tf.nn.conv2d(
input,
filter,
strides,
padding,
use_cudnn_on_gpu=True,
data_format='NHWC',
dilations=[1, 1, 1, 1],
name=None
)
Defined in tensorflow/python/ops/gen_nn_ops.py
.
See the guide: Neural Network > Convolution
Computes a 2-D convolution given 4-D input
and filter
tensors.
Given an input tensor of shape [batch, in_height, in_width, in_channels]
and a filter / kernel tensor of shape[filter_height, filter_width, in_channels, out_channels]
, this op performs the following:
- Flattens the filter to a 2-D matrix with shape
[filter_height * filter_width * in_channels, output_channels]
. - Extracts image patches from the input tensor to form a virtual tensor of shape
[batch, out_height, out_width, filter_height * filter_width * in_channels]
. - For each patch, right-multiplies the filter matrix and the image patch vector.
In detail, with the default NHWC format,
output[b, i, j, k] =
sum_{di, dj, q} input[b, strides[1] * i + di, strides[2] * j + dj, q] *
filter[di, dj, q, k]
output[b, i, j, k] =
sum_{di, dj, q} input[b, strides[1] * i + di, strides[2] * j + dj, q] *
filter[di, dj, q, k]
output[b, i, j, k] =
sum_{di, dj, q} input[b, strides[1] * i + di, strides[2] * j + dj, q] *
filter[di, dj, q, k]
output[b, i, j, k] =
sum_{di, dj, q} input[b, strides[1] * i + di, strides[2] * j + dj, q] *
filter[di, dj, q, k]
Must have strides[0] = strides[3] = 1
. For the most common case of the same horizontal and vertices strides, strides = [1, stride, stride, 1]
.
Args:
input
: ATensor
. Must be one of the following types:half
,bfloat16
,float32
,float64
. A 4-D tensor. The dimension order is interpreted according to the value ofdata_format
, see below for details.filter
: ATensor
. Must have the same type asinput
. A 4-D tensor of shape[filter_height, filter_width, in_channels, out_channels]
strides
: A list ofints
. 1-D tensor of length 4. The stride of the sliding window for each dimension ofinput
. The dimension order is determined by the value ofdata_format
, see below for details.padding
: Astring
from:"SAME", "VALID"
. The type of padding algorithm to use.use_cudnn_on_gpu
: An optionalbool
. Defaults toTrue
.data_format
: An optionalstring
from:"NHWC", "NCHW"
. Defaults to"NHWC"
. Specify the data format of the input and output data. With the default format "NHWC", the data is stored in the order of: [batch, height, width, channels]. Alternatively, the format could be "NCHW", the data storage order of: [batch, channels, height, width].dilations
: An optional list ofints
. Defaults to[1, 1, 1, 1]
. 1-D tensor of length 4. The dilation factor for each dimension ofinput
. If set to k > 1, there will be k-1 skipped cells between each filter element on that dimension. The dimension order is determined by the value ofdata_format
, see above for details. Dilations in the batch and depth dimensions must be 1.name
: A name for the operation (optional).
Returns:
A Tensor
. Has the same type as input
.
函数定义
tf.nn.conv2d(input, filter, strides, padding, use_cudnn_on_gpu=None, data_format=None, name=None)
功能:在给定4-D 输入和fliters的情况下,计算二维卷积。
input的shape: [batch, in_height, in_width, in_channels]
filter的shape: [filter_height, filter_width, in_channels, out_channels]
计算过程如下:
(1)展平filter成如下2-D matrix,其shape: [filter_height * filter_width * in_channels, output_channels]
(2)从input tensor中提取patches构成一个virtual tensor, 其shape: [batch, out_height, out_width, filter_height * filter_width * in_channels]
(3)对于每一个patch, 右乘上(1)中的filter matrix。即[batch, out_height, out_width, filter_height * filter_width * in_channels] x [filter_height * filter_width * in_channels, output_channels],其结果的shape就是[batch, out_height, out_width, output_channels]。
【注:必须有 strides[0] = strides[3] = 1】。绝大多数情况下,水平的stride和竖直的stride一样,即strides = [1, stride, stride, 1]。
输出结果的shape计算:
‘SAME’ 类型的padding,其输出的height和width计算如下:
out_height = ceil(float(in_height) / float(strides[1])) ceil:向上取整
out_width = ceil(float(in_width) / float(strides[2]))
‘VALID’类型的padding, 其输出的height和width计算如下:
out_height = ceil(float(in_height – filter_height + 1) / float(strides[1]))
out_width = ceil(float(in_width – filter_width + 1) / float(strides[2]))
【注:tensorflow中的卷积,严格上来说是cross-correlation,而不是卷积。因为在计算的过程中,没有对filter进行翻转,而严格的卷积计算是需要对filter进行翻转的!!!】