我正在使用下面粘贴的代码。代码的“前向”部分似乎通过“断言root_emb == 1 + emb [0] * emb [1]”传递。但是，一旦采取了训练步骤（断言后面的行），就会出现一个奇怪的错误，表明在循环过程中写入了TensorArray的问题。
tensorflow.python.framework.errors.InvalidArgumentError：TensorArray TensorArray @ gradient：无法读取TensorArray索引2，因为 它尚未写入。 [[节点： gradients / while / TensorArrayWrite_grad / TensorArrayRead = TensorArrayRead [_class = [“loc：@TensorArray”]，dtype = DT_FLOAT， _device = “/作业：本地主机/复制：0 /任务：0 / CPU：0”]（梯度/同时/ TensorArrayWrite_grad / TensorArrayGrad / TensorArrayGrad， 梯度/同时/ TensorArrayWrite_grad / TensorArrayRead / StackPop， 梯度/同时/ TensorArrayWrite_grad / TensorArrayGrad / gradient_flow）]] 由op u'gradients / while / TensorArrayWrite_grad / TensorArrayRead'引起， 定义于：文件“minimal.py”，第82行，in model = TreeRNN（8,1,1，degree = 2）文件“minimal.py”，第61行， init self.grad = tf.gradients（self.loss，self.params）文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/gradients.py”， 第481行，渐变 in_grads = _AsList（grad_fn（op，* out_grads））文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/tensor_array_grad.py”， 第115行，在_TensorArrayWriteGrad中 grad = g.read（index）File“/Library/Python/2.7/site-packages/tensorflow/python/ops/tensor_array_ops.py”， 第177行，正在阅读中 dtype = self._dtype，name = name）文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/gen_data_flow_ops.py”， 第781行，在_tensor_array_read中 flow_in = flow_in，dtype = dtype，name = name）文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/op_def_library.py”， 第694行，在apply_op中 op_def = op_def）文件“/Library/Python/2.7/site-packages/tensorflow/python/framework/ops.py”， 第2154行，在create_op中 original_op = self._default_original_op，op_def = op_def）文件“/Library/Python/2.7/site-packages/tensorflow/python/framework/ops.py”， 第1154行，在 init 中 self._traceback = _extract_stack（）
...最初创建为op u'while / TensorArrayWrite'， 定义于：文件“minimal.py”，第82行，in model = TreeRNN（8,1,1，degree = 2）文件“minimal.py”，第50行， init loop_vars =（self.time，node_emb，tf.zeros（[1]）））文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/control_flow_ops.py”， 第1681行，在 back_prop = back_prop，swap_memory = swap_memory，name = name）文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/control_flow_ops.py”， 第1671行，在while_loop中 result = context.BuildLoop（cond，body，loop_vars）File“/Library/Python/2.7/site-packages/tensorflow/python/ops/control_flow_ops.py”， 第1572行，在BuildLoop中 body_result = body（* vars_for_body_with_tensor_arrays）文件“minimal.py”，第43行，在_reurrence中 new_node_emb = node_emb.write（children_and_parent [-1]，parent_emb）文件 “/Library/Python/2.7/site-packages/tensorflow/python/ops/tensor_array_ops.py” 第200行，写在 name = name）文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/gen_data_flow_ops.py”， 第875行，在_tensor_array_write中 value = value，flow_in = flow_in，name = name）文件“/Library/Python/2.7/site-packages/tensorflow/python/ops/op_def_library.py”， 第694行，在apply_op中 op_def = op_def）
import numpy as np
import tensorflow as tf
from tensorflow.python.ops import tensor_array_ops, control_flow_ops
class TreeRNN(object):
def __init__(self, num_emb, emb_dim, output_dim, degree=2, learning_rate=0.01):
self.num_emb = num_emb
self.emb_dim = emb_dim
self.output_dim = output_dim
self.degree= degree
self.learning_rate = tf.Variable(float(learning_rate), trainable=False)
self.embeddings = tf.Variable(self.init_matrix([self.num_emb, self.emb_dim]))
self.recursive_unit = self.create_recursive_unit()
self.W_out = tf.Variable(self.init_matrix([self.output_dim, self.emb_dim]))
self.b_out = tf.Variable(self.init_vector([self.output_dim]))
self.x = tf.placeholder(tf.int32, shape=[None]) # word indices
self.tree = tf.placeholder(tf.int32, shape=[None, self.degree + 1])
self.y = tf.placeholder(tf.float32, shape=[self.output_dim])
num_words, = tf.unpack(tf.shape(self.x), 1) # also num leaves
emb_x = tf.gather(self.embeddings, self.x)
node_emb = tensor_array_ops.TensorArray(
dtype=tf.float32, size=num_words - 1, dynamic_size=True,
clear_after_read=False)
node_emb = node_emb.unpack(emb_x)
num_nodes, _ = tf.unpack(tf.shape(self.tree), 2) # num internal nodes
tree_traversal = tensor_array_ops.TensorArray(
dtype=tf.int32, size=num_nodes)
tree_traversal = tree_traversal.unpack(self.tree)
def _recurrence(t, node_emb, _):
node_info = tree_traversal.read(t)
children_and_parent = tf.unpack(node_info, self.degree + 1)
child_emb = []
for i in xrange(self.degree):
child_emb.append(node_emb.read(children_and_parent[i]))
parent_emb = self.recursive_unit(child_emb)
new_node_emb = node_emb.write(children_and_parent[-1], parent_emb)
return t + 1, new_node_emb, parent_emb
self.time = tf.constant(0, dtype=tf.int32, name='time')
_, _, final_emb = control_flow_ops.While(
cond=lambda t, _1, _2: t < num_nodes,
body=_recurrence,
loop_vars=(self.time, node_emb, tf.zeros([1])))
self.final_state = final_emb
self.pred_y = self.activation(
tf.matmul(self.W_out, tf.reshape(self.final_state, [self.emb_dim, 1]))
+ self.b_out)
self.loss = self.loss_fn(self.y, self.pred_y)
self.params = tf.trainable_variables()
opt = tf.train.GradientDescentOptimizer(self.learning_rate)
self.grad = tf.gradients(self.loss, self.params)
self.updates = opt.apply_gradients(zip(self.grad, self.params))
def init_matrix(self, shape):
return tf.random_normal(shape, stddev=0.1)
def init_vector(self, shape):
return tf.zeros(shape)
def create_recursive_unit(self):
def unit(child_emb): # very simple
return 1 + child_emb[0] * child_emb[1]
return unit
def activation(self, inp):
return tf.sigmoid(inp)
def loss_fn(self, y, pred_y):
return tf.reduce_sum(tf.square(y - pred_y))
model = TreeRNN(8, 1, 1, degree=2)
sess = tf.Session()
sess.run(tf.initialize_all_variables())
root_emb = sess.run([model.final_state],
feed_dict={model.x: np.array([0, 1]), model.tree: np.array([[0, 1, 2]])})
emb, = sess.run([model.embeddings])
assert root_emb == 1 + emb[0] * emb[1]
out = sess.run([model.updates, model.loss],
feed_dict={model.x: np.array([0, 1]),
model.tree: np.array([[0, 1, 2]]),
model.y: np.array([0])})
答案 0 :(得分：0)
在tf.while_loop
中设置parallel_iterations = 1