Skip to content

P
pytensor
提交 9dc07802 authored 作者: abergeron's avatar abergeron
浏览文件
操作

Merge pull request #2783 from nouiz/pool_average

Average pool CPU with python code
上级 54363a8d 8df6d348
隐藏空白字符变更
内嵌 并排
正在显示 包含 251 行增加 和 148 行删除
theano/sandbox/cuda/dnn.py
浏览文件 @ 9dc07802
......@@ -721,7 +721,8 @@ class GpuDnnPoolDesc(GpuOp):
:param ws: windows size
:param stride: (dx, dy)
:param mode: 'max' or 'average'
:param mode: 'max', 'average_inc_pad' or 'average_exc_pad'
The old deprecated name 'average' correspond to 'average_inc_pad'
:param pad: (padX, padY) padding information.
padX is the size of the left and right borders,
padY is the size of the top and bottom borders.
......@@ -744,7 +745,9 @@ class GpuDnnPoolDesc(GpuOp):
return False
def __init__(self, ws=(1, 1), stride=(1, 1), mode='max', pad=(0, 0)):
assert mode in ('max', 'average')
if mode == 'average':
mode = 'average_inc_pad'
assert mode in ('max', 'average_inc_pad', 'average_exc_pad')
self.mode = mode
assert len(ws) == 2
self.ws = ws
......@@ -772,8 +775,12 @@ class GpuDnnPoolDesc(GpuOp):
if self.mode == 'max':
mode_flag = 'CUDNN_POOLING_MAX'
elif self.mode == "average":
elif self.mode == "average_inc_pad":
mode_flag = 'CUDNN_POOLING_AVERAGE_COUNT_INCLUDE_PADDING'
elif self.mode == "average_exc_pad":
mode_flag = 'CUDNN_POOLING_AVERAGE_COUNT_EXCLUDE_PADDING'
if version() == -1:
raise Exception("cudnn v1 do not support average_exc_pad")
else:
raise NotImplementedError("Unsupported pooling model.")
......@@ -1194,7 +1201,8 @@ def dnn_pool(img, ws, stride=(1, 1), mode='max', pad=(0, 0)):
:param img: images to do the pooling over
:param ws: subsampling window size
:param stride: subsampling stride (default: (1, 1))
:param mode: one of 'max', 'average' (default: 'max')
:param mode: one of 'max', 'average_inc_pad' or 'average_exc_pad
(default: 'max')
:param pad: (padX, padY) padding information.
padX is the size of the left and right borders,
padY is the size of the top and bottom borders.
......@@ -1625,7 +1633,7 @@ if True:
@register_opt('cudnn')
@local_optimizer([DownsampleFactorMax])
def local_pool_dnn_stride(node):
def local_pool_dnn_alternative(node):
if not dnn_available():
return
if isinstance(node.op, DownsampleFactorMax):
......@@ -1635,9 +1643,10 @@ if True:
ds = node.op.ds
stride = node.op.st
pad = node.op.padding
mode = node.op.mode
if (img.owner and isinstance(img.owner.op, HostFromGpu)):
ret = dnn_pool(gpu_contiguous(img.owner.inputs[0]),
ds, stride=stride, pad=pad)
ds, stride=stride, pad=pad, mode=mode)
return [host_from_gpu(ret)]
@register_opt('cudnn')
......@@ -1667,12 +1676,13 @@ if True:
ds = node.op.ds
st = node.op.st
pad = node.op.padding
mode = node.op.mode
if ((inp.owner and isinstance(inp.owner.op, HostFromGpu)) or
(out.owner and isinstance(out.owner.op, HostFromGpu)) or
(inp_grad.owner and isinstance(inp_grad.owner.op,
HostFromGpu))):
desc = GpuDnnPoolDesc(ws=ds, stride=st, mode="max", pad=pad)()
desc = GpuDnnPoolDesc(ws=ds, stride=st, mode=mode, pad=pad)()
if not node.op.ignore_border:
return
ret = GpuDnnPoolGrad()(gpu_contiguous(inp),
......
theano/sandbox/cuda/opt.py
浏览文件 @ 9dc07802
......@@ -1648,8 +1648,9 @@ import theano.tensor.signal.downsample as downsample
def local_gpu_downsample_factor_max(node):
if (isinstance(node.op, downsample.DownsampleFactorMax)
and node.op.ds == node.op.st):
assert node.op.__props__ == ('ds', 'ignore_border', 'st', 'padding')
if node.op.padding != (0, 0):
assert node.op.__props__ == ('ds', 'ignore_border', 'st', 'padding',
'mode')
if node.op.padding != (0, 0) or node.op.mode != 'max':
return
x, = node.inputs
if (x.owner and isinstance(x.owner.op, HostFromGpu)):
......@@ -1662,8 +1663,9 @@ def local_gpu_downsample_factor_max(node):
def local_gpu_downsample_factor_max_grad(node):
if (isinstance(node.op, downsample.DownsampleFactorMaxGrad) and
node.op.ds == node.op.st):
assert node.op.__props__ == ('ds', 'ignore_border', 'st', 'padding')
if node.op.padding != (0, 0):
assert node.op.__props__ == ('ds', 'ignore_border', 'st', 'padding',
'mode')
if node.op.padding != (0, 0) or node.op.mode != 'max':
return
x, z, gz = node.inputs
if (x.owner and isinstance(x.owner.op, HostFromGpu)):
......@@ -1678,6 +1680,8 @@ def local_gpu_downsample_factor_max_grad(node):
@local_optimizer([downsample.DownsampleFactorMaxGradGrad])
def local_gpu_downsample_factor_max_grad_grad(node):
if isinstance(node.op, downsample.DownsampleFactorMaxGradGrad):
assert node.op.__props__ == ('ds', 'ignore_border', 'st')
x, z, gx = node.inputs
if (x.owner and isinstance(x.owner.op, HostFromGpu)):
op = GpuDownsampleFactorMaxGradGrad(node.op.ds,
......
theano/sandbox/cuda/tests/test_dnn.py
浏览文件 @ 9dc07802
......@@ -183,8 +183,12 @@ def test_pooling():
raise SkipTest(cuda.dnn.dnn_available.msg)
x = T.ftensor4()
for func, pad in product((T.max, T.mean),
for mode, pad in product(('max', 'average_inc_pad', 'average_exc_pad'),
((0, 0), (1, 0), (1, 0), (2, 3), (3, 2))):
if mode == 'max':
func = T.max
else:
func = T.mean
if pad != (0, 0) and cuda.dnn.version() == -1:
continue
......@@ -195,29 +199,23 @@ def test_pooling():
for stride in (2, 3):
if stride > ws:
continue
if func is T.max:
if pad[0] > stride or pad[1] > stride:
# Not implemented
continue
# We will check that the opt introduced it.
out1 = max_pool_2d(x, (ws, ws),
st=(stride, stride),
ignore_border=True,
padding=pad)
else:
out1 = cuda.dnn.dnn_pool(
x, ws=(ws, ws),
stride=(stride, stride),
pad=pad,
mode='max' if func is T.max else "average")
if pad[0] > stride or pad[1] > stride:
# Not implemented
continue
# We will check that the opt introduced it.
out1 = max_pool_2d(x, (ws, ws),
st=(stride, stride),
ignore_border=True,
padding=pad, mode=mode)
out2 = pool_2d_i2n(x, ds=(ws, ws), strides=(stride, stride),
pad=pad,
pool_function=func)
mode_without_gpu2 = mode_without_gpu.including()
mode_without_gpu2.check_isfinite = False
f1 = theano.function([x], out1, mode=mode_with_gpu)
assert any([isinstance(node.op, cuda.dnn.GpuDnnPool)
for node in f1.maker.fgraph.apply_nodes])
f2 = theano.function([x], out2, mode=mode_without_gpu)
f2 = theano.function([x], out2, mode=mode_without_gpu2)
assert not any([isinstance(node.op, cuda.dnn.GpuDnnPool)
for node in f2.maker.fgraph.apply_nodes])
for shp in [(1, 10, 100, 100),
......@@ -245,7 +243,7 @@ def test_pooling():
# This test the CPU grad + opt + GPU implemtentation
def fn(x):
return max_pool_2d(x, (ws, ws), ignore_border=True,
padding=pad)
padding=pad, mode=mode)
theano.tests.unittest_tools.verify_grad(fn, [data],
cast_to_output_type=False,
mode=mode_with_gpu)
......@@ -261,7 +259,7 @@ def test_pooling():
x, ws=(ws, ws),
stride=(stride, stride),
pad=pad,
mode='max' if func is T.max else "average")
mode=mode)
return dnn_op
theano.tests.unittest_tools.verify_grad(
fn, [data],
......@@ -274,17 +272,16 @@ def test_pooling():
for node in fg.maker.fgraph.toposort()])
g_out = fg(data)
if func is T.max:
# Compare again the CPU result
out = max_pool_2d(x, (ws, ws),
padding=pad,
ignore_border=True)
fc = theano.function([x], theano.grad(out.sum(), x),
mode=mode_without_gpu)
assert any([isinstance(node.op, DownsampleFactorMaxGrad)
for node in fc.maker.fgraph.toposort()])
c_out = fc(data)
assert numpy.allclose(c_out, g_out)
# Compare again the CPU result
out = max_pool_2d(x, (ws, ws),
padding=pad,
ignore_border=True, mode=mode)
fc = theano.function([x], theano.grad(out.sum(), x),
mode=mode_without_gpu)
assert any([isinstance(node.op, DownsampleFactorMaxGrad)
for node in fc.maker.fgraph.toposort()])
c_out = fc(data)
assert numpy.allclose(c_out, g_out)
def test_pooling_opt():
......@@ -523,7 +520,7 @@ class TestDnnInferShapes(utt.InferShapeTester):
for params in product(
[(1, 1), (2, 2), (3, 3)],
[(1, 1), (2, 2), (3, 3)],
['max', 'average']
['max', 'average_inc_pad', 'average_exc_pad']
):
desc = dnn.GpuDnnPoolDesc(
ws=params[0],
......@@ -559,7 +556,7 @@ class TestDnnInferShapes(utt.InferShapeTester):
for params in product(
[(1, 1), (2, 2), (3, 3)],
[(1, 1), (2, 2), (3, 3)],
['max', 'average']
['max', 'average_inc_pad']
):
desc = dnn.GpuDnnPoolDesc(
ws=params[0],
......
theano/tensor/signal/downsample.py
浏览文件 @ 9dc07802
......@@ -38,7 +38,8 @@ def max_pool_2d_same_size(input, patch_size):
return outs
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),
mode='max'):
"""
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
......@@ -62,11 +63,17 @@ def max_pool_2d(input, ds, ignore_border=False, st=None, padding=(0, 0)):
of the images, pad_h is the size of the top and bottom margins,
and pad_w is the size of the left and right margins.
:type padding: tuple of two ints
:param mode: 'max', 'average_inc_pad' or 'average_exc_pad'.
Operation executed on each window. `max` always excludes the padding
in the computation. `average` gives you the choice to include or
exclude it.
:type mode: string
"""
if input.ndim < 2:
raise NotImplementedError('max_pool_2d requires a dimension >= 2')
if input.ndim == 4:
op = DownsampleFactorMax(ds, ignore_border, st=st, padding=padding)
op = DownsampleFactorMax(ds, ignore_border, st=st, padding=padding,
mode=mode)
output = op(input)
return output
......@@ -84,7 +91,8 @@ def max_pool_2d(input, ds, ignore_border=False, st=None, padding=(0, 0)):
input_4D = tensor.reshape(input, new_shape, ndim=4)
# downsample mini-batch of images
op = DownsampleFactorMax(ds, ignore_border, st=st, padding=padding)
op = DownsampleFactorMax(ds, ignore_border, st=st, padding=padding,
mode=mode)
output = op(input_4D)
# restore to original shape
......@@ -94,12 +102,11 @@ def max_pool_2d(input, ds, ignore_border=False, st=None, padding=(0, 0)):
class DownsampleFactorMax(Op):
"""For N-dimensional tensors, consider that the last two
dimensions span images. This Op downsamples these images by a
factor ds, by taking the max over non- overlapping rectangular
regions.
dimensions span images. This Op downsamples these images by
taking the max or average over different patch.
"""
__props__ = ('ds', 'ignore_border', 'st', 'padding')
__props__ = ('ds', 'ignore_border', 'st', 'padding', 'mode')
@staticmethod
def out_shape(imgshape, ds, ignore_border=False, st=None, padding=(0, 0)):
......@@ -178,8 +185,10 @@ 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),
mode='max'):
""" Take the max or average or different input patches.
:param ds: downsample factor over rows and column.
ds indicates the pool region size.
:type ds: list or tuple of two ints
......@@ -193,13 +202,17 @@ class DownsampleFactorMax(Op):
over rows/cols to get the the next pool region.
if st is None, it is considered equal to ds
(no overlap on pooling regions)
: type st: list or tuple of two ints
: type st: list or tuple of two ints or None
:param padding: (pad_h, pad_w), pad zeros to extend beyond four borders
of the images, pad_h is the size of the top and bottom margins,
and pad_w is the size of the left and right margins.
:type padding: tuple of two ints
:param mode: 'max', 'average_inc_pad', 'average_exc_pad'.
('average_inc_pad' exclude the padding from the count,
'average_exc_pad' include it)
"""
self.ds = tuple(ds)
if not all([isinstance(d, int) for d in ds]):
......@@ -208,6 +221,7 @@ class DownsampleFactorMax(Op):
" Got %s" % str(ds))
if st is None:
st = ds
assert isinstance(st, (tuple, list))
self.st = tuple(st)
self.ignore_border = ignore_border
self.padding = tuple(padding)
......@@ -217,11 +231,11 @@ class DownsampleFactorMax(Op):
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')
def __str__(self):
return '%s{%s, %s, %s, %s}' % (
self.__class__.__name__,
self.ds, self.st, self.ignore_border, self.padding)
if mode not in ['max', 'average_inc_pad', 'average_exc_pad']:
raise ValueError(
"DownsampleFactorMax mode parameter only support 'max',"
" 'average_inc_pad' and 'average_exc_pad'. Got %s" % mode)
self.mode = mode
def make_node(self, x):
if x.type.ndim != 4:
......@@ -251,27 +265,37 @@ class DownsampleFactorMax(Op):
pad_w = self.padding[1]
img_rows = x.shape[-2] + 2 * pad_h
img_cols = x.shape[-1] + 2 * pad_w
inc_pad = self.mode == 'average_inc_pad'
# pad the image
if self.padding != (0, 0):
fill = x.min()-1.
y = numpy.zeros(
(x.shape[0], x.shape[1], img_rows, img_cols),
dtype=x.dtype) + fill
dtype=x.dtype)
y[:, :, pad_h:(img_rows-pad_h), pad_w:(img_cols-pad_w)] = x
else:
y = x
# max pooling
func = numpy.max
if self.mode != 'max':
func = numpy.average
for n in xrange(x.shape[0]):
for k in xrange(x.shape[1]):
for r in xrange(pr):
row_st = r * st0
row_end = __builtin__.min(row_st + ds0, img_rows)
if not inc_pad:
row_st = __builtin__.max(row_st, self.padding[0])
row_end = __builtin__.min(row_end, x.shape[-2] + pad_h)
for c in xrange(pc):
col_st = c * st1
col_end = __builtin__.min(col_st + ds1, img_cols)
zz[n, k, r, c] = y[
n, k, row_st:row_end, col_st:col_end].max()
if not inc_pad:
col_st = __builtin__.max(col_st, self.padding[1])
col_end = __builtin__.min(col_end,
x.shape[-1] + pad_w)
zz[n, k, r, c] = func(y[
n, k, row_st:row_end, col_st:col_end])
def infer_shape(self, node, in_shapes):
shp = self.out_shape(in_shapes[0], self.ds,
......@@ -284,13 +308,16 @@ class DownsampleFactorMax(Op):
maxout = self(x)
return [DownsampleFactorMaxGrad(self.ds,
ignore_border=self.ignore_border,
st=self.st, padding=self.padding)(
st=self.st, padding=self.padding,
mode=self.mode)(
x, maxout, gz)]
def c_headers(self):
return ['<algorithm>']
def c_code(self, node, name, inp, out, sub):
if self.mode != 'max':
raise theano.gof.utils.MethodNotDefined()
x, = inp
z, = out
fail = sub['fail']
......@@ -441,20 +468,20 @@ class DownsampleFactorMax(Op):
class DownsampleFactorMaxGrad(Op):
__props__ = ('ds', 'ignore_border', 'st', 'padding')
__props__ = ('ds', 'ignore_border', 'st', 'padding', 'mode')
def __init__(self, ds, ignore_border, st=None, padding=(0, 0)):
def __init__(self, ds, ignore_border, st=None, padding=(0, 0), mode='max'):
self.ds = tuple(ds)
self.ignore_border = ignore_border
if st is None:
st = ds
self.st = tuple(st)
self.padding = tuple(padding)
def __str__(self):
return '%s{%s, %s, %s, %s}' % (
self.__class__.__name__,
self.ds, self.st, self.ignore_border, self.padding)
if mode not in ['max', 'average_inc_pad', 'average_exc_pad']:
raise ValueError(
"DownsampleFactorMax mode parameter only support 'max',"
" 'average_inc_pad' and 'average_exc_pad'. Got %s" % mode)
self.mode = mode
def make_node(self, x, maxout, gz):
# make_node should only be called by the grad function of
......@@ -469,6 +496,8 @@ class DownsampleFactorMaxGrad(Op):
return Apply(self, [x, maxout, gz], [x.type()])
def perform(self, node, inp, out):
if self.mode != 'max' and self.padding != (0, 0):
raise NotImplementedError()
x, maxout, gz = inp
gx_stg, = out
# number of pooling output rows
......@@ -481,28 +510,49 @@ class DownsampleFactorMaxGrad(Op):
pad_w = self.padding[1]
img_rows = x.shape[-2] + 2 * pad_h
img_cols = x.shape[-1] + 2 * pad_w
inc_pad = self.mode == 'average_inc_pad'
# pad the image
if self.padding != (0, 0):
fill = x.min()-1
y = numpy.zeros(
(x.shape[0], x.shape[1], img_rows, img_cols),
dtype=x.dtype) + fill
dtype=x.dtype)
y[:, :, pad_h:(img_rows-pad_h), pad_w:(img_cols-pad_w)] = x
else:
y = x
gx = numpy.zeros_like(y)
for n in xrange(x.shape[0]):
for k in xrange(x.shape[1]):
for r in xrange(pr):
row_st = r * st0
row_end = __builtin__.min(row_st + ds0, img_rows)
for c in xrange(pc):
col_st = c * st1
col_end = __builtin__.min(col_st + ds1, img_cols)
for row_ind in xrange(row_st, row_end):
for col_ind in xrange(col_st, col_end):
if (maxout[n, k, r, c] == y[n, k, row_ind, col_ind]):
gx[n, k, row_ind, col_ind] += gz[n, k, r, c]
if self.mode == 'max':
for n in xrange(x.shape[0]):
for k in xrange(x.shape[1]):
for r in xrange(pr):
row_st = __builtin__.max(r * st0, self.padding[0])
row_end = __builtin__.min(row_st + ds0, img_rows)
for c in xrange(pc):
col_st = __builtin__.max(c * st1, self.padding[1])
col_end = __builtin__.min(col_st + ds1, img_cols)
for row_ind in xrange(row_st, row_end):
for col_ind in xrange(col_st, col_end):
if (maxout[n, k, r, c] == y[n, k, row_ind, col_ind]):
gx[n, k, row_ind, col_ind] += gz[n, k, r, c]
else:
for n in xrange(x.shape[0]):
for k in xrange(x.shape[1]):
for r in xrange(pr):
if inc_pad:
row_st = r * st0
else:
row_st = __builtin__.max(r * st0, self.padding[0])
row_end = __builtin__.min(row_st + ds0, img_rows)
for c in xrange(pc):
if inc_pad:
col_st = c * st1
else:
col_st = __builtin__.max(c * st1,
self.padding[1])
col_end = __builtin__.min(col_st + ds1, img_cols)
val = gz[n, k, r, c] / ((row_end - row_st) *
(col_end - col_st))
gx[n, k, row_st:row_end, col_st:col_end] += val
# unpad the image
gx = gx[:, :, pad_h:(img_rows-pad_h), pad_w:(img_cols-pad_w)]
gx_stg[0] = gx
......@@ -513,7 +563,7 @@ class DownsampleFactorMaxGrad(Op):
def grad(self, inp, grads):
x, maxout, gz = inp
ggx, = grads
if self.padding == (0, 0):
if self.padding == (0, 0) and self.mode == 'max':
return [theano.tensor.zeros_like(x),
theano.tensor.zeros_like(maxout),
DownsampleFactorMaxGradGrad(
......@@ -528,6 +578,8 @@ class DownsampleFactorMaxGrad(Op):
def c_code(self, node, name, inp, out, sub):
if self.ds != self.st or self.padding != (0, 0):
raise theano.gof.utils.MethodNotDefined()
if self.mode != 'max':
raise theano.gof.utils.MethodNotDefined()
x, z, gz = inp
gx, = out
fail = sub['fail']
......@@ -624,6 +676,7 @@ class DownsampleFactorMaxGrad(Op):
class DownsampleFactorMaxGradGrad(Op):
__props__ = ('ds', 'ignore_border', 'st')
@staticmethod
def out_shape(imgshape, ds, ignore_border=False, st=None):
......@@ -702,20 +755,6 @@ class DownsampleFactorMaxGradGrad(Op):
st = ds
self.st = tuple(st)
def __eq__(self, other):
return (type(self) == type(other)
and self.ds == other.ds
and self.st == other.st
and self.ignore_border == other.ignore_border)
def __hash__(self):
return hash(type(self)) ^ hash(self.ds) ^ \
hash(self.st) ^ hash(self.ignore_border)
def __str__(self):
return '%s{%s,%s,%s}' % (self.__class__.__name__,
self.ds, self.st, self.ignore_border)
def make_node(self, x, maxout, gz):
# make_node should only be called by the grad function of
# DownsampleFactorMaxGrad, so these asserts should not fail.
......
theano/tensor/signal/tests/test_downsample.py
浏览文件 @ 9dc07802
from itertools import product
import unittest
import __builtin__
import numpy
import theano
import theano.tensor as tensor
from theano.tests import unittest_tools as utt
......@@ -12,7 +15,7 @@ from theano import function
class TestDownsampleFactorMax(utt.InferShapeTester):
@staticmethod
def numpy_max_pool_2d(input, ds, ignore_border=False):
def numpy_max_pool_2d(input, ds, ignore_border=False, mode='max'):
'''Helper function, implementing max_pool_2d in pure numpy'''
if len(input.shape) < 2:
raise NotImplementedError('input should have at least 2 dim,'
......@@ -29,6 +32,9 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
out_shp.append(input.shape[-2] / ds[0] + xi)
out_shp.append(input.shape[-1] / ds[1] + yi)
output_val = numpy.zeros(out_shp)
func = numpy.max
if mode != 'max':
func = numpy.average
for k in numpy.ndindex(*input.shape[:-2]):
for i in range(output_val.shape[-2]):
......@@ -36,12 +42,12 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
for j in range(output_val.shape[-1]):
jj = j * ds[1]
patch = input[k][ii:ii + ds[0], jj:jj + ds[1]]
output_val[k][i, j] = numpy.max(patch)
output_val[k][i, j] = func(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), mode='max'):
pad_h = padding[0]
pad_w = padding[1]
h = x.shape[-2]
......@@ -50,14 +56,12 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
assert ds[1] > pad_w
def pad_img(x):
fill = x.min()-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
lr_bar = (numpy.zeros((pad_h * 2 + h, pad_w)) + fill)[
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)
y = numpy.zeros(
(x.shape[0], x.shape[1],
x.shape[2]+pad_h*2, x.shape[3]+pad_w*2),
dtype=x.dtype)
y[:, :, pad_h:(x.shape[2]+pad_h), pad_w:(x.shape[3]+pad_w)] = x
return y
img_rows = h + 2 * pad_h
img_cols = w + 2 * pad_w
......@@ -71,19 +75,31 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
output_val = numpy.zeros(out_shp)
tt = []
y = pad_img(x)
func = numpy.max
if mode != 'max':
func = numpy.average
inc_pad = mode == 'average_inc_pad'
for k in numpy.ndindex(*x.shape[:-2]):
for i in range(output_val.shape[-2]):
ii_st = i * st[0]
ii_end = __builtin__.min(ii_st + ds[0], img_rows)
if not inc_pad:
ii_st = __builtin__.max(ii_st, pad_h)
ii_end = __builtin__.min(ii_end, h + pad_h)
for j in range(output_val.shape[-1]):
jj_st = j * st[1]
jj_end = __builtin__.min(jj_st + ds[1], img_cols)
if not inc_pad:
jj_st = __builtin__.max(jj_st, pad_w)
jj_end = __builtin__.min(jj_end, w + pad_w)
patch = y[k][ii_st:ii_end, jj_st:jj_end]
output_val[k][i, j] = numpy.max(patch)
output_val[k][i, j] = func(patch)
return output_val
@staticmethod
def numpy_max_pool_2d_stride(input, ds, ignore_border=False, st=None):
def numpy_max_pool_2d_stride(input, ds, ignore_border=False, st=None,
mode='max'):
'''Helper function, implementing max_pool_2d in pure numpy
this function provides st input to indicate the stide size
for the pooling regions. if not indicated, st == sd.'''
......@@ -128,6 +144,10 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
out_shp.append(out_r)
out_shp.append(out_c)
func = numpy.max
if mode != 'max':
func = numpy.average
output_val = numpy.zeros(out_shp)
for k in numpy.ndindex(*input.shape[:-2]):
for i in range(output_val.shape[-2]):
......@@ -137,32 +157,37 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
jj_st = j * st[1]
jj_end = __builtin__.min(jj_st + ds[1], img_cols)
patch = input[k][ii_st:ii_end, jj_st:jj_end]
output_val[k][i, j] = numpy.max(patch)
output_val[k][i, j] = func(patch)
return output_val
def test_DownsampleFactorMax(self):
rng = numpy.random.RandomState(utt.fetch_seed())
# generate random images
maxpoolshps = ((1, 1), (2, 2), (3, 3), (2, 3))
imval = rng.rand(4, 10, 64, 64)
imval = rng.rand(4, 2, 16, 16)
images = tensor.dtensor4()
for maxpoolshp in maxpoolshps:
for ignore_border in [True, False]:
for maxpoolshp, ignore_border, mode in product(maxpoolshps,
[True, False],
['max',
'average_inc_pad',
'average_exc_pad']):
# print 'maxpoolshp =', maxpoolshp
# print 'ignore_border =', ignore_border
# Pure Numpy computation
numpy_output_val = self.numpy_max_pool_2d(imval, maxpoolshp,
ignore_border)
output = max_pool_2d(images, maxpoolshp, ignore_border)
ignore_border,
mode=mode)
output = max_pool_2d(images, maxpoolshp, ignore_border,
mode=mode)
f = function([images, ], [output, ])
output_val = f(imval)
assert numpy.all(output_val == numpy_output_val)
# DownsampleFactorMax op
maxpool_op = DownsampleFactorMax(maxpoolshp,
ignore_border=ignore_border)(images)
ignore_border=ignore_border,
mode=mode)(images)
f = function([images], maxpool_op)
output_val = f(imval)
utt.assert_allclose(output_val, numpy_output_val)
......@@ -179,24 +204,30 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
(4, 10, 14, 14), (4, 10, 6, 6), (4, 10, 4, 3),
(4, 10, 12, 14), (4, 10, 4, 5), (4, 10, 3, 2),
(4, 10, 12, 14), (4, 10, 5, 6), (4, 10, 4, 3))
# The same for each mode
outputshps = outputshps + outputshps + outputshps
images = tensor.dtensor4()
indx = 0
for maxpoolshp in maxpoolshps:
for ignore_border in [True, False]:
for mode, maxpoolshp, ignore_border in product(['max',
'average_inc_pad',
'average_exc_pad'],
maxpoolshps,
[True, False]):
for stride in stridesizes:
outputshp = outputshps[indx]
indx += 1
# DownsampleFactorMax op
numpy_output_val = \
self.numpy_max_pool_2d_stride(imval, maxpoolshp,
ignore_border, stride)
ignore_border, stride,
mode)
assert numpy_output_val.shape == outputshp, (
"outshape is %s, calculated shape is %s"
% (outputshp, numpy_output_val.shape))
maxpool_op = \
DownsampleFactorMax(maxpoolshp,
ignore_border=ignore_border,
st=stride)(images)
st=stride, mode=mode)(images)
f = function([images], maxpool_op)
output_val = f(imval)
utt.assert_allclose(output_val, numpy_output_val)
......@@ -219,7 +250,9 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
imval = rng.rand(4, 10, imvsize[0], imvsize[1])
stride = stridesizes[indx]
maxpoolshp = maxpoolshps[indx]
for ignore_border in [True, False]:
for ignore_border, mode in product([True, False],
['max', 'average_inc_pad',
'average_exc_pad']):
indx_out = indx * 2
if not ignore_border:
indx_out += 1
......@@ -227,14 +260,14 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
# DownsampleFactorMax op
numpy_output_val = \
self.numpy_max_pool_2d_stride(imval, maxpoolshp,
ignore_border, stride)
ignore_border, stride, mode)
assert numpy_output_val.shape == outputshp, (
"outshape is %s, calculated shape is %s"
% (outputshp, numpy_output_val.shape))
maxpool_op = \
DownsampleFactorMax(maxpoolshp,
ignore_border=ignore_border,
st=stride)(images)
st=stride, mode=mode)(images)
f = function([images], maxpool_op)
output_val = f(imval)
utt.assert_allclose(output_val, numpy_output_val)
......@@ -247,20 +280,24 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
paddingsizes = [(2, 2), (1, 2), (2, 1), (0, 0), (1, 1)]
imgsizes = [(5, 5), (5, 5), (5, 6), (6, 5), (5, 5)]
m = 4 # minibatch
c = 10 # channel size
c = 2 # channel size
images = tensor.dtensor4()
for indx in numpy.arange(len(maxpoolsizes)):
for indx, mode in product(numpy.arange(len(maxpoolsizes)),
['max', 'average_inc_pad',
'average_exc_pad']):
imgsize = imgsizes[indx]
imval = rng.rand(m, c, imgsize[0], imgsize[1])
imval = rng.rand(m, c, imgsize[0], imgsize[1]) - 0.5
stridesize = stridesizes[indx]
maxpoolsize = maxpoolsizes[indx]
paddingsize = paddingsizes[indx]
numpy_output_val = self.numpy_max_pool_2d_stride_padding(
imval, maxpoolsize, ignore_border, stridesize, paddingsize)
imval, maxpoolsize, ignore_border,
stridesize, paddingsize, mode)
maxpool_op = DownsampleFactorMax(
maxpoolsize,
ignore_border=ignore_border,
st=stridesize, padding=paddingsize)(images)
st=stridesize, padding=paddingsize, mode=mode)(images)
f = function([images], maxpool_op)
output_val = f(imval)
utt.assert_allclose(output_val, numpy_output_val)
......@@ -277,7 +314,7 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
maxpoolsize = maxpoolsizes[i]
stridesize = stridesizes[i]
paddingsize = paddingsizes[i]
def mp(input):
return DownsampleFactorMax(
maxpoolsize, ignore_border=True,
......@@ -447,20 +484,26 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
imval = rng.rand(4, 5)
images = tensor.dmatrix()
for maxpoolshp in maxpoolshps:
for ignore_border in [True, False]:
for maxpoolshp, ignore_border, mode in product(maxpoolshps,
[True, False],
['max',
'average_inc_pad',
'average_exc_pad']):
# print 'maxpoolshp =', maxpoolshp
# print 'ignore_border =', ignore_border
numpy_output_val = self.numpy_max_pool_2d(imval, maxpoolshp,
ignore_border)
output = max_pool_2d(images, maxpoolshp, ignore_border)
ignore_border,
mode=mode)
output = max_pool_2d(images, maxpoolshp, ignore_border,
mode=mode)
output_val = function([images], output)(imval)
assert numpy.all(output_val == numpy_output_val), (
"output_val is %s, numpy_output_val is %s"
% (output_val, numpy_output_val))
def mp(input):
return max_pool_2d(input, maxpoolshp, ignore_border)
return max_pool_2d(input, maxpoolshp, ignore_border,
mode=mode)
utt.verify_grad(mp, [imval], rng=rng)
def test_max_pool_2d_2D_same_size(self):
......@@ -492,13 +535,18 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
imval = rng.rand(2, 3, 4)
images = tensor.dtensor3()
for maxpoolshp in maxpoolshps:
for ignore_border in [True, False]:
for maxpoolshp, ignore_border, mode in product(maxpoolshps,
[True, False],
['max',
'average_inc_pad',
'average_exc_pad']):
# print 'maxpoolshp =', maxpoolshp
# print 'ignore_border =', ignore_border
numpy_output_val = self.numpy_max_pool_2d(imval, maxpoolshp,
ignore_border)
output = max_pool_2d(images, maxpoolshp, ignore_border)
ignore_border,
mode)
output = max_pool_2d(images, maxpoolshp, ignore_border,
mode=mode)
output_val = function([images], output)(imval)
assert numpy.all(output_val == numpy_output_val), (
"output_val is %s, numpy_output_val is %s"
......@@ -524,13 +572,18 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
imval = rng.rand(2, 1, 1, 1, 3, 4)
images = tensor.TensorType('float64', [False] * 6)()
for maxpoolshp in maxpoolshps:
for ignore_border in [True, False]:
for maxpoolshp, ignore_border, mode in product(maxpoolshps,
[True, False],
['max',
'average_inc_pad',
'average_exc_pad']):
# print 'maxpoolshp =', maxpoolshp
# print 'ignore_border =', ignore_border
numpy_output_val = self.numpy_max_pool_2d(imval, maxpoolshp,
ignore_border)
output = max_pool_2d(images, maxpoolshp, ignore_border)
ignore_border,
mode=mode)
output = max_pool_2d(images, maxpoolshp, ignore_border,
mode=mode)
output_val = function([images], output)(imval)
assert numpy.all(output_val == numpy_output_val)
......
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