                 no              no
ERR_RAISE                 no              no
ERR_WARN                  no              no
FLOATING_POINT_SUPPORT    no              no
FPE_DIVIDEBYZERO          no              no
FPE_INVALID               no              no
FPE_OVERFLOW              no              no
FPE_UNDERFLOW             no              no
False_                    no              no
Inf                       no              no
Infinity                  no              no
MAXDIMS                   no              no
MachAr                    no              no
ModuleDeprecationWarning  no              no
NAN                       no              no
NINF                      no              no
NZERO                     no              no
NaN                       no              no
PINF                      no              no
PZERO                     no              no
PackageLoader             no              no
RAISE                     no              no
RankWarning               no              no
SHIFT_DIVIDEBYZERO        no              no
SHIFT_INVALID             no              no
SHIFT_OVERFLOW            no              no
SHIFT_UNDERFLOW           no              no
ScalarType                no              no
Tester                    no              no
True_                     no              no
UFUNC_BUFSIZE_DEFAULT     no              no
UFUNC_PYVALS_NAME         no              no
WRAP                      no              no
__NUMPY_SETUP__           no              no
__all__                   no              no
__builtins__              yes
__config__                no              no
__doc__                   yes
__file__                  yes
__git_revision__          no              no
__name__                  yes
__package__               yes
__path__                  yes
__version__               no              no
_import_tools             no              no
_mat                      no              no
_old_rtld                 no              no
abs                       no              no
absolute                  no              no
absolute_import           no              no
add                       yes
add_docstring             no              no
add_newdoc                no              no
add_newdoc_ufunc          no              no
add_newdocs               no              no
alen                      no              no
all                       yes
allclose                  no              no
alltrue                   no              no
alterdot                  no              no
amax                      no              no
amin                      no              no
angle                     yes
any                       yes
append                    no              no
apply_along_axis          no              no
apply_over_axes           no              no
arange                    yes
arccos                    yes
arccosh                   yes
arcsin                    yes
arcsinh                   yes
arctan                    yes
arctan2                   yes
arctanh                   yes
argmax                    yes
argmin                    yes
argpartition              no              no
argsort                   yes
argwhere                  no              no
around                    no              no
array                     no              no
array2string              no              no
array_equal               no              no
array_equiv               no              no
array_repr                no              no
array_split               no              no
array_str                 no              no
asanyarray                no              no
asarray                   no              no
asarray_chkfinite         no              no
ascontiguousarray         no              no
asfarray                  no              no
asfortranarray            no              no
asmatrix                  no              no
asscalar                  no              no
atleast_1d                no              no
atleast_2d                no              no
atleast_3d                no              no
average                   no              no
bartlett                  yes
base_repr                 no              no
bench                     no              no
binary_repr               no              no
bincount                  yes
bitwise_and               yes
bitwise_not               yes
bitwise_or                yes
bitwise_xor               yes
blackman                  no              no
bmat                      no              no
bool                      no              no
bool8                     no              no
bool_                     no              no
broadcast                 no              no
broadcast_arrays          no              no
busday_count              no              no
busday_offset             no              no
busdaycalendar            no              no
byte                      no              no
byte_bounds               no              no
bytes_                    no              no
c_                        no              no
can_cast                  no              no
cast                      yes
cdouble                   no              no
ceil                      yes
cfloat                    no              no
char                      no              no
character                 no              no
chararray                 no              no
choose                    yes
clip                      yes
clongdouble               no              no
clongfloat                no              no
column_stack              no              no
common_type               no              no
compare_chararrays        no              no
compat                    no              no
complex                   yes
complex128                no              no
complex256                no              no
complex64                 no              no
complex_                  no              no
complexfloating           no              no
compress                  no              no
concatenate               yes
conj                      yes
conjugate                 no              no
convolve                  no              no
copy                      yes
copysign                  no              no
copyto                    no              no
core                      no              no
corrcoef                  no              no
correlate                 no              no
cos                       yes
cosh                      yes
count_nonzero             no              no
cov                       no              no
cross                     no              no
csingle                   no              no
ctypeslib                 no              no
cumprod                   yes
cumproduct                no              no
cumsum                    yes
datetime64                no              no
datetime_as_string        no              no
datetime_data             no              no
deg2rad                   yes
degrees                   no              no
delete                    no              no
deprecate                 no              no
deprecate_with_doc        no              no
diag                      yes
diag_indices              no              no
diag_indices_from         no              no
diagflat                  no              no
diagonal                  yes
diff                      no              no
digitize                  no              no
disable_rtld_global       no              no
disp                      no              no
distutils                 no              no
divide                    no              no
division                  no              no
dot                       yes
double                    no              no
dsplit                    no              no
dstack                    no              no
dtype                     no              no
dual                      no              no
e                         no              no
ediff1d                   no              no
einsum                    no              no
emath                     no              no
empty                     no              no
empty_like                no              no
enable_rtld_global        no              no
equal                     no              no
errstate                  no              no
euler_gamma               no              no
exp                       yes
exp2                      yes
expand_dims               no              no
expm1                     yes
extract                   no              no
eye                       yes
fabs                      no              no
fastCopyAndTranspose      no              no
fft                       no              no
fill_diagonal             yes
find_common_type          no              no
finfo                     no              no
fix                       no              no
flatiter                  no              no
flatnonzero               yes
flexible                  no              no
fliplr                    no              no
flipud                    no              no
float                     no              no
float128                  no              no
float16                   no              no
float32                   no              no
float64                   no              no
float_                    no              no
floating                  no              no
floor                     yes
floor_divide              no              no
fmax                      no              no
fmin                      no              no
fmod                      no              no
format_parser             no              no
frexp                     no              no
frombuffer                no              no
fromfile                  no              no
fromfunction              no              no
fromiter                  no              no
frompyfunc                no              no
fromregex                 no              no
fromstring                no              no
full                      no              no
full_like                 no              no
fv                        no              no
generic                   no              no
genfromtxt                no              no
get_array_wrap            no              no
get_include               no              no
get_numarray_include      no              no
get_printoptions          no              no
getbuffer                 no              no
getbufsize                no              no
geterr                    no              no
geterrcall                no              no
geterrobj                 no              no
gradient                  no              no
greater                   no              no
greater_equal             no              no
half                      no              no
hamming                   no              no
hanning                   no              no
histogram                 no              no
histogram2d               no              no
histogramdd               no              no
hsplit                    no              no
hstack                    no              no
hypot                     no              no
i0                        no              no
identity                  no              no
iinfo                     no              no
imag                      yes
in1d                      no              no
index_exp                 no              no
indices                   no              no
inexact                   no              no
inf                       no              no
info                      no              no
infty                     no              no
inner                     no              no
insert                    no              no
int                       no              no
int0                      no              no
int16                     no              no
int32                     no              no
int64                     no              no
int8                      no              no
int_                      no              no
int_asbuffer              no              no
intc                      no              no
integer                   no              no
interp                    no              no
intersect1d               no              no
intp                      no              no
invert                    yes
ipmt                      no              no
irr                       no              no
is_busday                 no              no
isclose                   no              no
iscomplex                 no              no
iscomplexobj              no              no
isfinite                  no              no
isfortran                 no              no
isinf                     yes
isnan                     yes
isneginf                  no              no
isposinf                  no              no
isreal                    no              no
isrealobj                 no              no
isscalar                  no              no
issctype                  no              no
issubclass_               no              no
issubdtype                no              no
issubsctype               no              no
iterable                  no              no
ix_                       no              no
kaiser                    no              no
kron                      no              no
ldexp                     no              no
left_shift                no              no
less                      no              no
less_equal                no              no
lexsort                   no              no
lib                       no              no
linalg                    no              no
linspace                  no              no
little_endian             no              no
load                      yes
loads                     no              no
loadtxt                   no              no
log                       yes
log10                     yes
log1p                     yes
log2                      yes
logaddexp                 no              no
logaddexp2                no              no
logical_and               no              no
logical_not               no              no
logical_or                no              no
logical_xor               no              no
logspace                  no              no
long                      no              no
longcomplex               no              no
longdouble                no              no
longfloat                 no              no
longlong                  no              no
lookfor                   no              no
ma                        no              no
mafromtxt                 no              no
mask_indices              no              no
mat                       no              no
math                      no              no
matrix                    yes
matrixlib                 no              no
max                       yes
maximum                   yes
maximum_sctype            no              no
may_share_memory          no              no
mean                      yes
median                    no              no
memmap                    no              no
meshgrid                  no              no
mgrid                     no              no
min                       yes
min_scalar_type           no              no
minimum                   yes
mintypecode               no              no
mirr                      no              no
mod                       yes
modf                      no              no
msort                     no              no
multiply                  no              no
nan                       no              no
nan_to_num                no              no
nanargmax                 no              no
nanargmin                 no              no
nanmax                    no              no
nanmean                   no              no
nanmin                    no              no
nanstd                    no              no
nansum                    no              no
nanvar                    no              no
nbytes                    no              no
ndarray                   no              no
ndenumerate               no              no
ndfromtxt                 no              no
ndim                      no              no
ndindex                   no              no
nditer                    no              no
negative                  no              no
nested_iters              no              no
newaxis                   no              no
newbuffer                 no              no
nextafter                 no              no
nonzero                   yes
not_equal                 no              no
nper                      no              no
npv                       no              no
number                    no              no
obj2sctype                no              no
object                    no              no
object0                   no              no
object_                   no              no
ogrid                     no              no
ones                      yes
ones_like                 yes
outer                     yes
packbits                  no              no
pad                       no              no
partition                 no              no
percentile                no              no
pi                        no              no
piecewise                 no              no
pkgload                   no              no
place                     no              no
pmt                       no              no
poly                      no              no
poly1d                    no              no
polyadd                   no              no
polyder                   no              no
polydiv                   no              no
polyfit                   no              no
polyint                   no              no
polymul                   no              no
polynomial                no              no
polysub                   no              no
polyval                   no              no
power                     yes
ppmt                      no              no
print_function            no              no
prod                      yes
product                   no              no
promote_types             no              no
ptp                       yes
put                       no              no
putmask                   no              no
pv                        no              no
r_                        no              no
rad2deg                   yes
radians                   no              no
random                    yes
rank                      no              no
rate                      no              no
ravel                     no              no
ravel_multi_index         no              no
real                      yes
real_if_close             no              no
rec                       no              no
recarray                  no              no
recfromcsv                no              no
recfromtxt                no              no
reciprocal                no              no
record                    no              no
remainder                 no              no
repeat                    yes
require                   no              no
reshape                   yes
resize                    no              no
restoredot                no              no
result_type               no              no
right_shift               no              no
rint                      no              no
roll                      yes
rollaxis                  no              no
roots                     no              no
rot90                     no              no
round                     yes
round_                    no              no
row_stack                 no              no
s_                        no              no
safe_eval                 no              no
save                      no              no
savetxt                   no              no
savez                     no              no
savez_compressed          no              no
sctype2char               no              no
sctypeDict                no              no
sctypeNA                  no              no
sctypes                   no              no
searchsorted              no              no
select                    no              no
set_numeric_ops           no              no
set_printoptions          no              no
set_string_function       no              no
setbufsize                no              no
setdiff1d                 no              no
seterr                    no              no
seterrcall                no              no
seterrobj                 no              no
setxor1d                  no              no
shape                     yes
short                     no              no
show_config               no              no
sign                      no              no
signbit                   no              no
signedinteger             no              no
sin                       yes
sinc                      no              no
single                    no              no
singlecomplex             no              no
sinh                      yes
size                      no              no
sometrue                  no              no
sort                      yes
sort_complex              no              no
source                    no              no
spacing                   no              no
split                     yes
sqrt                      yes
square                    yes
squeeze                   yes
std                       yes
str                       no              no
str_                      no              no
string0                   no              no
string_                   no              no
subtract                  no              no
sum                       yes
swapaxes                  yes
sys                       yes
take                      yes
tan                       yes
tanh                      yes
tensordot                 yes
test                      no              no
testing                   no              no
tile                      yes
timedelta64               no              no
trace                     no              no
transpose                 yes
trapz                     no              no
tri                       yes
tril                      yes
tril_indices              no              no
tril_indices_from         no              no
trim_zeros                no              no
triu                      yes
triu_indices              no              no
triu_indices_from         no              no
true_divide               no              no
trunc                     yes
typeDict                  no              no
typeNA                    no              no
typecodes                 no              no
typename                  no              no
ubyte                     no              no
ufunc                     no              no
uint                      no              no
uint0                     no              no
uint16                    no              no
uint32                    no              no
uint64                    no              no
uint8                     no              no
uintc                     no              no
uintp                     no              no
ulonglong                 no              no
unicode                   no              no
unicode0                  no              no
unicode_                  no              no
union1d                   no              no
unique                    no              no
unpackbits                no              no
unravel_index             no              no
unsignedinteger           no              no
unwrap                    no              no
use_fastnumpy             no              no
ushort                    no              no
vander                    no              no
var                       yes
vdot                      no              no
vectorize                 no              no
version                   no              no
void                      no              no
void0                     no              no
vsplit                    no              no
vstack                    no              no
warnings                  yes
where                     yes
who                       no              no
zeros                     yes
zeros_like                yes
@yingzha

This comment has been minimized.

Show comment
Hide comment
@yingzha

yingzha Dec 2, 2014

Contributor

haven't finished yet, I'll fill the blanks as soon as possible.

Contributor

yingzha commented Dec 2, 2014

haven't finished yet, I'll fill the blanks as soon as possible.

@abergeron

This comment has been minimized.

Show comment
Hide comment
@abergeron

abergeron Dec 2, 2014

Member

While it is a goal to provide a numpy-like interface, I don't think we aim for name-to-name compliance with the imported modules. So I think your list should exclude external modules like __buitins__ or sys.

Member

abergeron commented Dec 2, 2014

While it is a goal to provide a numpy-like interface, I don't think we aim for name-to-name compliance with the imported modules. So I think your list should exclude external modules like __buitins__ or sys.

@vdumoulin

This comment has been minimized.

Show comment
Hide comment
@vdumoulin

vdumoulin Dec 2, 2014

Contributor

Yeah, I knew this would generate lots of false positives, but it's a good
starting point. I'll edit that list to exclude the most obvious ones.

Vincent Dumoulin, B.Sc. Physique et Informatique
Étudiant au Ph.D.
LISA, Université de Montréal
Bureau: 3336, pavillon André-Aisenstadt
vincent.dumoulin@umontreal.ca

On Tue, Dec 2, 2014 at 12:49 PM, abergeron notifications@github.com wrote:

While it is a goal to provide a numpy-like interface, I don't think we aim
for name-to-name compliance with the imported modules. So I think your list
should exclude external modules like buitins or sys.


Reply to this email directly or view it on GitHub
#1916 (comment).

Contributor

vdumoulin commented Dec 2, 2014

Yeah, I knew this would generate lots of false positives, but it's a good
starting point. I'll edit that list to exclude the most obvious ones.

Vincent Dumoulin, B.Sc. Physique et Informatique
Étudiant au Ph.D.
LISA, Université de Montréal
Bureau: 3336, pavillon André-Aisenstadt
vincent.dumoulin@umontreal.ca

On Tue, Dec 2, 2014 at 12:49 PM, abergeron notifications@github.com wrote:

While it is a goal to provide a numpy-like interface, I don't think we aim
for name-to-name compliance with the imported modules. So I think your list
should exclude external modules like buitins or sys.


Reply to this email directly or view it on GitHub
#1916 (comment).

@yingzha

This comment has been minimized.

Show comment
Hide comment
@yingzha

yingzha Dec 16, 2014

Contributor
numpy.*                   theano.tensor.* interface compatible
ALLOW_THREADS             no              no
BUFSIZE                   no              no
CLIP                      no              no
ComplexWarning            no              no
DataSource                no              no
ERR_CALL                  no              no
ERR_DEFAULT               no              no
ERR_DEFAULT2              no              no
ERR_IGNORE                no              no
ERR_LOG                   no              no
ERR_PRINT                 no              no
ERR_RAISE                 no              no
ERR_WARN                  no              no
FLOATING_POINT_SUPPORT    no              no
FPE_DIVIDEBYZERO          no              no
FPE_INVALID               no              no
FPE_OVERFLOW              no              no
FPE_UNDERFLOW             no              no
False_                    no              no
Inf                       no              no
Infinity                  no              no
MAXDIMS                   no              no
MachAr                    no              no
ModuleDeprecationWarning  no              no
NAN                       no              no
NINF                      no              no
NZERO                     no              no
NaN                       no              no
PINF                      no              no
PZERO                     no              no
PackageLoader             no              no
RAISE                     no              no
RankWarning               no              no
SHIFT_DIVIDEBYZERO        no              no
SHIFT_INVALID             no              no
SHIFT_OVERFLOW            no              no
SHIFT_UNDERFLOW           no              no
ScalarType                no              no
Tester                    no              no
True_                     no              no
UFUNC_BUFSIZE_DEFAULT     no              no
UFUNC_PYVALS_NAME         no              no
WRAP                      no              no
__NUMPY_SETUP__           no              no
__all__                   no              no
__builtins__              yes             yes
__config__                no              no
__doc__                   yes             yes
__file__                  yes             yes
__git_revision__          no              no
__name__                  yes             yes
__package__               yes             yes
__path__                  yes             yes
__version__               no              no
_import_tools             no              no
_mat                      no              no
_old_rtld                 no              no
abs                       no              no
absolute                  no              no
absolute_import           no              no
add                       yes             lacks the parameter ‘out’
add_docstring             no              no
add_newdoc                no              no
add_newdoc_ufunc          no              no
add_newdocs               no              no
alen                      no              no
all                       yes             lacks the parameter ‘out’
allclose                  no              no
alltrue                   no              no
alterdot                  no              no
amax                      no              no
amin                      no              no
angle                     yes             lacks the parameter ‘deg’
any                       yes             lacks the parameter ‘out’
append                    no              no
apply_along_axis          no              no
apply_over_axes           no              no
arange                    yes             yes
arccos                    yes             lacks the parameter ‘out’
arccosh                   yes             lacks the parameter ‘out’
arcsin                    yes             lacks the parameter ‘out’
arcsinh                   yes             lacks the parameter ‘out’
arctan                    yes             lacks the parameter ‘out’
arctan2                   yes             lacks the parameter ‘out’
arctanh                   yes             lacks the parameter ‘out’
argmax                    yes             added the param ‘keepdims’
argmin                    yes             added the param ‘keepdims’
argpartition              no              no
argsort                   yes             yes
argwhere                  no              no
around                    no              no
array                     no              no
array2string              no              no
array_equal               no              no
array_equiv               no              no
array_repr                no              no
array_split               no              no
array_str                 no              no
asanyarray                no              no
asarray                   no              no
asarray_chkfinite         no              no
ascontiguousarray         no              no
asfarray                  no              no
asfortranarray            no              no
asmatrix                  no              no
asscalar                  no              no
atleast_1d                no              no
atleast_2d                no              no
atleast_3d                no              no
average                   no              no
bartlett                  yes             yes
base_repr                 no              no
bench                     no              no
binary_repr               no              no
bincount                  yes             yes
bitwise_and               yes             lacks the parameter ‘out’
bitwise_not               yes             lacks the parameter ‘out’
bitwise_or                yes             lacks the parameter ‘out’
bitwise_xor               yes             lacks the parameter ‘out’
blackman                  no              no
bmat                      no              no
bool                      no              no
bool8                     no              no
bool_                     no              no
broadcast                 no              no
broadcast_arrays          no              no
busday_count              no              no
busday_offset             no              no
busdaycalendar            no              no
byte                      no              no
byte_bounds               no              no
bytes_                    no              no
c_                        no              no
can_cast                  no              no
cast                      yes             no, numpy.cast is Base object for a dictionary for look-up with any alias for an array dtype.
cdouble                   no              no
ceil                      yes             lacks the parameter ‘out’
cfloat                    no              no
char                      no              no
character                 no              no
chararray                 no              no
choose                    yes             yes
clip                      yes             lacks the parameter ‘out’
clongdouble               no              no
clongfloat                no              no
column_stack              no              no
common_type               no              no
compare_chararrays        no              no
compat                    no              no
complex                   yes             yes
complex128                no              no
complex256                no              no
complex64                 no              no
complex_                  no              no
complexfloating           no              no
compress                  no              no
concatenate               yes             yes
conj                      yes             lacks the parameter ‘out’
conjugate                 no              no
convolve                  no              no
copy                      yes             lacks the param ‘order’
copysign                  no              no
copyto                    no              no
core                      no              no
corrcoef                  no              no
correlate                 no              no
cos                       yes             lacks the parameter ‘out’
cosh                      yes             lacks the parameter ‘out’
count_nonzero             no              no
cov                       no              no
cross                     no              no
csingle                   no              no
ctypeslib                 no              no
cumprod                   yes             lacks the params ‘dtype’,’out’
cumproduct                no              no
cumsum                    yes             lacks the params ‘dtype’, ‘out’
datetime64                no              no
datetime_as_string        no              no
datetime_data             no              no
deg2rad                   yes             lacks the parameter ‘out’
degrees                   no              no
delete                    no              no
deprecate                 no              no
deprecate_with_doc        no              no
diag                      yes             yes
diag_indices              no              no
diag_indices_from         no              no
diagflat                  no              no
diagonal                  yes             yes
diff                      no              no
digitize                  no              no
disable_rtld_global       no              no
disp                      no              no
distutils                 no              no
divide                    no              no
division                  no              no
dot                       yes             lacks the param ‘out’
double                    no              no
dsplit                    no              no
dstack                    no              no
dtype                     no              no
dual                      no              no
e                         no              no
ediff1d                   no              no
einsum                    no              no
emath                     no              no
empty                     no              no
empty_like                no              no
enable_rtld_global        no              no
equal                     no              no
errstate                  no              no
euler_gamma               no              no
exp                       yes             lacks the param ’out’
exp2                      yes             lacks the param ‘out’
expand_dims               no              no
expm1                     yes             lacks the param ‘out’
extract                   no              no
eye                       yes             yes
fabs                      no              no
fastCopyAndTranspose      no              no
fft                       no              no
fill_diagonal             yes             lacks the param ‘wrap’
find_common_type          no              no
finfo                     no              no
fix                       no              no
flatiter                  no              no
flatnonzero               yes             yes
flexible                  no              no
fliplr                    no              no
flipud                    no              no
float                     no              no
float128                  no              no
float16                   no              no
float32                   no              no
float64                   no              no
float_                    no              no
floating                  no              no
floor                     yes             lacks the param ‘out’
floor_divide              no              no
fmax                      no              no
fmin                      no              no
fmod                      no              no
format_parser             no              no
frexp                     no              no
frombuffer                no              no
fromfile                  no              no
fromfunction              no              no
fromiter                  no              no
frompyfunc                no              no
fromregex                 no              no
fromstring                no              no
full                      no              no
full_like                 no              no
fv                        no              no
generic                   no              no
genfromtxt                no              no
get_array_wrap            no              no
get_include               no              no
get_numarray_include      no              no
get_printoptions          no              no
getbuffer                 no              no
getbufsize                no              no
geterr                    no              no
geterrcall                no              no
geterrobj                 no              no
gradient                  no              no
greater                   no              no
greater_equal             no              no
half                      no              no
hamming                   no              no
hanning                   no              no
histogram                 no              no
histogram2d               no              no
histogramdd               no              no
hsplit                    no              no
hstack                    no              no
hypot                     no              no
i0                        no              no
identity                  no              no
iinfo                     no              no
imag                      yes             yes
in1d                      no              no
index_exp                 no              no
indices                   no              no
inexact                   no              no
inf                       no              no
info                      no              no
infty                     no              no
inner                     no              no
insert                    no              no
int                       no              no
int0                      no              no
int16                     no              no
int32                     no              no
int64                     no              no
int8                      no              no
int_                      no              no
int_asbuffer              no              no
intc                      no              no
integer                   no              no
interp                    no              no
intersect1d               no              no
intp                      no              no
invert                    yes             lacks the param ‘out’
ipmt                      no              no
irr                       no              no
is_busday                 no              no
isclose                   no              no
iscomplex                 no              no
iscomplexobj              no              no
isfinite                  no              no
isfortran                 no              no
isinf                     yes             lacks the param ‘out’ 
isnan                     yes             lacks the param ‘out’
isneginf                  no              no
isposinf                  no              no
isreal                    no              no
isrealobj                 no              no
isscalar                  no              no
issctype                  no              no
issubclass_               no              no
issubdtype                no              no
issubsctype               no              no
iterable                  no              no
ix_                       no              no
kaiser                    no              no
kron                      no              no
ldexp                     no              no
left_shift                no              no
less                      no              no
less_equal                no              no
lexsort                   no              no
lib                       no              no
linalg                    no              no
linspace                  no              no
little_endian             no              no
load                      yes             added params ‘dtype’,’broadcastable’, load an array from an .npy file, while numpy.load loads arrays or pickled objects from .npy, .npz or pickled files. 
loads                     no              no
loadtxt                   no              no
log                       yes             lacks the param ‘out’
log10                     yes             lacks the param ‘out’
log1p                     yes             lacks the param ‘out’
log2                      yes             lacks the param ‘out’
logaddexp                 no              no
logaddexp2                no              no
logical_and               no              no
logical_not               no              no
logical_or                no              no
logical_xor               no              no
logspace                  no              no
long                      no              no
longcomplex               no              no
longdouble                no              no
longfloat                 no              no
longlong                  no              no
lookfor                   no              no
ma                        no              no
mafromtxt                 no              no
mask_indices              no              no
mat                       no              no
math                      no              no
matrix                    yes             lacks the param ‘copy’
matrixlib                 no              no
max                       yes             lacks the param ‘out’
maximum                   yes             lacks the param ‘out’
maximum_sctype            no              no
may_share_memory          no              no
mean                      yes             added the params ‘op’, ‘acc_dtype’, lacking the param ‘out’
median                    no              no
memmap                    no              no
meshgrid                  no              no
mgrid                     no              no
min                       yes             lacks the param ‘out’
min_scalar_type           no              no
minimum                   yes             lacks the param ‘out’
mintypecode               no              no
mirr                      no              no
mod                       yes             lacks the param ‘out’
modf                      no              no
msort                     no              no
multiply                  no              no
nan                       no              no
nan_to_num                no              no
nanargmax                 no              no
nanargmin                 no              no
nanmax                    no              no
nanmean                   no              no
nanmin                    no              no
nanstd                    no              no
nansum                    no              no
nanvar                    no              no
nbytes                    no              no
ndarray                   no              no
ndenumerate               no              no
ndfromtxt                 no              no
ndim                      no              no
ndindex                   no              no
nditer                    no              no
negative                  no              no
nested_iters              no              no
newaxis                   no              no
newbuffer                 no              no
nextafter                 no              no
nonzero                   yes             if param ‘retrun_matrix=False’, the same as Numpy, else the same as Theano Op
not_equal                 no              no
nper                      no              no
npv                       no              no
number                    no              no
obj2sctype                no              no
object                    no              no
object0                   no              no
object_                   no              no
ogrid                     no              no
ones                      yes             lacks the param ‘order’
ones_like                 yes             lacks the param ‘order’,’subok’
outer                     yes             lacks the param ‘out’
packbits                  no              no
pad                       no              no
partition                 no              no
percentile                no              no
pi                        no              no
piecewise                 no              no
pkgload                   no              no
place                     no              no
pmt                       no              no
poly                      no              no
poly1d                    no              no
polyadd                   no              no
polyder                   no              no
polydiv                   no              no
polyfit                   no              no
polyint                   no              no
polymul                   no              no
polynomial                no              no
polysub                   no              no
polyval                   no              no
power                     yes             lacks the param ‘out’
ppmt                      no              no
print_function            no              no
prod                      yes             added the params ‘acc_dtype’, ‘no_zeros_in_input’, lacks the param ‘out’
product                   no              no
promote_types             no              no
ptp                       yes             lacks the param ‘out’
put                       no              no
putmask                   no              no
pv                        no              no
r_                        no              no
rad2deg                   yes             lacks the param ‘out’
radians                   no              no
random                    yes             yes
rank                      no              no
rate                      no              no
ravel                     no              no
ravel_multi_index         no              no
real                      yes             yes
real_if_close             no              no
rec                       no              no
recarray                  no              no
recfromcsv                no              no
recfromtxt                no              no
reciprocal                no              no
record                    no              no
remainder                 no              no
repeat                    yes             yes
require                   no              no
reshape                   yes             added the params ‘ndim’, ‘name’, lacks the param ‘order’
resize                    no              no
restoredot                no              no
result_type               no              no
right_shift               no              no
rint                      no              no
roll                      yes             yes
rollaxis                  no              no
roots                     no              no
rot90                     no              no
round                     yes             added the param ‘mode’, lacks the params ‘decimals’ and ‘out’
round_                    no              no
row_stack                 no              no
s_                        no              no
safe_eval                 no              no
save                      no              no
savetxt                   no              no
savez                     no              no
savez_compressed          no              no
sctype2char               no              no
sctypeDict                no              no
sctypeNA                  no              no
sctypes                   no              no
searchsorted              no              no
select                    no              no
set_numeric_ops           no              no
set_printoptions          no              no
set_string_function       no              no
setbufsize                no              no
setdiff1d                 no              no
seterr                    no              no
seterrcall                no              no
seterrobj                 no              no
setxor1d                  no              no
shape                     yes             yes
short                     no              no
show_config               no              no
sign                      no              no
signbit                   no              no
signedinteger             no              no
sin                       yes             lacking the param ‘out’, uf
sinc                      no              no
single                    no              no
singlecomplex             no              no
sinh                      yes             lacking the param ‘out’, uf
size                      no              no
sometrue                  no              no
sort                      yes             yes
sort_complex              no              no
source                    no              no
spacing                   no              no
split                     yes             added the params ‘splits_size’ and ’n_splits’, lacks the param ‘indices_or_sections’
sqrt                      yes             lacks the param ‘out’
square                    yes             lacks the param ‘out’
squeeze                   yes             lacks the param ‘axis’
std                       yes             lacks the params ‘dtype’, ‘out’, ‘ddof’
str                       no              no
str_                      no              no
string0                   no              no
string_                   no              no
subtract                  no              no
sum                       yes             adding the param ‘acc_dtype’, lacks the param ‘out’ 
swapaxes                  yes             yes
sys                       yes             yes
take                      yes             lacks the param ‘out’
tan                       yes             lacks the param ‘out’
tanh                      yes             lacks the param ‘out’
tensordot                 yes             yes
test                      no              no
testing                   no              no
tile                      yes             added the param ‘ndim’
timedelta64               no              no
trace                     no              no
transpose                 yes             yes
trapz                     no              no
tri                       yes             yes
tril                      yes             yes
tril_indices              no              no
tril_indices_from         no              no
trim_zeros                no              no
triu                      yes             yes
triu_indices              no              no
triu_indices_from         no              no
true_divide               no              no
trunc                     yes             lacks the param ‘out’
typeDict                  no              no
typeNA                    no              no
typecodes                 no              no
typename                  no              no
ubyte                     no              no
ufunc                     no              no
uint                      no              no
uint0                     no              no
uint16                    no              no
uint32                    no              no
uint64                    no              no
uint8                     no              no
uintc                     no              no
uintp                     no              no
ulonglong                 no              no
unicode                   no              no
unicode0                  no              no
unicode_                  no              no
union1d                   no              no
unique                    no              no
unpackbits                no              no
unravel_index             no              no
unsignedinteger           no              no
unwrap                    no              no
use_fastnumpy             no              no
ushort                    no              no
vander                    no              no
var                       yes             lacks the params ‘dtype’, ’out’, ‘ddof’
vdot                      no              no
vectorize                 no              no
version                   no              no
void                      no              no
void0                     no              no
vsplit                    no              no
vstack                    no              no
warnings                  yes             yes
where                     yes             yes
who                       no              no
zeros                     yes             lacks the param ’order’ 
zeros_like                yes             lacks the params ‘order’,’subok’
Contributor

yingzha commented Dec 16, 2014

numpy.*                   theano.tensor.* interface compatible
ALLOW_THREADS             no              no
BUFSIZE                   no              no
CLIP                      no              no
ComplexWarning            no              no
DataSource                no              no
ERR_CALL                  no              no
ERR_DEFAULT               no              no
ERR_DEFAULT2              no              no
ERR_IGNORE                no              no
ERR_LOG                   no              no
ERR_PRINT                 no              no
ERR_RAISE                 no              no
ERR_WARN                  no              no
FLOATING_POINT_SUPPORT    no              no
FPE_DIVIDEBYZERO          no              no
FPE_INVALID               no              no
FPE_OVERFLOW              no              no
FPE_UNDERFLOW             no              no
False_                    no              no
Inf                       no              no
Infinity                  no              no
MAXDIMS                   no              no
MachAr                    no              no
ModuleDeprecationWarning  no              no
NAN                       no              no
NINF                      no              no
NZERO                     no              no
NaN                       no              no
PINF                      no              no
PZERO                     no              no
PackageLoader             no              no
RAISE                     no              no
RankWarning               no              no
SHIFT_DIVIDEBYZERO        no              no
SHIFT_INVALID             no              no
SHIFT_OVERFLOW            no              no
SHIFT_UNDERFLOW           no              no
ScalarType                no              no
Tester                    no              no
True_                     no              no
UFUNC_BUFSIZE_DEFAULT     no              no
UFUNC_PYVALS_NAME         no              no
WRAP                      no              no
__NUMPY_SETUP__           no              no
__all__                   no              no
__builtins__              yes             yes
__config__                no              no
__doc__                   yes             yes
__file__                  yes             yes
__git_revision__          no              no
__name__                  yes             yes
__package__               yes             yes
__path__                  yes             yes
__version__               no              no
_import_tools             no              no
_mat                      no              no
_old_rtld                 no              no
abs                       no              no
absolute                  no              no
absolute_import           no              no
add                       yes             lacks the parameter ‘out’
add_docstring             no              no
add_newdoc                no              no
add_newdoc_ufunc          no              no
add_newdocs               no              no
alen                      no              no
all                       yes             lacks the parameter ‘out’
allclose                  no              no
alltrue                   no              no
alterdot                  no              no
amax                      no              no
amin                      no              no
angle                     yes             lacks the parameter ‘deg’
any                       yes             lacks the parameter ‘out’
append                    no              no
apply_along_axis          no              no
apply_over_axes           no              no
arange                    yes             yes
arccos                    yes             lacks the parameter ‘out’
arccosh                   yes             lacks the parameter ‘out’
arcsin                    yes             lacks the parameter ‘out’
arcsinh                   yes             lacks the parameter ‘out’
arctan                    yes             lacks the parameter ‘out’
arctan2                   yes             lacks the parameter ‘out’
arctanh                   yes             lacks the parameter ‘out’
argmax                    yes             added the param ‘keepdims’
argmin                    yes             added the param ‘keepdims’
argpartition              no              no
argsort                   yes             yes
argwhere                  no              no
around                    no              no
array                     no              no
array2string              no              no
array_equal               no              no
array_equiv               no              no
array_repr                no              no
array_split               no              no
array_str                 no              no
asanyarray                no              no
asarray                   no              no
asarray_chkfinite         no              no
ascontiguousarray         no              no
asfarray                  no              no
asfortranarray            no              no
asmatrix                  no              no
asscalar                  no              no
atleast_1d                no              no
atleast_2d                no              no
atleast_3d                no              no
average                   no              no
bartlett                  yes             yes
base_repr                 no              no
bench                     no              no
binary_repr               no              no
bincount                  yes             yes
bitwise_and               yes             lacks the parameter ‘out’
bitwise_not               yes             lacks the parameter ‘out’
bitwise_or                yes             lacks the parameter ‘out’
bitwise_xor               yes             lacks the parameter ‘out’
blackman                  no              no
bmat                      no              no
bool                      no              no
bool8                     no              no
bool_                     no              no
broadcast                 no              no
broadcast_arrays          no              no
busday_count              no              no
busday_offset             no              no
busdaycalendar            no              no
byte                      no              no
byte_bounds               no              no
bytes_                    no              no
c_                        no              no
can_cast                  no              no
cast                      yes             no, numpy.cast is Base object for a dictionary for look-up with any alias for an array dtype.
cdouble                   no              no
ceil                      yes             lacks the parameter ‘out’
cfloat                    no              no
char                      no              no
character                 no              no
chararray                 no              no
choose                    yes             yes
clip                      yes             lacks the parameter ‘out’
clongdouble               no              no
clongfloat                no              no
column_stack              no              no
common_type               no              no
compare_chararrays        no              no
compat                    no              no
complex                   yes             yes
complex128                no              no
complex256                no              no
complex64                 no              no
complex_                  no              no
complexfloating           no              no
compress                  no              no
concatenate               yes             yes
conj                      yes             lacks the parameter ‘out’
conjugate                 no              no
convolve                  no              no
copy                      yes             lacks the param ‘order’
copysign                  no              no
copyto                    no              no
core                      no              no
corrcoef                  no              no
correlate                 no              no
cos                       yes             lacks the parameter ‘out’
cosh                      yes             lacks the parameter ‘out’
count_nonzero             no              no
cov                       no              no
cross                     no              no
csingle                   no              no
ctypeslib                 no              no
cumprod                   yes             lacks the params ‘dtype’,’out’
cumproduct                no              no
cumsum                    yes             lacks the params ‘dtype’, ‘out’
datetime64                no              no
datetime_as_string        no              no
datetime_data             no              no
deg2rad                   yes             lacks the parameter ‘out’
degrees                   no              no
delete                    no              no
deprecate                 no              no
deprecate_with_doc        no              no
diag                      yes             yes
diag_indices              no              no
diag_indices_from         no              no
diagflat                  no              no
diagonal                  yes             yes
diff                      no              no
digitize                  no              no
disable_rtld_global       no              no
disp                      no              no
distutils                 no              no
divide                    no              no
division                  no              no
dot                       yes             lacks the param ‘out’
double                    no              no
dsplit                    no              no
dstack                    no              no
dtype                     no              no
dual                      no              no
e                         no              no
ediff1d                   no              no
einsum                    no              no
emath                     no              no
empty                     no              no
empty_like                no              no
enable_rtld_global        no              no
equal                     no              no
errstate                  no              no
euler_gamma               no              no
exp                       yes             lacks the param ’out’
exp2                      yes             lacks the param ‘out’
expand_dims               no              no
expm1                     yes             lacks the param ‘out’
extract                   no              no
eye                       yes             yes
fabs                      no              no
fastCopyAndTranspose      no              no
fft                       no              no
fill_diagonal             yes             lacks the param ‘wrap’
find_common_type          no              no
finfo                     no              no
fix                       no              no
flatiter                  no              no
flatnonzero               yes             yes
flexible                  no              no
fliplr                    no              no
flipud                    no              no
float                     no              no
float128                  no              no
float16                   no              no
float32                   no              no
float64                   no              no
float_                    no              no
floating                  no              no
floor                     yes             lacks the param ‘out’
floor_divide              no              no
fmax                      no              no
fmin                      no              no
fmod                      no              no
format_parser             no              no
frexp                     no              no
frombuffer                no              no
fromfile                  no              no
fromfunction              no              no
fromiter                  no              no
frompyfunc                no              no
fromregex                 no              no
fromstring                no              no
full                      no              no
full_like                 no              no
fv                        no              no
generic                   no              no
genfromtxt                no              no
get_array_wrap            no              no
get_include               no              no
get_numarray_include      no              no
get_printoptions          no              no
getbuffer                 no              no
getbufsize                no              no
geterr                    no              no
geterrcall                no              no
geterrobj                 no              no
gradient                  no              no
greater                   no              no
greater_equal             no              no
half                      no              no
hamming                   no              no
hanning                   no              no
histogram                 no              no
histogram2d               no              no
histogramdd               no              no
hsplit                    no              no
hstack                    no              no
hypot                     no              no
i0                        no              no
identity                  no              no
iinfo                     no              no
imag                      yes             yes
in1d                      no              no
index_exp                 no              no
indices                   no              no
inexact                   no              no
inf                       no              no
info                      no              no
infty                     no              no
inner                     no              no
insert                    no              no
int                       no              no
int0                      no              no
int16                     no              no
int32                     no              no
int64                     no              no
int8                      no              no
int_                      no              no
int_asbuffer              no              no
intc                      no              no
integer                   no              no
interp                    no              no
intersect1d               no              no
intp                      no              no
invert                    yes             lacks the param ‘out’
ipmt                      no              no
irr                       no              no
is_busday                 no              no
isclose                   no              no
iscomplex                 no              no
iscomplexobj              no              no
isfinite                  no              no
isfortran                 no              no
isinf                     yes             lacks the param ‘out’ 
isnan                     yes             lacks the param ‘out’
isneginf                  no              no
isposinf                  no              no
isreal                    no              no
isrealobj                 no              no
isscalar                  no              no
issctype                  no              no
issubclass_               no              no
issubdtype                no              no
issubsctype               no              no
iterable                  no              no
ix_                       no              no
kaiser                    no              no
kron                      no              no
ldexp                     no              no
left_shift                no              no
less                      no              no
less_equal                no              no
lexsort                   no              no
lib                       no              no
linalg                    no              no
linspace                  no              no
little_endian             no              no
load                      yes             added params ‘dtype’,’broadcastable’, load an array from an .npy file, while numpy.load loads arrays or pickled objects from .npy, .npz or pickled files. 
loads                     no              no
loadtxt                   no              no
log                       yes             lacks the param ‘out’
log10                     yes             lacks the param ‘out’
log1p                     yes             lacks the param ‘out’
log2                      yes             lacks the param ‘out’
logaddexp                 no              no
logaddexp2                no              no
logical_and               no              no
logical_not               no              no
logical_or                no              no
logical_xor               no              no
logspace                  no              no
long                      no              no
longcomplex               no              no
longdouble                no              no
longfloat                 no              no
longlong                  no              no
lookfor                   no              no
ma                        no              no
mafromtxt                 no              no
mask_indices              no              no
mat                       no              no
math                      no              no
matrix                    yes             lacks the param ‘copy’
matrixlib                 no              no
max                       yes             lacks the param ‘out’
maximum                   yes             lacks the param ‘out’
maximum_sctype            no              no
may_share_memory          no              no
mean                      yes             added the params ‘op’, ‘acc_dtype’, lacking the param ‘out’
median                    no              no
memmap                    no              no
meshgrid                  no              no
mgrid                     no              no
min                       yes             lacks the param ‘out’
min_scalar_type           no              no
minimum                   yes             lacks the param ‘out’
mintypecode               no              no
mirr                      no              no
mod                       yes             lacks the param ‘out’
modf                      no              no
msort                     no              no
multiply                  no              no
nan                       no              no
nan_to_num                no              no
nanargmax                 no              no
nanargmin                 no              no
nanmax                    no              no
nanmean                   no              no
nanmin                    no              no
nanstd                    no              no
nansum                    no              no
nanvar                    no              no
nbytes                    no              no
ndarray                   no              no
ndenumerate               no              no
ndfromtxt                 no              no
ndim                      no              no
ndindex                   no              no
nditer                    no              no
negative                  no              no
nested_iters              no              no
newaxis                   no              no
newbuffer                 no              no
nextafter                 no              no
nonzero                   yes             if param ‘retrun_matrix=False’, the same as Numpy, else the same as Theano Op
not_equal                 no              no
nper                      no              no
npv                       no              no
number                    no              no
obj2sctype                no              no
object                    no              no
object0                   no              no
object_                   no              no
ogrid                     no              no
ones                      yes             lacks the param ‘order’
ones_like                 yes             lacks the param ‘order’,’subok’
outer                     yes             lacks the param ‘out’
packbits                  no              no
pad                       no              no
partition                 no              no
percentile                no              no
pi                        no              no
piecewise                 no              no
pkgload                   no              no
place                     no              no
pmt                       no              no
poly                      no              no
poly1d                    no              no
polyadd                   no              no
polyder                   no              no
polydiv                   no              no
polyfit                   no              no
polyint                   no              no
polymul                   no              no
polynomial                no              no
polysub                   no              no
polyval                   no              no
power                     yes             lacks the param ‘out’
ppmt                      no              no
print_function            no              no
prod                      yes             added the params ‘acc_dtype’, ‘no_zeros_in_input’, lacks the param ‘out’
product                   no              no
promote_types             no              no
ptp                       yes             lacks the param ‘out’
put                       no              no
putmask                   no              no
pv                        no              no
r_                        no              no
rad2deg                   yes             lacks the param ‘out’
radians                   no              no
random                    yes             yes
rank                      no              no
rate                      no              no
ravel                     no              no
ravel_multi_index         no              no
real                      yes             yes
real_if_close             no              no
rec                       no              no
recarray                  no              no
recfromcsv                no              no
recfromtxt                no              no
reciprocal                no              no
record                    no              no
remainder                 no              no
repeat                    yes             yes
require                   no              no
reshape                   yes             added the params ‘ndim’, ‘name’, lacks the param ‘order’
resize                    no              no
restoredot                no              no
result_type               no              no
right_shift               no              no
rint                      no              no
roll                      yes             yes
rollaxis                  no              no
roots                     no              no
rot90                     no              no
round                     yes             added the param ‘mode’, lacks the params ‘decimals’ and ‘out’
round_                    no              no
row_stack                 no              no
s_                        no              no
safe_eval                 no              no
save                      no              no
savetxt                   no              no
savez                     no              no
savez_compressed          no              no
sctype2char               no              no
sctypeDict                no              no
sctypeNA                  no              no
sctypes                   no              no
searchsorted              no              no
select                    no              no
set_numeric_ops           no              no
set_printoptions          no              no
set_string_function       no              no
setbufsize                no              no
setdiff1d                 no              no
seterr                    no              no
seterrcall                no              no
seterrobj                 no              no
setxor1d                  no              no
shape                     yes             yes
short                     no              no
show_config               no              no
sign                      no              no
signbit                   no              no
signedinteger             no              no
sin                       yes             lacking the param ‘out’, uf
sinc                      no              no
single                    no              no
singlecomplex             no              no
sinh                      yes             lacking the param ‘out’, uf
size                      no              no
sometrue                  no              no
sort                      yes             yes
sort_complex              no              no
source                    no              no
spacing                   no              no
split                     yes             added the params ‘splits_size’ and ’n_splits’, lacks the param ‘indices_or_sections’
sqrt                      yes             lacks the param ‘out’
square                    yes             lacks the param ‘out’
squeeze                   yes             lacks the param ‘axis’
std                       yes             lacks the params ‘dtype’, ‘out’, ‘ddof’
str                       no              no
str_                      no              no
string0                   no              no
string_                   no              no
subtract                  no              no
sum                       yes             adding the param ‘acc_dtype’, lacks the param ‘out’ 
swapaxes                  yes             yes
sys                       yes             yes
take                      yes             lacks the param ‘out’
tan                       yes             lacks the param ‘out’
tanh                      yes             lacks the param ‘out’
tensordot                 yes             yes
test                      no              no
testing                   no              no
tile                      yes             added the param ‘ndim’
timedelta64               no              no
trace                     no              no
transpose                 yes             yes
trapz                     no              no
tri                       yes             yes
tril                      yes             yes
tril_indices              no              no
tril_indices_from         no              no
trim_zeros                no              no
triu                      yes             yes
triu_indices              no              no
triu_indices_from         no              no
true_divide               no              no
trunc                     yes             lacks the param ‘out’
typeDict                  no              no
typeNA                    no              no
typecodes                 no              no
typename                  no              no
ubyte                     no              no
ufunc                     no              no
uint                      no              no
uint0                     no              no
uint16                    no              no
uint32                    no              no
uint64                    no              no
uint8                     no              no
uintc                     no              no
uintp                     no              no
ulonglong                 no              no
unicode                   no              no
unicode0                  no              no
unicode_                  no              no
union1d                   no              no
unique                    no              no
unpackbits                no              no
unravel_index             no              no
unsignedinteger           no              no
unwrap                    no              no
use_fastnumpy             no              no
ushort                    no              no
vander                    no              no
var                       yes             lacks the params ‘dtype’, ’out’, ‘ddof’
vdot                      no              no
vectorize                 no              no
version                   no              no
void                      no              no
void0                     no              no
vsplit                    no              no
vstack                    no              no
warnings                  yes             yes
where                     yes             yes
who                       no              no
zeros                     yes             lacks the param ’order’ 
zeros_like                yes             lacks the params ‘order’,’subok’
@vdumoulin

This comment has been minimized.

Show comment
Hide comment
@vdumoulin

vdumoulin Jan 6, 2015

Contributor

Here's the same list, with some false positives removed:

numpy.*                   theano.tensor.* interface compatible
abs                       no              no
absolute                  no              no
add                       yes             lacks the parameter ‘out’
alen                      no              no
all                       yes             lacks the parameter ‘out’
allclose                  no              no
alltrue                   no              no
alterdot                  no              no
amax                      no              no
amin                      no              no
angle                     yes             lacks the parameter ‘deg’
any                       yes             lacks the parameter ‘out’
append                    no              no
apply_along_axis          no              no
apply_over_axes           no              no
arange                    yes             yes
arccos                    yes             lacks the parameter ‘out’
arccosh                   yes             lacks the parameter ‘out’
arcsin                    yes             lacks the parameter ‘out’
arcsinh                   yes             lacks the parameter ‘out’
arctan                    yes             lacks the parameter ‘out’
arctan2                   yes             lacks the parameter ‘out’
arctanh                   yes             lacks the parameter ‘out’
argmax                    yes             added the param ‘keepdims’
argmin                    yes             added the param ‘keepdims’
argpartition              no              no
argsort                   yes             yes
argwhere                  no              no
around                    no              no
array                     no              no
array_equal               no              no
array_equiv               no              no
array_split               no              no
asmatrix                  no              no
asscalar                  no              no
atleast_1d                no              no
atleast_2d                no              no
atleast_3d                no              no
average                   no              no
bartlett                  yes             yes
bincount                  yes             yes
bitwise_and               yes             lacks the parameter ‘out’
bitwise_not               yes             lacks the parameter ‘out’
bitwise_or                yes             lacks the parameter ‘out’
bitwise_xor               yes             lacks the parameter ‘out’
blackman                  no              no
bmat                      no              no
broadcast                 no              no
broadcast_arrays          no              no
byte                      no              no
byte_bounds               no              no
bytes_                    no              no
c_                        no              no
can_cast                  no              no
cast                      yes             no, numpy.cast is Base object for a dictionary for look-up with any alias for an array dtype.
cdouble                   no              no
ceil                      yes             lacks the parameter ‘out’
cfloat                    no              no
char                      no              no
character                 no              no
chararray                 no              no
choose                    yes             yes
clip                      yes             lacks the parameter ‘out’
clongdouble               no              no
clongfloat                no              no
column_stack              no              no
compat                    no              no
complex                   yes             yes
complex128                no              no
complex256                no              no
complex64                 no              no
complex_                  no              no
complexfloating           no              no
compress                  no              no
concatenate               yes             yes
conj                      yes             lacks the parameter ‘out’
conjugate                 no              no
convolve                  no              no
copy                      yes             lacks the param ‘order’
copysign                  no              no
copyto                    no              no
corrcoef                  no              no
correlate                 no              no
cos                       yes             lacks the parameter ‘out’
cosh                      yes             lacks the parameter ‘out’
count_nonzero             no              no
cov                       no              no
cross                     no              no
csingle                   no              no
cumprod                   yes             lacks the params ‘dtype’,’out’
cumproduct                no              no
cumsum                    yes             lacks the params ‘dtype’, ‘out’
datetime64                no              no
datetime_data             no              no
deg2rad                   yes             lacks the parameter ‘out’
degrees                   no              no
delete                    no              no
diag                      yes             yes
diag_indices              no              no
diag_indices_from         no              no
diagflat                  no              no
diagonal                  yes             yes
diff                      no              no
digitize                  no              no
disp                      no              no
distutils                 no              no
divide                    no              no
division                  no              no
dot                       yes             lacks the param ‘out’
double                    no              no
dsplit                    no              no
dstack                    no              no
dtype                     no              no
dual                      no              no
e                         no              no
ediff1d                   no              no
einsum                    no              no
emath                     no              no
empty                     no              no
empty_like                no              no
equal                     no              no
errstate                  no              no
euler_gamma               no              no
exp                       yes             lacks the param ’out’
exp2                      yes             lacks the param ‘out’
expand_dims               no              no
expm1                     yes             lacks the param ‘out’
extract                   no              no
eye                       yes             yes
fabs                      no              no
fft                       no              no
fill_diagonal             yes             lacks the param ‘wrap’
find_common_type          no              no
finfo                     no              no
fix                       no              no
flatiter                  no              no
flatnonzero               yes             yes
flexible                  no              no
fliplr                    no              no
flipud                    no              no
float                     no              no
float128                  no              no
float16                   no              no
float32                   no              no
float64                   no              no
float_                    no              no
floating                  no              no
floor                     yes             lacks the param ‘out’
floor_divide              no              no
fmax                      no              no
fmin                      no              no
fmod                      no              no
frexp                     no              no
frombuffer                no              no
fromfile                  no              no
fromfunction              no              no
fromiter                  no              no
frompyfunc                no              no
fromregex                 no              no
fromstring                no              no
full                      no              no
full_like                 no              no
fv                        no              no
generic                   no              no
genfromtxt                no              no
get_array_wrap            no              no
get_include               no              no
get_numarray_include      no              no
get_printoptions          no              no
getbuffer                 no              no
getbufsize                no              no
geterr                    no              no
geterrcall                no              no
geterrobj                 no              no
gradient                  no              no
greater                   no              no
greater_equal             no              no
half                      no              no
hamming                   no              no
hanning                   no              no
histogram                 no              no
histogram2d               no              no
histogramdd               no              no
hsplit                    no              no
hstack                    yes(horizontal_stack)              no
hypot                     no              no
i0                        no              no
identity                  no              no
iinfo                     no              no
imag                      yes             yes
in1d                      no              no
index_exp                 no              no
indices                   no              no
inexact                   no              no
inf                       no              no
info                      no              no
infty                     no              no
inner                     no              no
insert                    no              no
int                       no              no
int0                      no              no
int16                     no              no
int32                     no              no
int64                     no              no
int8                      no              no
int_                      no              no
int_asbuffer              no              no
intc                      no              no
integer                   no              no
interp                    no              no
intersect1d               no              no
intp                      no              no
invert                    yes             lacks the param ‘out’
ipmt                      no              no
irr                       no              no
is_busday                 no              no
isclose                   no              no
iscomplex                 no              no
iscomplexobj              no              no
isfinite                  no              no
isfortran                 no              no
isinf                     yes             lacks the param ‘out’ 
isnan                     yes             lacks the param ‘out’
isneginf                  no              no
isposinf                  no              no
isreal                    no              no
isrealobj                 no              no
isscalar                  no              no
issctype                  no              no
issubclass_               no              no
issubdtype                no              no
issubsctype               no              no
iterable                  no              no
ix_                       no              no
kaiser                    no              no
kron                      no              no
ldexp                     no              no
left_shift                no              no
less                      no              no
less_equal                no              no
lexsort                   no              no
lib                       no              no
linalg                    no              no
linspace                  no              no
little_endian             no              no
load                      yes             added params ‘dtype’,’broadcastable’, load an array from an .npy file, while numpy.load loads arrays or pickled objects from .npy, .npz or pickled files. 
loads                     no              no
loadtxt                   no              no
log                       yes             lacks the param ‘out’
log10                     yes             lacks the param ‘out’
log1p                     yes             lacks the param ‘out’
log2                      yes             lacks the param ‘out’
logaddexp                 no              no
logaddexp2                no              no
logical_and               no              no
logical_not               no              no
logical_or                no              no
logical_xor               no              no
logspace                  no              no
long                      no              no
longcomplex               no              no
longdouble                no              no
longfloat                 no              no
longlong                  no              no
lookfor                   no              no
ma                        no              no
mafromtxt                 no              no
mask_indices              no              no
mat                       no              no
math                      no              no
matrix                    yes             lacks the param ‘copy’
matrixlib                 no              no
max                       yes             lacks the param ‘out’
maximum                   yes             lacks the param ‘out’
maximum_sctype            no              no
may_share_memory          no              no
mean                      yes             added the params ‘op’, ‘acc_dtype’, lacking the param ‘out’
median                    no              no
memmap                    no              no
meshgrid                  no              no
mgrid                     no              no
min                       yes             lacks the param ‘out’
min_scalar_type           no              no
minimum                   yes             lacks the param ‘out’
mintypecode               no              no
mirr                      no              no
mod                       yes             lacks the param ‘out’
modf                      no              no
msort                     no              no
multiply                  no              no
nan                       no              no
nan_to_num                no              no
nanargmax                 no              no
nanargmin                 no              no
nanmax                    no              no
nanmean                   no              no
nanmin                    no              no
nanstd                    no              no
nansum                    no              no
nanprod                    no              no (added in dev version: https://github.com/numpy/numpy/pull/5418)
nanvar                    no              no
nbytes                    no              no
ndarray                   no              no
ndenumerate               no              no
ndfromtxt                 no              no
ndim                      no              no
ndindex                   no              no
nditer                    no              no
negative                  no              no
nested_iters              no              no
newaxis                   no              no
newbuffer                 no              no
nextafter                 no              no
nonzero                   yes             if param ‘retrun_matrix=False’, the same as Numpy, else the same as Theano Op
not_equal                 no              no
nper                      no              no
npv                       no              no
number                    no              no
obj2sctype                no              no
object                    no              no
object0                   no              no
object_                   no              no
ogrid                     no              no
ones                      yes             lacks the param ‘order’
ones_like                 yes             lacks the param ‘order’,’subok’
outer                     yes             lacks the param ‘out’
packbits                  no              no
pad                       no              no
partition                 no              no
percentile                no              no
pi                        no              no
piecewise                 no              no
pkgload                   no              no
place                     no              no
pmt                       no              no
poly                      no              no
poly1d                    no              no
polyadd                   no              no
polyder                   no              no
polydiv                   no              no
polyfit                   no              no
polyint                   no              no
polymul                   no              no
polynomial                no              no
polysub                   no              no
polyval                   no              no
power                     yes             lacks the param ‘out’
ppmt                      no              no
print_function            no              no
prod                      yes             added the params ‘acc_dtype’, ‘no_zeros_in_input’, lacks the param ‘out’
product                   no              no
promote_types             no              no
ptp                       yes             lacks the param ‘out’
put                       no              no
putmask                   no              no
pv                        no              no
r_                        no              no
rad2deg                   yes             lacks the param ‘out’
radians                   no              no
random                    yes             yes
rank                      no              no
rate                      no              no
ravel                     no              no
ravel_multi_index         no              no
real                      yes             yes
real_if_close             no              no
rec                       no              no
recarray                  no              no
recfromcsv                no              no
recfromtxt                no              no
reciprocal                no              no
record                    no              no
remainder                 no              no
repeat                    yes             yes
require                   no              no
reshape                   yes             added the params ‘ndim’, ‘name’, lacks the param ‘order’
resize                    no              no
restoredot                no              no
result_type               no              no
right_shift               no              no
rint                      no              no
roll                      yes             yes
rollaxis                  no              no
roots                     no              no
rot90                     no              no
round                     yes             added the param ‘mode’, lacks the params ‘decimals’ and ‘out’
round_                    no              no
row_stack                 no              no
s_                        no              no
safe_eval                 no              no
save                      no              no
savetxt                   no              no
savez                     no              no
savez_compressed          no              no
sctype2char               no              no
sctypeDict                no              no
sctypeNA                  no              no
sctypes                   no              no
searchsorted              no              no
select                    no              no
set_numeric_ops           no              no
set_printoptions          no              no
set_string_function       no              no
setbufsize                no              no
setdiff1d                 no              no
seterr                    no              no
seterrcall                no              no
seterrobj                 no              no
setxor1d                  no              no
shape                     yes             yes
short                     no              no
show_config               no              no
sign                      no              no
signbit                   no              no
signedinteger             no              no
sin                       yes             lacking the param ‘out’, uf
sinc                      no              no
single                    no              no
singlecomplex             no              no
sinh                      yes             lacking the param ‘out’, uf
size                      no              no
sometrue                  no              no
sort                      yes             yes
sort_complex              no              no
source                    no              no
spacing                   no              no
split                     yes             added the params ‘splits_size’ and ’n_splits’, lacks the param ‘indices_or_sections’
sqrt                      yes             lacks the param ‘out’
square                    yes             lacks the param ‘out’
squeeze                   yes             lacks the param ‘axis’
std                       yes             lacks the params ‘dtype’, ‘out’, ‘ddof’
str                       no              no
str_                      no              no
string0                   no              no
string_                   no              no
subtract                  no              no
sum                       yes             adding the param ‘acc_dtype’, lacks the param ‘out’ 
swapaxes                  yes             yes
sys                       yes             yes
take                      yes             lacks the param ‘out’
tan                       yes             lacks the param ‘out’
tanh                      yes             lacks the param ‘out’
tensordot                 yes             yes
test                      no              no
testing                   no              no
tile                      yes             added the param ‘ndim’
timedelta64               no              no
trace                     no              no
transpose                 yes             yes
trapz                     no              no
tri                       yes             yes
tril                      yes             yes
tril_indices              no              no
tril_indices_from         no              no
trim_zeros                no              no
triu                      yes             yes
triu_indices              no              no
triu_indices_from         no              no
true_divide               no              no
trunc                     yes             lacks the param ‘out’
typeDict                  no              no
typeNA                    no              no
typecodes                 no              no
typename                  no              no
ubyte                     no              no
ufunc                     no              no
uint                      no              no
uint0                     no              no
uint16                    no              no
uint32                    no              no
uint64                    no              no
uint8                     no              no
uintc                     no              no
uintp                     no              no
ulonglong                 no              no
unicode                   no              no
unicode0                  no              no
unicode_                  no              no
union1d                   no              no
unique                    no              no
unpackbits                no              no
unravel_index             no              no
unsignedinteger           no              no
unwrap                    no              no
use_fastnumpy             no              no
ushort                    no              no
vander                    no              no
var                       yes             lacks the params ‘dtype’, ’out’, ‘ddof’
vdot                      no              no
vectorize                 no              no
version                   no              no
void                      no              no
void0                     no              no
vsplit                    no              no
vstack                    yes (vertical_stack)              no
where                     yes             yes
who                       no              no
zeros                     yes             lacks the param ’order’ 
zeros_like                yes             lacks the params ‘order’,’subok’
Contributor

vdumoulin commented Jan 6, 2015

Here's the same list, with some false positives removed:

numpy.*                   theano.tensor.* interface compatible
abs                       no              no
absolute                  no              no
add                       yes             lacks the parameter ‘out’
alen                      no              no
all                       yes             lacks the parameter ‘out’
allclose                  no              no
alltrue                   no              no
alterdot                  no              no
amax                      no              no
amin                      no              no
angle                     yes             lacks the parameter ‘deg’
any                       yes             lacks the parameter ‘out’
append                    no              no
apply_along_axis          no              no
apply_over_axes           no              no
arange                    yes             yes
arccos                    yes             lacks the parameter ‘out’
arccosh                   yes             lacks the parameter ‘out’
arcsin                    yes             lacks the parameter ‘out’
arcsinh                   yes             lacks the parameter ‘out’
arctan                    yes             lacks the parameter ‘out’
arctan2                   yes             lacks the parameter ‘out’
arctanh                   yes             lacks the parameter ‘out’
argmax                    yes             added the param ‘keepdims’
argmin                    yes             added the param ‘keepdims’
argpartition              no              no
argsort                   yes             yes
argwhere                  no              no
around                    no              no
array                     no              no
array_equal               no              no
array_equiv               no              no
array_split               no              no
asmatrix                  no              no
asscalar                  no              no
atleast_1d                no              no
atleast_2d                no              no
atleast_3d                no              no
average                   no              no
bartlett                  yes             yes
bincount                  yes             yes
bitwise_and               yes             lacks the parameter ‘out’
bitwise_not               yes             lacks the parameter ‘out’
bitwise_or                yes             lacks the parameter ‘out’
bitwise_xor               yes             lacks the parameter ‘out’
blackman                  no              no
bmat                      no              no
broadcast                 no              no
broadcast_arrays          no              no
byte                      no              no
byte_bounds               no              no
bytes_                    no              no
c_                        no              no
can_cast                  no              no
cast                      yes             no, numpy.cast is Base object for a dictionary for look-up with any alias for an array dtype.
cdouble                   no              no
ceil                      yes             lacks the parameter ‘out’
cfloat                    no              no
char                      no              no
character                 no              no
chararray                 no              no
choose                    yes             yes
clip                      yes             lacks the parameter ‘out’
clongdouble               no              no
clongfloat                no              no
column_stack              no              no
compat                    no              no
complex                   yes             yes
complex128                no              no
complex256                no              no
complex64                 no              no
complex_                  no              no
complexfloating           no              no
compress                  no              no
concatenate               yes             yes
conj                      yes             lacks the parameter ‘out’
conjugate                 no              no
convolve                  no              no
copy                      yes             lacks the param ‘order’
copysign                  no              no
copyto                    no              no
corrcoef                  no              no
correlate                 no              no
cos                       yes             lacks the parameter ‘out’
cosh                      yes             lacks the parameter ‘out’
count_nonzero             no              no
cov                       no              no
cross                     no              no
csingle                   no              no
cumprod                   yes             lacks the params ‘dtype’,’out’
cumproduct                no              no
cumsum                    yes             lacks the params ‘dtype’, ‘out’
datetime64                no              no
datetime_data             no              no
deg2rad                   yes             lacks the parameter ‘out’
degrees                   no              no
delete                    no              no
diag                      yes             yes
diag_indices              no              no
diag_indices_from         no              no
diagflat                  no              no
diagonal                  yes             yes
diff                      no              no
digitize                  no              no
disp                      no              no
distutils                 no              no
divide                    no              no
division                  no              no
dot                       yes             lacks the param ‘out’
double                    no              no
dsplit                    no              no
dstack                    no              no
dtype                     no              no
dual                      no              no
e                         no              no
ediff1d                   no              no
einsum                    no              no
emath                     no              no
empty                     no              no
empty_like                no              no
equal                     no              no
errstate                  no              no
euler_gamma               no              no
exp                       yes             lacks the param ’out’
exp2                      yes             lacks the param ‘out’
expand_dims               no              no
expm1                     yes             lacks the param ‘out’
extract                   no              no
eye                       yes             yes
fabs                      no              no
fft                       no              no
fill_diagonal             yes             lacks the param ‘wrap’
find_common_type          no              no
finfo                     no              no
fix                       no              no
flatiter                  no              no
flatnonzero               yes             yes
flexible                  no              no
fliplr                    no              no
flipud                    no              no
float                     no              no
float128                  no              no
float16                   no              no
float32                   no              no
float64                   no              no
float_                    no              no
floating                  no              no
floor                     yes             lacks the param ‘out’
floor_divide              no              no
fmax                      no              no
fmin                      no              no
fmod                      no              no
frexp                     no              no
frombuffer                no              no
fromfile                  no              no
fromfunction              no              no
fromiter                  no              no
frompyfunc                no              no
fromregex                 no              no
fromstring                no              no
full                      no              no
full_like                 no              no
fv                        no              no
generic                   no              no
genfromtxt                no              no
get_array_wrap            no              no
get_include               no              no
get_numarray_include      no              no
get_printoptions          no              no
getbuffer                 no              no
getbufsize                no              no
geterr                    no              no
geterrcall                no              no
geterrobj                 no              no
gradient                  no              no
greater                   no              no
greater_equal             no              no
half                      no              no
hamming                   no              no
hanning                   no              no
histogram                 no              no
histogram2d               no              no
histogramdd               no              no
hsplit                    no              no
hstack                    yes(horizontal_stack)              no
hypot                     no              no
i0                        no              no
identity                  no              no
iinfo                     no              no
imag                      yes             yes
in1d                      no              no
index_exp                 no              no
indices                   no              no
inexact                   no              no
inf                       no              no
info                      no              no
infty                     no              no
inner                     no              no
insert                    no              no
int                       no              no
int0                      no              no
int16                     no              no
int32                     no              no
int64                     no              no
int8                      no              no
int_                      no              no
int_asbuffer              no              no
intc                      no              no
integer                   no              no
interp                    no              no
intersect1d               no              no
intp                      no              no
invert                    yes             lacks the param ‘out’
ipmt                      no              no
irr                       no              no
is_busday                 no              no
isclose                   no              no
iscomplex                 no              no
iscomplexobj              no              no
isfinite                  no              no
isfortran                 no              no
isinf                     yes             lacks the param ‘out’ 
isnan                     yes             lacks the param ‘out’
isneginf                  no              no
isposinf                  no              no
isreal                    no              no
isrealobj                 no              no
isscalar                  no              no
issctype                  no              no
issubclass_               no              no
issubdtype                no              no
issubsctype               no              no
iterable                  no              no
ix_                       no              no
kaiser                    no              no
kron                      no              no
ldexp                     no              no
left_shift                no              no
less                      no              no
less_equal                no              no
lexsort                   no              no
lib                       no              no
linalg                    no              no
linspace                  no              no
little_endian             no              no
load                      yes             added params ‘dtype’,’broadcastable’, load an array from an .npy file, while numpy.load loads arrays or pickled objects from .npy, .npz or pickled files. 
loads                     no              no
loadtxt                   no              no
log                       yes             lacks the param ‘out’
log10                     yes             lacks the param ‘out’
log1p                     yes             lacks the param ‘out’
log2                      yes             lacks the param ‘out’
logaddexp                 no              no
logaddexp2                no              no
logical_and               no              no
logical_not               no              no
logical_or                no              no
logical_xor               no              no
logspace                  no              no
long                      no              no
longcomplex               no              no
longdouble                no              no
longfloat                 no              no
longlong                  no              no
lookfor                   no              no
ma                        no              no
mafromtxt                 no              no
mask_indices              no              no
mat                       no              no
math                      no              no
matrix                    yes             lacks the param ‘copy’
matrixlib                 no              no
max                       yes             lacks the param ‘out’
maximum                   yes             lacks the param ‘out’
maximum_sctype            no              no
may_share_memory          no              no
mean                      yes             added the params ‘op’, ‘acc_dtype’, lacking the param ‘out’
median                    no              no
memmap                    no              no
meshgrid                  no              no
mgrid                     no              no
min                       yes             lacks the param ‘out’
min_scalar_type           no              no
minimum                   yes             lacks the param ‘out’
mintypecode               no              no
mirr                      no              no
mod                       yes             lacks the param ‘out’
modf                      no              no
msort                     no              no
multiply                  no              no
nan                       no              no
nan_to_num                no              no
nanargmax                 no              no
nanargmin                 no              no
nanmax                    no              no
nanmean                   no              no
nanmin                    no              no
nanstd                    no              no
nansum                    no              no
nanprod                    no              no (added in dev version: https://github.com/numpy/numpy/pull/5418)
nanvar                    no              no
nbytes                    no              no
ndarray                   no              no
ndenumerate               no              no
ndfromtxt                 no              no
ndim                      no              no
ndindex                   no              no
nditer                    no              no
negative                  no              no
nested_iters              no              no
newaxis                   no              no
newbuffer                 no              no
nextafter                 no              no
nonzero                   yes             if param ‘retrun_matrix=False’, the same as Numpy, else the same as Theano Op
not_equal                 no              no
nper                      no              no
npv                       no              no
number                    no              no
obj2sctype                no              no
object                    no              no
object0                   no              no
object_                   no              no
ogrid                     no              no
ones                      yes             lacks the param ‘order’
ones_like                 yes             lacks the param ‘order’,’subok’
outer                     yes             lacks the param ‘out’
packbits                  no              no
pad                       no              no
partition                 no              no
percentile                no              no
pi                        no              no
piecewise                 no              no
pkgload                   no              no
place                     no              no
pmt                       no              no
poly                      no              no
poly1d                    no              no
polyadd                   no              no
polyder                   no              no
polydiv                   no              no
polyfit                   no              no
polyint                   no              no
polymul                   no              no
polynomial                no              no
polysub                   no              no
polyval                   no              no
power                     yes             lacks the param ‘out’
ppmt                      no              no
print_function            no              no
prod                      yes             added the params ‘acc_dtype’, ‘no_zeros_in_input’, lacks the param ‘out’
product                   no              no
promote_types             no              no
ptp                       yes             lacks the param ‘out’
put                       no              no
putmask                   no              no
pv                        no              no
r_                        no              no
rad2deg                   yes             lacks the param ‘out’
radians                   no              no
random                    yes             yes
rank                      no              no
rate                      no              no
ravel                     no              no
ravel_multi_index         no              no
real                      yes             yes
real_if_close             no              no
rec                       no              no
recarray                  no              no
recfromcsv                no              no
recfromtxt                no              no
reciprocal                no              no
record                    no              no
remainder                 no              no
repeat                    yes             yes
require                   no              no
reshape                   yes             added the params ‘ndim’, ‘name’, lacks the param ‘order’
resize                    no              no
restoredot                no              no
result_type               no              no
right_shift               no              no
rint                      no              no
roll                      yes             yes
rollaxis                  no              no
roots                     no              no
rot90                     no              no
round                     yes             added the param ‘mode’, lacks the params ‘decimals’ and ‘out’
round_                    no              no
row_stack                 no              no
s_                        no              no
safe_eval                 no              no
save                      no              no
savetxt                   no              no
savez                     no              no
savez_compressed          no              no
sctype2char               no              no
sctypeDict                no              no
sctypeNA                  no              no
sctypes                   no              no
searchsorted              no              no
select                    no              no
set_numeric_ops           no              no
set_printoptions          no              no
set_string_function       no              no
setbufsize                no              no
setdiff1d                 no              no
seterr                    no              no
seterrcall                no              no
seterrobj                 no              no
setxor1d                  no              no
shape                     yes             yes
short                     no              no
show_config               no              no
sign                      no              no
signbit                   no              no
signedinteger             no              no
sin                       yes             lacking the param ‘out’, uf
sinc                      no              no
single                    no              no
singlecomplex             no              no
sinh                      yes             lacking the param ‘out’, uf
size                      no              no
sometrue                  no              no
sort                      yes             yes
sort_complex              no              no
source                    no              no
spacing                   no              no
split                     yes             added the params ‘splits_size’ and ’n_splits’, lacks the param ‘indices_or_sections’
sqrt                      yes             lacks the param ‘out’
square                    yes             lacks the param ‘out’
squeeze                   yes             lacks the param ‘axis’
std                       yes             lacks the params ‘dtype’, ‘out’, ‘ddof’
str                       no              no
str_                      no              no
string0                   no              no
string_                   no              no
subtract                  no              no
sum                       yes             adding the param ‘acc_dtype’, lacks the param ‘out’ 
swapaxes                  yes             yes
sys                       yes             yes
take                      yes             lacks the param ‘out’
tan                       yes             lacks the param ‘out’
tanh                      yes             lacks the param ‘out’
tensordot                 yes             yes
test                      no              no
testing                   no              no
tile                      yes             added the param ‘ndim’
timedelta64               no              no
trace                     no              no
transpose                 yes             yes
trapz                     no              no
tri                       yes             yes
tril                      yes             yes
tril_indices              no              no
tril_indices_from         no              no
trim_zeros                no              no
triu                      yes             yes
triu_indices              no              no
triu_indices_from         no              no
true_divide               no              no
trunc                     yes             lacks the param ‘out’
typeDict                  no              no
typeNA                    no              no
typecodes                 no              no
typename                  no              no
ubyte                     no              no
ufunc                     no              no
uint                      no              no
uint0                     no              no
uint16                    no              no
uint32                    no              no
uint64                    no              no
uint8                     no              no
uintc                     no              no
uintp                     no              no
ulonglong                 no              no
unicode                   no              no
unicode0                  no              no
unicode_                  no              no
union1d                   no              no
unique                    no              no
unpackbits                no              no
unravel_index             no              no
unsignedinteger           no              no
unwrap                    no              no
use_fastnumpy             no              no
ushort                    no              no
vander                    no              no
var                       yes             lacks the params ‘dtype’, ’out’, ‘ddof’
vdot                      no              no
vectorize                 no              no
version                   no              no
void                      no              no
void0                     no              no
vsplit                    no              no
vstack                    yes (vertical_stack)              no
where                     yes             yes
who                       no              no
zeros                     yes             lacks the param ’order’ 
zeros_like                yes             lacks the params ‘order’,’subok’
@nouiz

This comment has been minimized.

Show comment
Hide comment
@nouiz

nouiz Jan 6, 2015

Member

I added in the last comment nanprod that just got added in numpy.

Member

nouiz commented Jan 6, 2015

I added in the last comment nanprod that just got added in numpy.

@lamblin

This comment has been minimized.

Show comment
Hide comment
@lamblin

lamblin Jan 21, 2015

Member

A couple of general remarks:

  • the 'out' arg for elemwise functions does not make sense in Theano, because we cannot change or update existing variables, we are always creating new ones
  • for the moment, the parameters 'order' does not make sense, as we try not to make assumptions about the C/Fortran order of the underlying memory, but I guess that could change in the future
  • 'subok' is not really applicable to tensors either, as we do not have a subclass of Tensor.
  • objects in the numpy space (as opposed to functions), such as int do not have to be in theano.tensor
  • also, functions that do not represent mathematical operations on arrays, such as array, fromfile will not have equivalent.
Member

lamblin commented Jan 21, 2015

A couple of general remarks:

  • the 'out' arg for elemwise functions does not make sense in Theano, because we cannot change or update existing variables, we are always creating new ones
  • for the moment, the parameters 'order' does not make sense, as we try not to make assumptions about the C/Fortran order of the underlying memory, but I guess that could change in the future
  • 'subok' is not really applicable to tensors either, as we do not have a subclass of Tensor.
  • objects in the numpy space (as opposed to functions), such as int do not have to be in theano.tensor
  • also, functions that do not represent mathematical operations on arrays, such as array, fromfile will not have equivalent.
@nouiz

This comment has been minimized.

Show comment
Hide comment
@nouiz

nouiz Jan 21, 2015

Member

Why not wrap fromfile? The users would do it most of the time outside
Theano, so I won't spend time wrapping them, but if someone do it, I would
accept the PR. Do agree with that?

On Wed, Jan 21, 2015 at 11:42 AM, Pascal Lamblin notifications@github.com
wrote:

A couple of general remarks:

  • the 'out' arg for elemwise functions does not make sense in Theano,
    because we cannot change or update existing variables, we are always
    creating new ones
  • for the moment, the parameters 'order' does not make sense, as we
    try not to make assumptions about the C/Fortran order of the underlying
    memory, but I guess that could change in the future
  • 'subok' is not really applicable to tensors either, as we do not
    have a subclass of Tensor.
  • objects in the numpy space (as opposed to functions), such as int do
    not have to be in theano.tensor
  • also, functions that do not represent mathematical operations on
    arrays, such as array, fromfile will not have equivalent.


Reply to this email directly or view it on GitHub
#1916 (comment).

Member

nouiz commented Jan 21, 2015

Why not wrap fromfile? The users would do it most of the time outside
Theano, so I won't spend time wrapping them, but if someone do it, I would
accept the PR. Do agree with that?

On Wed, Jan 21, 2015 at 11:42 AM, Pascal Lamblin notifications@github.com
wrote:

A couple of general remarks:

  • the 'out' arg for elemwise functions does not make sense in Theano,
    because we cannot change or update existing variables, we are always
    creating new ones
  • for the moment, the parameters 'order' does not make sense, as we
    try not to make assumptions about the C/Fortran order of the underlying
    memory, but I guess that could change in the future
  • 'subok' is not really applicable to tensors either, as we do not
    have a subclass of Tensor.
  • objects in the numpy space (as opposed to functions), such as int do
    not have to be in theano.tensor
  • also, functions that do not represent mathematical operations on
    arrays, such as array, fromfile will not have equivalent.


Reply to this email directly or view it on GitHub
#1916 (comment).

@lamblin

This comment has been minimized.

Show comment
Hide comment
@lamblin

lamblin Jan 22, 2015

Member

What do you mean by wrapping? numpy.fromfile returns an ndarray from a file. What would the theano equivalent be? Loading a symbolic tensor from a file does not really make sense, as it has no value.
Or do you mean wrapping it as a Theano Op, that loads from the disk? In that case, yes, that might work.

Member

lamblin commented Jan 22, 2015

What do you mean by wrapping? numpy.fromfile returns an ndarray from a file. What would the theano equivalent be? Loading a symbolic tensor from a file does not really make sense, as it has no value.
Or do you mean wrapping it as a Theano Op, that loads from the disk? In that case, yes, that might work.

@nouiz

This comment has been minimized.

Show comment
Hide comment
@nouiz

nouiz Jan 22, 2015

Member

I mean the last case.

On Wed, Jan 21, 2015 at 4:57 PM, Pascal Lamblin notifications@github.com
wrote:

What do you mean by wrapping? numpy.fromfile returns an ndarray from a
file. What would the theano equivalent be? Loading a symbolic tensor from a
file does not really make sense, as it has no value.
Or do you mean wrapping it as a Theano Op, that loads from the disk? In
that case, yes, that might work.


Reply to this email directly or view it on GitHub
#1916 (comment).

Member

nouiz commented Jan 22, 2015

I mean the last case.

On Wed, Jan 21, 2015 at 4:57 PM, Pascal Lamblin notifications@github.com
wrote:

What do you mean by wrapping? numpy.fromfile returns an ndarray from a
file. What would the theano equivalent be? Loading a symbolic tensor from a
file does not really make sense, as it has no value.
Or do you mean wrapping it as a Theano Op, that loads from the disk? In
that case, yes, that might work.


Reply to this email directly or view it on GitHub
#1916 (comment).

@nouiz

This comment has been minimized.

Show comment
Hide comment
@nouiz

nouiz Jun 16, 2015

Member

unique is done.

Member

nouiz commented Jun 16, 2015

unique is done.

@Sentient07

This comment has been minimized.

Show comment
Hide comment
@Sentient07

Sentient07 Mar 3, 2016

Contributor

@nouiz , could i take this if no one is working on this currently ?

Contributor

Sentient07 commented Mar 3, 2016

@nouiz , could i take this if no one is working on this currently ?

@AnishShah

This comment has been minimized.

Show comment
Hide comment
@AnishShah

AnishShah Mar 7, 2016

Contributor

@Sentient07 Are you working on this? Because I would like to work on this.
I will try to work on argwhere.

Contributor

AnishShah commented Mar 7, 2016

@Sentient07 Are you working on this? Because I would like to work on this.
I will try to work on argwhere.

@Sentient07

This comment has been minimized.

Show comment
Hide comment
@Sentient07

Sentient07 Mar 7, 2016

Contributor

@AnishShah I need confirmation from @nouiz regarding what is to be done and who is working on it. He will assign when he is back.

Contributor

Sentient07 commented Mar 7, 2016

@AnishShah I need confirmation from @nouiz regarding what is to be done and who is working on it. He will assign when he is back.

@AnishShah

This comment has been minimized.

Show comment
Hide comment
@AnishShah

AnishShah Mar 7, 2016

Contributor

No problem 👍

Contributor

AnishShah commented Mar 7, 2016

No problem 👍

@shiretzet

This comment has been minimized.

Show comment
Hide comment
@shiretzet

shiretzet Nov 16, 2016

I've implemented the "cov" function, but it has an issue with GPU.
See #5233

I've implemented the "cov" function, but it has an issue with GPU.
See #5233

@nouiz

This comment has been minimized.

Show comment
Hide comment
@nouiz

nouiz Nov 16, 2016

Member

Can you make a pr with the code? It help debug the problem.

Le 16 nov. 2016 03:06, "shiretzet" notifications@github.com a écrit :

I've implemented the "cov" function, but it has an issue with GPU.
See #5233 #5233


You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#1916 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AALC-zUDFqCF38yUR79zzQDfkQhfshKmks5q-rmigaJpZM4CEBfq
.

Member

nouiz commented Nov 16, 2016

Can you make a pr with the code? It help debug the problem.

Le 16 nov. 2016 03:06, "shiretzet" notifications@github.com a écrit :

I've implemented the "cov" function, but it has an issue with GPU.
See #5233 #5233


You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#1916 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AALC-zUDFqCF38yUR79zzQDfkQhfshKmks5q-rmigaJpZM4CEBfq
.

@f0k f0k referenced this issue in Lasagne/Lasagne Apr 11, 2017

Closed

tensor.median? #827

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment