鉴于训练有素的LSTM模型,我想对单个时间步进行推理,即下例中的seq_length = 1
。在每个时间步之后,需要记住内部LSTM(存储器和隐藏)状态以用于下一个“批次”。在推理的最开始,给定输入计算内部LSTM状态init_c, init_h
。然后将它们存储在传递给LSTM的LSTMStateTuple
对象中。在训练期间,每个时间步都更新此状态。但是对于推理,我希望state
保存在批次之间,即初始状态只需要在开始时计算,之后LSTM状态应该在每个“批次”之后保存(n = 1) 。
我发现了这个与StackOverflow相关的问题:Tensorflow, best way to save state in RNNs?。但是这仅在state_is_tuple=False
时有效,但TensorFlow很快就会弃用此行为(请参阅rnn_cell.py)。 Keras似乎有一个很好的包装器可以使有状态 LSTM成为可能,但我不知道在TensorFlow中实现这一目标的最佳方法。 TensorFlow GitHub上的这个问题也与我的问题有关:https://github.com/tensorflow/tensorflow/issues/2838
建立有状态LSTM模型的任何好建议?
inputs = tf.placeholder(tf.float32, shape=[None, seq_length, 84, 84], name="inputs")
targets = tf.placeholder(tf.float32, shape=[None, seq_length], name="targets")
num_lstm_layers = 2
with tf.variable_scope("LSTM") as scope:
lstm_cell = tf.nn.rnn_cell.LSTMCell(512, initializer=initializer, state_is_tuple=True)
self.lstm = tf.nn.rnn_cell.MultiRNNCell([lstm_cell] * num_lstm_layers, state_is_tuple=True)
init_c = # compute initial LSTM memory state using contents in placeholder 'inputs'
init_h = # compute initial LSTM hidden state using contents in placeholder 'inputs'
self.state = [tf.nn.rnn_cell.LSTMStateTuple(init_c, init_h)] * num_lstm_layers
outputs = []
for step in range(seq_length):
if step != 0:
scope.reuse_variables()
# CNN features, as input for LSTM
x_t = # ...
# LSTM step through time
output, self.state = self.lstm(x_t, self.state)
outputs.append(output)
答案 0 :(得分:21)
我发现保存占位符中所有图层的整个状态最简单。
init_state = np.zeros((num_layers, 2, batch_size, state_size))
...
state_placeholder = tf.placeholder(tf.float32, [num_layers, 2, batch_size, state_size])
然后解压缩并创建一个LSTMStateTuples元组,然后再使用本机tensorflow RNN Api。
l = tf.unpack(state_placeholder, axis=0)
rnn_tuple_state = tuple(
[tf.nn.rnn_cell.LSTMStateTuple(l[idx][0], l[idx][1])
for idx in range(num_layers)]
)
RNN传递API:
cell = tf.nn.rnn_cell.LSTMCell(state_size, state_is_tuple=True)
cell = tf.nn.rnn_cell.MultiRNNCell([cell]*num_layers, state_is_tuple=True)
outputs, state = tf.nn.dynamic_rnn(cell, x_input_batch, initial_state=rnn_tuple_state)
state
- 变量将作为占位符被送到下一批。
答案 1 :(得分:6)
Tensorflow,在RNN中保存状态的最佳方法?实际上是我原来的问题。下面的代码是我如何使用状态元组。
with tf.variable_scope('decoder') as scope:
rnn_cell = tf.nn.rnn_cell.MultiRNNCell \
([
tf.nn.rnn_cell.LSTMCell(512, num_proj = 256, state_is_tuple = True),
tf.nn.rnn_cell.LSTMCell(512, num_proj = WORD_VEC_SIZE, state_is_tuple = True)
], state_is_tuple = True)
state = [[tf.zeros((BATCH_SIZE, sz)) for sz in sz_outer] for sz_outer in rnn_cell.state_size]
for t in range(TIME_STEPS):
if t:
last = y_[t - 1] if TRAINING else y[t - 1]
else:
last = tf.zeros((BATCH_SIZE, WORD_VEC_SIZE))
y[t] = tf.concat(1, (y[t], last))
y[t], state = rnn_cell(y[t], state)
scope.reuse_variables()
我只是创建一个工作正常的列表列表,而不是使用tf.nn.rnn_cell.LSTMStateTuple
。在这个例子中,我没有保存状态。但是,您可以轻松地从变量中取出状态,只使用assign来保存值。