将自定义卷积从PyTorch转换为Tensorflow（2.2.0）

Question

我目前正在尝试将自定义卷积从PyTorch转换为Tensorflow（V.2.2.0）。

卷积在PyTorch中定义为：

std::complex<double>

其中 self.quantizer = q = nn.Conv1d(1, 2*nq, kernel_size=1, bias=True) a = (nq-1) / gap #1st half = lines passing to (min+x,1) and (min+x+1/a,0) with x = {nq-1..0}*gap/(nq-1) q.weight.data[:nq] = -a q.bias.data[:nq] = torch.from_numpy(a*min + np.arange(nq, 0, -1)) # b = 1 + a*(min+x) #2nd half = lines passing to (min+x,1) and (min+x-1/a,0) with x = {nq-1..0}*gap/(nq-1) q.weight.data[nq:] = a q.bias.data[nq:] = torch.from_numpy(np.arange(2-nq, 2, 1) - a*min) # b = 1 - a*(min+x) # first and last one are special: just horizontal straight line q.weight.data[0] = q.weight.data[-1] = 0 q.bias.data[0] = q.bias.data[-1] = 1 ，nq = 20和min = 0。

我的重新实现看起来像这样：

max = 1

这些功能用作权重和偏差初始化：

my_weight = my_init_weight((1,1,nq*2))
q = tf.nn.convolution(input_q, my_weight)
q = tf.nn.bias_add(q, my_init_bias((40,1), tf.float32))

输入是一个矩阵，其形状用于PyTorch（首先使用通道），def my_init_weight(shape, dtype=None): weights = np.zeros(shape, dtype=np.float32) weights[:, :, :nq] = -a weights[:, :, nq:] = a weights[:, :, 0] = weights[:, :, -1] = 0 return tf.convert_to_tensor(weights, dtype=tf.float32) def my_init_bias(shape, dtype=None): weights = np.zeros(shape[0], dtype=np.float32) weights[:nq] = a*min + np.arange(nq, 0, -1) weights[nq:] = np.arange(2-nq, 2, 1) - a*min weights[0] = weights[-1] = 1 return weights ，用于Tensorflow（最后使用通道）的1681, 1, 1600，输出是1681, 1600, 1或1681, 40, 1600。因此应该是正确的，但是两个卷积的输出是不同的。

输入，输出：随机1681, 1600, 40图像上的Tensorflow：

100, 100

输入，输出：随机my_weight = my_init_weight((1,1,nq*2)) my_weight = tf.nn.bias_add(my_weight, my_init_bias((40,1), tf.float32)) q = tf.nn.convolution(test_conv, my_weight) q_left, q_right = tf.split(q, 2, axis=2) q = tf.math.minimum(q_left, q_right) nbs = tf.reduce_sum(q, axis=0) 图像上的PyTorch：

100, 100

Answer 1

好的，我找到了解决方法：

我忘记添加.clamp(min=0)。

将q = tf.clip_by_value(q, 0, tf.keras.backend.max(q))添加到

my_weight = my_init_weight((1,1,nq*2))
my_weight = tf.nn.bias_add(my_weight, my_init_bias((40,1), tf.float32))
q = tf.nn.convolution(test_conv, my_weight)

q_left, q_right = tf.split(q, 2, axis=2)
q = tf.math.minimum(q_left, q_right)
q = tf.clip_by_value(q, 0, tf.keras.backend.max(q)) <-----------
nbs = tf.reduce_sum(q, axis=0)

解决了这个问题。