TensorFlow计算tf.nn.conv2d

x_raw=np.array([
[2,5,3],
[3,4,2],
[41,1,1]
)

f_raw=np.数组(
[[
[2,1],
[3,4]
]
[4,1],
[1,2]
])

f=tf.常量(f_raw,dtype=tf.float32)
x=tf.常量(x_raw,dtype=tf.float32)

滤波器=tf.整形(f,[2,2,1,2])
图像=tf.重塑(x,[1,3,3,1])

tf.nn.conv2d(图像,过滤器,[1,1,1,1],“有效”).eval()
< /代码> 

但是TensorFlow的输出是关闭的:

< Buff行情>
数组([[[35.,33.],[37.,25.],[[35.,25.],[19.,15.]]],dtype=float32)
< /块引用>

我做错了什么?

我想用TensorFlow重现同样的结果tf.nn.conv2d:

x_raw = np.array([
    [2,5,3],
    [3,4,2],
    [4,1,1]
])

f_raw = np.array(
[[
    [2,1],
    [3,4]
],[
    [4,1],
    [1,2]   
]])

f = tf.constant(f_raw, dtype=tf.float32)
x = tf.constant(x_raw, dtype=tf.float32)

filter = tf.reshape(f, [2, 2, 1, 2])
image  = tf.reshape(x, [1, 3, 3, 1])

tf.nn.conv2d(image, filter, [1, 1, 1, 1], "VALID").eval()


但是TensorFlow的输出是关闭的:


数组([[[35.,33.],[37.,25.],[[35.,25.],[19.,15.]]],dtype=float32)


我做错了什么?