finish average pool python code on the CPU.

theano/sandbox/cuda/dnn.py

@@ -720,7 +720,8 @@ 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.
@@ -743,7 +744,9 @@ 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
@@ -771,8 +774,12 @@ 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.")
 
@@ -1193,7 +1200,8 @@ 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.

theano/sandbox/cuda/tests/test_dnn.py

@@ -88,12 +88,12 @@ 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 func is T.max:
-            mode = 'max'
+        if mode == 'max':
+            func = T.max
         else:
-            mode = 'average'
+            func = T.mean
         if pad != (0, 0) and cuda.dnn.version() == -1:
             continue
 
@@ -164,7 +164,7 @@ 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],
@@ -427,7 +427,7 @@ 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],
@@ -463,7 +463,7 @@ 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

@@ -63,7 +63,9 @@ 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' or 'average', the operation executed on each window.
+    :param mode: 'max', 'average_inc_pad' or 'average_exc_pad',
+        the operation executed on each window. The max always exclude the
+        padding in the computation of the max. For average, you have the choise.
     :type mode: string
     """
     if input.ndim < 2:
@@ -185,8 +187,7 @@ class DownsampleFactorMax(Op):
 
     def __init__(self, ds, ignore_border=False, st=None, padding=(0, 0),
                  mode='max'):
-        """
-        :param ds: downsample factor over rows and column.
+        """:param ds: downsample factor over rows and column.
                    ds indicates the pool region size.
         :type ds: list or tuple of two ints
 
@@ -206,7 +207,9 @@ class DownsampleFactorMax(Op):
             and pad_w is the size of the left and right margins.
         :type padding: tuple of two ints
 
-        :param mode: 'max' or 'average'
+        :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)
@@ -226,10 +229,10 @@ 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')
-        if mode not in ['max', 'average']:
+        if mode not in ['max', 'average_inc_pad', 'average_exc_pad']:
             raise ValueError(
-                "DownsampleFactorMax mode parameter only support 'max' and"
-                " 'average'. Got %s" % mode)
+                "DownsampleFactorMax mode parameter only support 'max',"
+                " 'average_inc_pad' and 'average_exc_pad'. Got %s" % mode)
         self.mode = mode
 
     def __str__(self):
@@ -245,8 +248,6 @@ class DownsampleFactorMax(Op):
         return gof.Apply(self, [x], [x.type()])
 
     def perform(self, node, inp, out):
-        if self.mode != 'max' and self.padding != (0, 0):
-            raise NotImplementedError()
         x, = inp
         z, = out
         if len(x.shape) != 4:
@@ -267,18 +268,18 @@ 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
         func = numpy.max
-        if self.mode == 'average':
+        if self.mode != 'max':
             func = numpy.average
         # max pooling
         for n in xrange(x.shape[0]):
@@ -286,9 +287,16 @@ class DownsampleFactorMax(Op):
                 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)
+                        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])
 
@@ -472,10 +480,10 @@ class DownsampleFactorMaxGrad(Op):
             st = ds
         self.st = tuple(st)
         self.padding = tuple(padding)
-        if mode not in ['max', 'average']:
+        if mode not in ['max', 'average_inc_pad', 'average_exc_pad']:
             raise ValueError(
-                "DownsampleFactorMax mode parameter only support 'max' and"
-                " 'average'. Got %s" % mode)
+                "DownsampleFactorMax mode parameter only support 'max',"
+                " 'average_inc_pad' and 'average_exc_pad'. Got %s" % mode)
         self.mode = mode
 
     def __str__(self):
@@ -510,12 +518,13 @@ 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
@@ -524,29 +533,34 @@ class DownsampleFactorMaxGrad(Op):
             for n in xrange(x.shape[0]):
                 for k in xrange(x.shape[1]):
                     for r in xrange(pr):
-                        row_st = r * st0
+                        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 = c * st1
+                            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]
-        elif self.mode == 'average':
+        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
-        else:
-            raise ValueError('mode %s not know' % self.mode)
         # unpad the image
         gx = gx[:, :, pad_h:(img_rows-pad_h), pad_w:(img_cols-pad_w)]
         gx_stg[0] = gx

theano/tensor/signal/tests/test_downsample.py

@@ -33,7 +33,7 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
         out_shp.append(input.shape[-1] / ds[1] + yi)
         output_val = numpy.zeros(out_shp)
         func = numpy.max
-        if mode == 'average':
+        if mode != 'max':
             func = numpy.average
 
         for k in numpy.ndindex(*input.shape[:-2]):
@@ -47,7 +47,7 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
 
     @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]
@@ -56,14 +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
@@ -77,15 +75,26 @@ 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
@@ -136,7 +145,7 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
         out_shp.append(out_c)
 
         func = numpy.max
-        if mode == 'average':
+        if mode != 'max':
             func = numpy.average
 
         output_val = numpy.zeros(out_shp)
@@ -159,7 +168,9 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
         images = tensor.dtensor4()
         for maxpoolshp, ignore_border, mode in product(maxpoolshps,
                                                        [True, False],
-                                                       ['max', 'average']):
+                                                       ['max',
+                                                        'average_inc_pad',
+                                                        'average_exc_pad']):
                 # print 'maxpoolshp =', maxpoolshp
                 # print 'ignore_border =', ignore_border
 
@@ -193,15 +204,13 @@ 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))
-        outputshps += ((4, 10, 16, 16), (4, 10, 6, 6), (4, 10, 4, 3),
-                      (4, 10, 16, 16), (4, 10, 6, 6), (4, 10, 4, 3),
-                      (4, 10, 14, 14), (4, 10, 5, 5), (4, 10, 3, 2),
-                      (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 mode, maxpoolshp, ignore_border in product(['max', 'average'],
+        for mode, maxpoolshp, ignore_border in product(['max',
+                                                        'average_inc_pad',
+                                                        'average_exc_pad'],
                                                        maxpoolshps,
                                                        [True, False]):
                 for stride in stridesizes:
@@ -242,7 +251,8 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
             stride = stridesizes[indx]
             maxpoolshp = maxpoolshps[indx]
             for ignore_border, mode in product([True, False],
-                                               ['max', 'average']):
+                                               ['max', 'average_inc_pad',
+                                                'average_exc_pad']):
                 indx_out = indx * 2
                 if not ignore_border:
                     indx_out += 1
@@ -270,20 +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)
@@ -472,7 +486,9 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
 
         for maxpoolshp, ignore_border, mode in product(maxpoolshps,
                                                        [True, False],
-                                                       ['max', 'average']):
+                                                       ['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,
@@ -521,7 +537,9 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
 
         for maxpoolshp, ignore_border, mode in product(maxpoolshps,
                                                        [True, False],
-                                                       ['max', 'average']):
+                                                       ['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,
@@ -556,7 +574,9 @@ class TestDownsampleFactorMax(utt.InferShapeTester):
 
         for maxpoolshp, ignore_border, mode in product(maxpoolshps,
                                                        [True, False],
-                                                       ['max', 'average']):
+                                                       ['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,