TensorFlow 2.0 dataset.__iter__() 仅在启用即时执行时才受支持

Question

我在 TensorFlow 2 中使用以下自定义训练代码：

def parse_function(filename, filename2):
    image = read_image(fn)
    def ret1(): return image, read_image(fn2), 0
    def ret2(): return image, preprocess(image), 1
    return tf.case({tf.less(tf.random.uniform([1])[0], tf.constant(0.5)): ret2}, default=ret1)

dataset = tf.data.Dataset.from_tensor_slices((train,shuffled_train))
dataset = dataset.shuffle(len(train))
dataset = dataset.map(parse_function, num_parallel_calls=4)
dataset = dataset.batch(1)
dataset = dataset.prefetch(buffer_size=4)

@tf.function
def train(model, dataset, optimizer):
    for x1, x2, y in enumerate(dataset):
        with tf.GradientTape() as tape:
            left, right = model([x1, x2])
            loss = contrastive_loss(left, right, tf.cast(y, tf.float32))
        gradients = tape.gradient(loss, model.trainable_variables)
        optimizer.apply_gradients(zip(gradients, model.trainable_variables))

siamese_net.compile(optimizer=tf.keras.optimizers.RMSprop(learning_rate=1e-3))
train(siamese_net, dataset, tf.keras.optimizers.RMSprop(learning_rate=1e-3))

此代码给出错误：

dataset.__iter__() is only supported when eager execution is enabled.

但是，它在 TensorFlow 2.0 中默认启用。 tf.executing_eagerly() 也 returns 'True'.

Answer 1

我通过将 train 函数更改为以下内容来解决此问题：

def train(model, dataset, optimizer):
    for step, (x1, x2, y) in enumerate(dataset):
        with tf.GradientTape() as tape:
            left, right = model([x1, x2])
            loss = contrastive_loss(left, right, tf.cast(y, tf.float32))
        gradients = tape.gradient(loss, model.trainable_variables)
        optimizer.apply_gradients(zip(gradients, model.trainable_variables))

两个更改是删除@tf.function 和修复枚举。

Answer 2

我通过在导入 tensorflow 后启用 eager execution 来修复它：

import tensorflow as tf

tf.enable_eager_execution()

参考：Tensorflow

Answer 3

如果您在

之后使用 Jupyter notebook

import tensorflow as tf

tf.enable_eager_execution()

你需要重启内核才能正常工作