多数的前向推理框架不支持AdaptivePooing操作，此时需要将AdaptivePooing操作转换为普通的Pooling操作。AdaptivePooling与Max/AvgPooling相互转换提供了一种转换方法，但我在Pytorch1.6中的测试结果是错误的。通过查看Pytorch源码（pytorch-master\aten\src\ATen\native\AdaptiveAveragePooling.cpp）我找出了正确的转换方式。
inline int start_index(int a, int b, int c) { return (int)std::floor((float)(a * c) / b); } inline int end_index(int a, int b, int c) { return (int)std::ceil((float)((a + 1) * c) / b); } template <typename scalar_t> static void adaptive_avg_pool2d_single_out_frame( scalar_t *input_p, scalar_t *output_p, int64_t sizeD, int64_t isizeH, int64_t isizeW, int64_t osizeH, int64_t osizeW, int64_t istrideD, int64_t istrideH, int64_t istrideW) { at::parallel_for(0, sizeD, 0, [&](int64_t start, int64_t end) { for (auto d = start; d < end; d++) { /* loop over output */ int64_t oh, ow; for(oh = 0; oh < osizeH; oh++) { int istartH = start_index(oh, osizeH, isizeH); int iendH = end_index(oh, osizeH, isizeH); int kH = iendH - istartH; for(ow = 0; ow < osizeW; ow++) { int istartW = start_index(ow, osizeW, isizeW); int iendW = end_index(ow, osizeW, isizeW); int kW = iendW - istartW; /* local pointers */ scalar_t *ip = input_p + d*istrideD + istartH*istrideH + istartW*istrideW; scalar_t *op = output_p + d*osizeH*osizeW + oh*osizeW + ow; /* compute local average: */ scalar_t sum = 0; int ih, iw; for(ih = 0; ih < kH; ih++) { for(iw = 0; iw < kW; iw++) { scalar_t val = *(ip + ih*istrideH + iw*istrideW); sum += val; } } /* set output to local average */ *op = sum / kW / kH; } } } }); }上述代码段中isizeH，isizeW分别表示输入张量的宽高osizeH，osizeW则表示输出宽高。关注第二个for循环for(oh = 0; oh < osizeH; oh++){.....}中的内容。假设输入的宽高均为223isizeH = isizeW = 223，输出的宽高均为7osizeH = osizeW = 224,然后简单分析一下oh=0,1,2时的情况：
oh=0, istartH = 0, iendH = ceil(223/7)=32, kH = 32oh=1, istartH = floor(223/7) = 31, iendH = ceil(223*2/7)=64, kH = 33oh=2, istartH = floor(223*2/7) = 63, iendH = ceil(223*3/7)=96, kH = 33这里的kH就是kernel_size的大小. oh=0时的kernel_size比其他情况要小，所以需要在输入上添加padding，让oh=0时的kernel_size与其他情况相同。添加的padding大小为1，等价于让istartH从-1开始，即kH = 32-(-1) = 33. 下一个需要获取的参数是stride，stride = istartH[oh=i]-istartH[oh=i-1], 在上述例子中即为32。按照上述的例子分析输入宽高为224的情况可以发现padding=0，所以padding也是一个需要转换的参数。下面给出3个参数的转换公式：
stride = ceil(input_size / output_size)kernel_size = ceil(2 * input_size / output_size) - floor(input_size / output_size)padding = ceil(input_size / output_size) - floor(input_size / output_size)在上述的代码中最后部分，可以看见均值使用*op = sum / kW / kH计算得到的。这表明在边缘部分计算均值没有考虑padding，所以对应的AvgPool中的count_include_pad应该设为False。下面贴出我的测试代码：
def test(size): import numpy as np import torch x = torch.randn(1,1,size,size) input_size = np.array(x.shape[2:]) output_size = np.array([7,7]) # stride = ceil(input_size / output_size) # kernel_size = ceil(2 * input_size / output_size) - floor(input_size / output_size) # padding = ceil(input_size / output_size) - floor(input_size / output_size) stride = numpy.ceil(input_size / output_size).astype(int) kernel_size = (numpy.ceil(2 * input_size / output_size) - numpy.floor(input_size / output_size)).astype(int) padding = (numpy.ceil(input_size / output_size) - numpy.floor(input_size / output_size)).astype(int) print(stride) print(kernel_size) print(padding) avg1 = nn.AdaptiveAvgPool2d(list(output_size)) avg2 = nn.AvgPool2d(kernel_size=kernel_size.tolist(), stride=stride.tolist(), padding=padding.tolist(), ceil_mode=False, count_include_pad=False) max1 = nn.AdaptiveMaxPool2d(list(output_size)) max2 = nn.MaxPool2d(kernel_size=kernel_size.tolist(), stride=stride.tolist(), padding=padding.tolist(), ceil_mode=False ) avg1_out = avg1(x) avg2_out = avg2(x) max1_out = max1(x) max2_out = max2(x) print(avg1_out-avg2_out) print(max1_out-max2_out) print(torch.__version__)
