keras.layers.LSTM参数 API

tf.keras.layers.LSTM(
    units,
    activation="tanh",
    recurrent_activation="sigmoid",
    use_bias=True,
    kernel_initializer="glorot_uniform",
    recurrent_initializer="orthogonal",
    bias_initializer="zeros",
    unit_forget_bias=True,
    kernel_regularizer=None,
    recurrent_regularizer=None,
    bias_regularizer=None,
    activity_regularizer=None,
    kernel_constraint=None,
    recurrent_constraint=None,
    bias_constraint=None,
    dropout=0.0,
    recurrent_dropout=0.0,
    implementation=2,
    return_sequences=False,
    return_state=False,
    go_backwards=False,
    stateful=False,
    time_major=False,
    unroll=False,
    **kwargs
)

 

For example:

 

>>> inputs = tf.random.normal([32, 10, 8])
>>> lstm = tf.keras.layers.LSTM(4)
>>> output = lstm(inputs)
>>> print(output.shape)
(32, 4)
>>> lstm = tf.keras.layers.LSTM(4, return_sequences=True, return_state=True)
>>> whole_seq_output, final_memory_state, final_carry_state = lstm(inputs)
>>> print(whole_seq_output.shape)
(32, 10, 4)
>>> print(final_memory_state.shape)
(32, 4)
>>> print(final_carry_state.shape)
(32, 4)

 

Arguments

  • units: Positive integer, dimensionality of the output space.
  • activation: Activation function to use. Default: hyperbolic tangent (tanh). If you pass None, no activation is applied (ie. "linear" activation: a(x) = x).
  • recurrent_activation: Activation function to use for the recurrent step. Default: sigmoid (sigmoid). If you pass None, no activation is applied (ie. "linear" activation: a(x) = x).
  • use_bias: Boolean (default True), whether the layer uses a bias vector.
  • kernel_initializer: Initializer for the kernel weights matrix, used for the linear transformation of the inputs. Default: glorot_uniform.
  • recurrent_initializer: Initializer for the recurrent_kernel weights matrix, used for the linear transformation of the recurrent state. Default: orthogonal.
  • bias_initializer: Initializer for the bias vector. Default: zeros.
  • unit_forget_bias: Boolean (default True). If True, add 1 to the bias of the forget gate at initialization. Setting it to true will also force bias_initializer="zeros". This is recommended in Jozefowicz et al..
  • kernel_regularizer: Regularizer function applied to the kernel weights matrix. Default: None.
  • recurrent_regularizer: Regularizer function applied to the recurrent_kernel weights matrix. Default: None.
  • bias_regularizer: Regularizer function applied to the bias vector. Default: None.
  • activity_regularizer: Regularizer function applied to the output of the layer (its "activation"). Default: None.
  • kernel_constraint: Constraint function applied to the kernel weights matrix. Default: None.
  • recurrent_constraint: Constraint function applied to the recurrent_kernel weights matrix. Default: None.
  • bias_constraint: Constraint function applied to the bias vector. Default: None.
  • dropout: Float between 0 and 1. Fraction of the units to drop for the linear transformation of the inputs. Default: 0.
  • recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for the linear transformation of the recurrent state. Default: 0.
  • implementation: Implementation mode, either 1 or 2. Mode 1 will structure its operations as a larger number of smaller dot products and additions, whereas mode 2 will batch them into fewer, larger operations. These modes will have different performance profiles on different hardware and for different applications. Default: 2.
  • return_sequences: Boolean. Whether to return the last output. in the output sequence, or the full sequence. Default: False.
  • return_state: Boolean. Whether to return the last state in addition to the output. Default: False.
  • go_backwards: Boolean (default False). If True, process the input sequence backwards and return the reversed sequence.
  • stateful: Boolean (default False). If True, the last state for each sample at index i in a batch will be used as initial state for the sample of index i in the following batch.
  • time_major: The shape format of the inputs and outputs tensors. If True, the inputs and outputs will be in shape [timesteps, batch, feature], whereas in the False case, it will be [batch, timesteps, feature]. Using time_major = True is a bit more efficient because it avoids transposes at the beginning and end of the RNN calculation. However, most TensorFlow data is batch-major, so by default this function accepts input and emits output in batch-major form.
  • unroll: Boolean (default False). If True, the network will be unrolled, else a symbolic loop will be used. Unrolling can speed-up a RNN, although it tends to be more memory-intensive. Unrolling is only suitable for short sequences.

Call arguments

  • inputs: A 3D tensor with shape [batch, timesteps, feature].
  • mask: Binary tensor of shape [batch, timesteps] indicating whether a given timestep should be masked (optional, defaults to None).
  • training: Python boolean indicating whether the layer should behave in training mode or in inference mode. This argument is passed to the cell when calling it. This is only relevant if dropout or recurrent_dropout is used (optional, defaults to None).
  • initial_state: List of initial state tensors to be passed to the first call of the cell (optional, defaults to None which causes creation of zero-filled initial state tensors).

 

https://keras.io/api/layers/recurrent_layers/lstm/

 

  • 2
    点赞
  • 19
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
LSTM(长短时记忆网络)是一种常用的循环神经网络。在Keras中,可以使用keras.layers.LSTM来构建LSTM模型。 keras.layers.LSTM的主要参数如下: - units:LSTM层中的神经元个数。 - activation:激活函数。默认为"tanh"。 - recurrent_activation:循环层的激活函数。默认为"hard_sigmoid"。 - use_bias:是否使用偏置项。默认为True。 - kernel_initializer:权重矩阵的初始化方法。默认为"glorot_uniform"。 - recurrent_initializer:循环权重矩阵的初始化方法。默认为"orthogonal"。 - bias_initializer:偏置项的初始化方法。默认为"zeros"。 - unit_forget_bias:是否为LSTM的遗忘门添加偏置项。默认为True。 - kernel_regularizer:权重矩阵的正则化方法。默认为None。 - recurrent_regularizer:循环权重矩阵的正则化方法。默认为None。 - bias_regularizer:偏置项的正则化方法。默认为None。 - activity_regularizer:输出的正则化方法。默认为None。 - dropout:dropout比率,用于输入的遗忘门。默认为0。 - recurrent_dropout:dropout比率,用于循环层的遗忘门。默认为0。 - implementation:实现模式,可以选择使用GPU进行计算。默认为1。 - return_sequences:是否返回所有时间步的输出。默认为False。 - return_state:是否返回最后一个时间步的状态。默认为False。 - go_backwards:是否反向处理输入序列。默认为False。 - stateful:是否在时间步之间保留网络状态。默认为False。 - unroll:是否展开循环。默认为False。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值