Skip to content

P
pytensor
提交 bd5dc5a8 authored 作者: Li's avatar Li
浏览文件
操作

pep8, and grad finally worked

上级 33551a31
隐藏空白字符变更
内嵌 并排
正在显示 包含 55 行增加 和 45 行删除
theano/tensor/signal/downsample.py
浏览文件 @ bd5dc5a8
......@@ -12,13 +12,14 @@ import numpy
import theano
from theano import gof, Op, tensor, Variable, Apply
def max_pool2D(*args, **kwargs):
import sys
print >> sys.stderr, "DEPRECATION: max_pool2D renamed to max_pool_2d"
return max_pool_2d(*args, **kwargs)
def max_pool_2d(input, ds, ignore_border=False, st=None, padding=(0,0)):
def max_pool_2d(input, ds, ignore_border=False, st=None, padding=(0, 0)):
"""
Takes as input a N-D tensor, where N >= 2. It downscales the input image by
the specified factor, by keeping only the maximum value of non-overlapping
......@@ -79,7 +80,7 @@ class DownsampleFactorMax(Op):
__props__ = ('ds', 'ignore_border', 'st', 'padding')
@staticmethod
def out_shape(imgshape, ds, ignore_border=False, st=None, padding=(0,0)):
def out_shape(imgshape, ds, ignore_border=False, st=None, padding=(0, 0)):
"""Return the shape of the output from this op, for input of given
shape and flags.
......@@ -100,8 +101,9 @@ class DownsampleFactorMax(Op):
extra row/col of partial downsampling (False) or ignore it (True).
:type ignore_border: bool
:param padding: pad zeros on four borders of the images
:type padding: tuple of two ints
:param padding: (pad_h, pad_w), pad zeros on four borders
of the images, pad_h for padding rows, and pad_w for columns.
:type padding: tuple of two ints
:rtype: list
:returns: the shape of the output from this op, for input of given
......@@ -117,7 +119,7 @@ class DownsampleFactorMax(Op):
r, c = imgshape[-2:]
r += padding[0] * 2
c += padding[1] * 2
if ignore_border:
out_r = (r - ds[0]) // st[0] + 1
out_c = (c - ds[1]) // st[1] + 1
......@@ -153,7 +155,7 @@ class DownsampleFactorMax(Op):
rval = list(imgshape[:-2]) + [nr, nc]
return rval
def __init__(self, ds, ignore_border=False, st=None, padding=(0,0)):
def __init__(self, ds, ignore_border=False, st=None, padding=(0, 0)):
"""
:param ds: downsample factor over rows and column.
ds indicates the pool region size.
......@@ -170,8 +172,9 @@ class DownsampleFactorMax(Op):
(no overlap on pooling regions)
: type st: list or tuple of two ints
:param padding: pad zeros on four borders of the images
:type padding: tuple of two ints
:param padding: (pad_h, pad_w), pad zeros on four borders
of the images, pad_h for padding rows, and pad_w for columns.
:type padding: tuple of two ints
"""
self.ds = tuple(ds)
......@@ -185,13 +188,16 @@ class DownsampleFactorMax(Op):
self.ignore_border = ignore_border
self.padding = tuple(padding)
self.padding = padding
if padding != (0,0) and not ignore_border:
raise NotImplementedError('padding works only with ignore_boarder=True')
if padding != (0, 0) and not ignore_border:
raise NotImplementedError(
'padding works only with ignore_boarder=True')
if self.padding[0] >= self.ds[0] or self.padding[1] >= self.ds[1]:
raise NotImplementedError('padding_h and padding_w must be smaller than strides')
raise NotImplementedError(
'padding_h and padding_w must be smaller than strides')
def __str__(self):
return '%s{%s, %s, %s, %s}' % (self.__class__.__name__,
self.ds, self.st, self.ignore_border,self.padding)
self.ds, self.st, self.ignore_border, self.padding)
def make_node(self, x):
if x.type.ndim != 4:
......@@ -214,7 +220,6 @@ class DownsampleFactorMax(Op):
self.ignore_border, self.st, self.padding),
dtype=x.dtype)
zz = z[0]
#number of pooling output rows
pr = zz.shape[-2]
#number of pooling output cols
......@@ -228,7 +233,9 @@ class DownsampleFactorMax(Op):
# pad the image
fill = x.min()-1.
y = numpy.zeros((x.shape[0], x.shape[1], img_rows, img_cols), dtype=x.dtype) + fill
y = numpy.zeros(
(x.shape[0], x.shape[1], img_rows, img_cols),
dtype=x.dtype) + fill
y[:, :, pad_h:(img_rows-pad_h), pad_w:(img_cols-pad_w)] = x
# max pooling
for n in xrange(x.shape[0]):
......@@ -253,7 +260,7 @@ class DownsampleFactorMax(Op):
maxout = self(x)
return [DownsampleFactorMaxGrad(self.ds,
ignore_border=self.ignore_border,
st=self.st)(
st=self.st, padding=self.padding)(
x, maxout, gz)]
def c_code(self, node, name, inp, out, sub):
......@@ -261,7 +268,7 @@ class DownsampleFactorMax(Op):
# the stride size and the pooling size are different.
# An exception is raised for such a case.
if self.ds != self.st:
raise theano.gof.utils.MethodNotDefined()
raise theano.gof.utils.MethodNotDefined()
x, = inp
z, = out
fail = sub['fail']
......@@ -340,7 +347,7 @@ class DownsampleFactorMax(Op):
class DownsampleFactorMaxGrad(Op):
__props__ = ('ds', 'ignore_border', 'st', 'padding')
def __init__(self, ds, ignore_border, st=None, padding=(0,0)):
def __init__(self, ds, ignore_border, st=None, padding=(0, 0)):
self.ds = tuple(ds)
self.ignore_border = ignore_border
if st is None:
......@@ -350,7 +357,7 @@ class DownsampleFactorMaxGrad(Op):
def __str__(self):
return '%s{%s, %s, %s, %s}' % (self.__class__.__name__,
self.ds, self.st, self.ignore_border,self.padding)
self.ds, self.st, self.ignore_border, self.padding)
def make_node(self, x, maxout, gz):
# make_node should only be called by the grad function of
......@@ -364,7 +371,6 @@ class DownsampleFactorMaxGrad(Op):
def perform(self, node, inp, out):
x, maxout, gz = inp
gx_stg, = out
#number of pooling output rows
pr = maxout.shape[-2]
#number of pooling output cols
......@@ -375,11 +381,10 @@ class DownsampleFactorMaxGrad(Op):
img_cols = x.shape[-1] + 2 * self.padding[1]
pad_h = self.padding[0]
pad_w = self.padding[1]
# pad the image
fill = x.min()-1
y = numpy.zeros((x.shape[0], x.shape[1], img_rows, img_cols), dtype=x.dtype) + fill
y = numpy.zeros(
(x.shape[0], x.shape[1], img_rows, img_cols), dtype=x.dtype) + fill
y[:, :, pad_h:(img_rows-pad_h), pad_w:(img_cols-pad_w)] = x
gx = numpy.zeros_like(y)
for n in xrange(x.shape[0]):
......@@ -408,15 +413,17 @@ class DownsampleFactorMaxGrad(Op):
return [theano.tensor.zeros_like(x),
theano.tensor.zeros_like(maxout),
DownsampleFactorMaxGradGrad(
self.ds, ignore_border=self.ignore_border, st=self.st)(x, maxout, ggx)]
self.ds, ignore_border=self.ignore_border,
st=self.st)(x, maxout, ggx)]
else:
return [theano.tensor.zeros_like(x),
theano.tensor.zeros_like(maxout),
theano.gradients.grad_not_implemented(
self, 2, gz,'Hessian not implemented with padding')]
self, 2, gz, 'Hessian not implemented with padding')]
def c_code(self, node, name, inp, out, sub):
if self.ds != self.st:
raise theano.gof.utils.MethodNotDefined()
raise theano.gof.utils.MethodNotDefined()
x, z, gz = inp
gx, = out
fail = sub['fail']
......
theano/tensor/signal/tests/test_downsample.py
浏览文件 @ bd5dc5a8
......@@ -37,22 +37,24 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
patch = input[k][ii:ii + ds[0], jj:jj + ds[1]]
output_val[k][i, j] = numpy.max(patch)
return output_val
@staticmethod
def numpy_max_pool_2d_stride_padding(
x, ds, ignore_border=True, st=None, padding=(0,0)):
x, ds, ignore_border=True, st=None, padding=(0, 0)):
pad_h = padding[0]
pad_w = padding[1]
h = x.shape[-2]
w = x.shape[-1]
assert ds[0] > pad_h
assert ds[1] > pad_w
assert ds[1] > pad_w
def pad_img(x):
fill = x.min()-1
t = numpy.ones((x.shape[0],x.shape[1],1,1))
t = numpy.ones((x.shape[0], x.shape[1], 1, 1))
ud_bar = (numpy.zeros((pad_h, w)) + fill)[
numpy.newaxis, numpy.newaxis,:,:] * t
numpy.newaxis, numpy.newaxis, :, :] * t
lr_bar = (numpy.zeros((pad_h * 2 + h, pad_w)) + fill)[
numpy.newaxis, numpy.newaxis,:,:] * t
numpy.newaxis, numpy.newaxis, :, :] * t
y = numpy.concatenate([ud_bar, x, ud_bar], axis=2)
y = numpy.concatenate([lr_bar, y, lr_bar], axis=3)
return y
......@@ -64,7 +66,7 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
out_shp.append(out_r)
out_shp.append(out_c)
ds0, ds1 = ds
st0, st1 = st
st0, st1 = st
output_val = numpy.zeros(out_shp)
tt = []
y = pad_img(x)
......@@ -236,14 +238,14 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
f = function([images], maxpool_op)
output_val = f(imval)
utt.assert_allclose(output_val, numpy_output_val)
def test_DownsampleFactorMaxPaddingStride(self):
ignore_border = True # padding does not support ignore_border=False
ignore_border = True # padding does not support ignore_border=False
rng = numpy.random.RandomState(utt.fetch_seed())
maxpoolsizes = [(3, 3), (4,4), (3,4), (4,3)]
stridesizes = [(2, 2), (2,2), (1,1), (1,2)]
paddingsizes = [(2, 2), (1,2), (2,1), (0,0)]
imgsizes = [(5, 5), (5,5), (5,6), (6,5)]
maxpoolsizes = [(3, 3), (4, 4), (3, 4), (4, 3)]
stridesizes = [(2, 2), (2, 2), (1, 1), (1, 2)]
paddingsizes = [(2, 2), (1, 2), (2, 1), (0, 0)]
imgsizes = [(5, 5), (5, 5), (5, 6), (6, 5)]
m = 4 # minibatch
c = 10 # channel size
images = tensor.dtensor4()
......@@ -254,21 +256,22 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
maxpoolsize = maxpoolsizes[indx]
paddingsize = paddingsizes[indx]
numpy_output_val = self.numpy_max_pool_2d_stride_padding(
imval, maxpoolsize,ignore_border, stridesize, paddingsize)
maxpool_op = DownsampleFactorMax(maxpoolsize,
ignore_border=ignore_border,
st=stridesize,padding=paddingsize)(images)
imval, maxpoolsize, ignore_border, stridesize, paddingsize)
maxpool_op = DownsampleFactorMax(
maxpoolsize,
ignore_border=ignore_border,
st=stridesize, padding=paddingsize)(images)
f = function([images], maxpool_op)
output_val = f(imval)
utt.assert_allclose(output_val, numpy_output_val)
def test_DownsampleFactorMaxPaddingStride_grad(self):
rng = numpy.random.RandomState(utt.fetch_seed())
imval = rng.rand(1, 1, 10, 10) * 10.0
maxpoolsize = (5, 3)
stridesize = (3, 2)
paddingsize = (2,2)
paddingsize = (2, 2)
def mp(input):
return DownsampleFactorMax(
maxpoolsize, ignore_border=True,
......@@ -276,7 +279,7 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
padding=paddingsize,
)(input)
utt.verify_grad(mp, [imval], rng=rng)
def test_DownsampleFactorMax_grad(self):
rng = numpy.random.RandomState(utt.fetch_seed())
maxpoolshps = ((1, 1), (3, 2), (2, 3))
......
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论