可以看出,我在“error_giving_notebook”中使用了 tf.function 装饰器,它会抛出 ValueError,而同一笔记本除了删除 tf.function 装饰器之外没有任何更改,在“non_problematic_notebook”中运行顺利。可能是什么原因?
当您尝试在 TF 2.0 中使用函数装饰器时,请在导入 TensorFlow 后使用以下行立即启用 run 函数:
tf.config.experimental_run_functions_eagerly(True)
由于上述内容已被弃用(不再是实验性的?),请使用以下内容:
tf.config.run_functions_eagerly(True)
如果您想了解更多信息,请参阅此链接。
这里的问题出在类 conv2d 的 call 方法的返回值上:
if self.bias:
if self.pad == 'REFLECT':
self.p = (self.filter_size - 1) // 2
self.x = tf.pad(inputs, [[0, 0], [self.p, self.p], [self.p, self.p], [0, 0]], 'REFLECT')
return Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride),
padding='VALID', use_bias=True, kernel_initializer=self.w, bias_initializer=self.b)(self.x)
else:
return Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride),
padding=self.pad, use_bias=True, kernel_initializer=self.w, bias_initializer=self.b)(inputs)
else:
if self.pad == 'REFLECT':
self.p = (self.filter_size - 1) // 2
self.x = tf.pad(inputs, [[0, 0], [self.p, self.p], [self.p, self.p], [0, 0]], 'REFLECT')
return Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride),
padding='VALID', use_bias=False, kernel_initializer=self.w)(self.x)
else:
return Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride),
padding=self.pad, use_bias=False, kernel_initializer=self.w)(inputs)
通过返回 Conv2D 对象,每次调用时都会创建 tf.Variable(权重、卷积层的偏差)
predictions = model(images)
在你的 tf 修饰的函数中。因此,例外。
解决此问题的一种可能方法是更改 conv2d 类中的构建和调用方法,如下所示:
def build(self, inputs):
self.w = tf.random_normal_initializer(mean=0.0, stddev=1e-4)
if self.bias:
self.b = tf.constant_initializer(0.0)
else:
self.b = None
self.conv_a = Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride), padding='VALID', use_bias=True, kernel_initializer=self.w, bias_initializer=self.b)
self.conv_b = Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride), padding=self.pad, use_bias=True, kernel_initializer=self.w, bias_initializer=self.b)
self.conv_c = Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride), padding='VALID', use_bias=False, kernel_initializer=self.w)
self.conv_d = Conv2D(filters=self.filter_num, kernel_size=(self.filter_size, self.filter_size), strides=(self.stride, self.stride),padding=self.pad, use_bias=False, kernel_initializer=self.w)
def call(self, inputs):
if self.bias:
if self.pad == 'REFLECT':
self.p = (self.filter_size - 1) // 2
self.x = tf.pad(inputs, [[0, 0], [self.p, self.p], [self.p, self.p], [0, 0]], 'REFLECT')
return self.conv_a(self.x)
else:
return self.conv_b(inputs)
else:
if self.pad == 'REFLECT':
self.p = (self.filter_size - 1) // 2
self.x = tf.pad(inputs, [[0, 0], [self.p, self.p], [self.p, self.p], [0, 0]], 'REFLECT')
return self.conv_c(self.x)
else:
return self.conv_d(inputs)
为了更好地理解 AutoGraph 以及 @tf.function 的工作原理,我建议看一下 this
当您尝试使用 TF 2.0 中的函数装饰器时,请在导入 TensorFlow 后使用以下行来立即启用 run 函数:
import tensorflow as tf
tf.config.run_functions_eagerly(True)
tf.config.run_functions_eagerly(True)
为我工作