如何在TensorFlow中累积梯度？

Question

我有一个类似于这个的问题。

由于资源有限，我使用深度模型（VGG-16）来训练三元组网络。我想要在大小为一的训练样例中积累 128 批次的梯度，然后传播误差并更新权重。

不清楚如何实现。我使用 tensorflow，但是任何实现/伪代码都可以。

- Hello Lili

---

为什么不使用您链接的问题中的答案？ - Pop

@Pop 因为我没有理解它们。我正在寻找更详细的东西（初学者级别）。 - Hello Lili

Answer 1

让我们一起来看看你链接中一个答案中提出的代码：

## Optimizer definition - nothing different from any classical example
opt = tf.train.AdamOptimizer()

## Retrieve all trainable variables you defined in your graph
tvs = tf.trainable_variables()
## Creation of a list of variables with the same shape as the trainable ones
# initialized with 0s
accum_vars = [tf.Variable(tf.zeros_like(tv.initialized_value()), trainable=False) for tv in tvs]
zero_ops = [tv.assign(tf.zeros_like(tv)) for tv in accum_vars]

## Calls the compute_gradients function of the optimizer to obtain... the list of gradients
gvs = opt.compute_gradients(rmse, tvs)

## Adds to each element from the list you initialized earlier with zeros its gradient (works because accum_vars and gvs are in the same order)
accum_ops = [accum_vars[i].assign_add(gv[0]) for i, gv in enumerate(gvs)]

## Define the training step (part with variable value update)
train_step = opt.apply_gradients([(accum_vars[i], gv[1]) for i, gv in enumerate(gvs)])

这段内容主要是向您的图形添加新的变量和运算符，以便您可以：

1. 在变量accum_vars中使用运算符accum_ops 累积梯度
2. 使用运算符train_step更新模型权重

然后，在训练时使用它，您需要按照下面的步骤进行（仍然来自您链接的答案）:

## The while loop for training
while ...:
    # Run the zero_ops to initialize it
    sess.run(zero_ops)
    # Accumulate the gradients 'n_minibatches' times in accum_vars using accum_ops
    for i in xrange(n_minibatches):
        sess.run(accum_ops, feed_dict=dict(X: Xs[i], y: ys[i]))
    # Run the train_step ops to update the weights based on your accumulated gradients
    sess.run(train_step)

- Pop

---

所以你把sess.run(train_step)放在了循环外面。这意味着权重更新会在计算完最后一个批次的梯度之后发生，对吗？如果我们将其放在循环内部，那么它会在每个时代之后发生，对吗？ - ARAT

Answer 2

TensorFlow 2.0兼容的解答：与上面提到的Pop答案和TensorFlow网站中提供的解释一致，以下是TensorFlow 2.0版本中累积梯度的代码：

def train(epochs):
  for epoch in range(epochs):
    for (batch, (images, labels)) in enumerate(dataset):
       with tf.GradientTape() as tape:
        logits = mnist_model(images, training=True)
        tvs = mnist_model.trainable_variables
        accum_vars = [tf.Variable(tf.zeros_like(tv.initialized_value()), trainable=False) for tv in tvs]
        zero_ops = [tv.assign(tf.zeros_like(tv)) for tv in accum_vars]
        loss_value = loss_object(labels, logits)

       loss_history.append(loss_value.numpy().mean())
       grads = tape.gradient(loss_value, tvs)
       #print(grads[0].shape)
       #print(accum_vars[0].shape)
       accum_ops = [accum_vars[i].assign_add(grad) for i, grad in enumerate(grads)]



    optimizer.apply_gradients(zip(grads, mnist_model.trainable_variables))
    print ('Epoch {} finished'.format(epoch))

# call the above function    
train(epochs = 3)

完整的代码可以在这个Github Gist中找到。

- user11530462

---

这句话的意思是：应该是 optimizer.apply_gradients(zip(accum_ops, mnist_model.trainable_variables)) 吗？ - Jared Nielsen

我也无法在tf.function中创建tf.Variable变量，你有什么建议吗？ - Jared Nielsen

我也遇到了跟随这段代码的困难，我在一个链接的问题中发布了一个可行的版本；https://dev59.com/4bjoa4cB1Zd3GeqPFdfj#62683800 - Ramiro R.C.