条件随机场（crf）及tensorflow代码实例

@tf_export("nn.sampled_softmax_loss")
def sampled_softmax_loss(weights,
                         biases,
                         labels,
                         inputs,
                         num_sampled,
                         num_classes,
                         num_true=1,
                         sampled_values=None,
                         remove_accidental_hits=True,
                         partition_strategy="mod",
                         name="sampled_softmax_loss",
                         seed=None):

    #num_sampled则是Sample Softmax时候用到的一个超参数，确定选几个词来对比优化
    
    '''
    weights: A `Tensor` of shape `[num_classes, dim]`, or a list of `Tensor`
        objects whose concatenation along dimension 0 has shape
        [num_classes, dim].  The (possibly-sharded) class embeddings.
    biases: A `Tensor` of shape `[num_classes]`.  The class biases.
    labels: A `Tensor` of type `int64` and shape `[batch_size,
        num_true]`. The target classes.  Note that this format differs from
        the `labels` argument of `nn.softmax_cross_entropy_with_logits`.
    inputs: A `Tensor` of shape `[batch_size, dim]`.  The forward
        activations of the input network.
    num_sampled: An `int`.  The number of classes to randomly sample per batch.
    num_classes: An `int`. The number of possible classes.
    '''

def crf_log_likelihood(inputs,
                       tag_indices,
                       sequence_lengths,
                       transition_params=None):
  """Computes the log-likelihood of tag sequences in a CRF.
  Args:
    inputs: A [batch_size, max_seq_len, num_tags] tensor of unary potentials
        to use as input to the CRF layer.
    tag_indices: A [batch_size, max_seq_len] matrix of tag indices for which we
        compute the log-likelihood.
    sequence_lengths: A [batch_size] vector of true sequence lengths.
    transition_params: A [num_tags, num_tags] transition matrix, if available.
    """

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
   
 
TIME_STEPS = 15#20 # backpropagation through time 的time_steps
BATCH_SIZE = 1#50
INPUT_SIZE = 1 # x数据输入size
LR = 0.05  # learning rate
num_tags = 2  
# 定义一个生成数据的 get_batch function:
def get_batch(): 
    xs = np.array([[2, 3, 4, 5, 5, 5, 1, 5, 3, 2, 5, 5, 5, 3, 5]])
    res = np.array([[0, 0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1]])
    return [xs[:, :, np.newaxis], res]
   
# 定义 CRF 的主体结构
class CRF(object):
    def __init__(self, n_steps, input_size, num_tags, batch_size):
        self.n_steps = n_steps
        self.input_size = input_size
        self.num_tags = num_tags
        self.batch_size = batch_size
        self.xs = tf.placeholder(tf.float32, [None, self.n_steps, self.input_size], name='xs')
        self.ys = tf.placeholder(tf.int32, [self.batch_size, self.n_steps], name='ys')
        #将输入 batch_size x seq_length x input_size   映射到 batch_size x seq_length x num_tags
        
        weights = tf.get_variable("weights", [self.input_size, self.num_tags])
        matricized_x_t = tf.reshape(self.xs, [-1, self.input_size])
        matricized_unary_scores = tf.matmul(matricized_x_t, weights)
        unary_scores = tf.reshape(matricized_unary_scores, [self.batch_size, self.n_steps, self.num_tags])   
          
        sequence_lengths = np.full(self.batch_size,self.n_steps,dtype=np.int32) 
         
        log_likelihood,transition_params = tf.contrib.crf.crf_log_likelihood(unary_scores,self.ys,sequence_lengths)
 
        self.pred, viterbi_score = tf.contrib.crf.crf_decode(unary_scores, transition_params, sequence_lengths)
        # add a training op to tune the parameters.
        self.cost = tf.reduce_mean(-log_likelihood)
        self.train_op = tf.train.AdamOptimizer(LR).minimize(self.cost)
   
     
   
   
# 训练 CRF
if __name__ == '__main__':
      
    # 搭建 CRF 模型
    model = CRF(TIME_STEPS, INPUT_SIZE, num_tags, BATCH_SIZE)
    sess = tf.Session()
    sess.run(tf.global_variables_initializer())
     
    # matplotlib可视化
    plt.ion()  # 设置连续 plot
    plt.show()   
    # 训练多次
    for i in range(150):
        xs, res = get_batch()  # 提取 batch data
        #print(res.shape)
        # 初始化 data
        feed_dict = {
            model.xs: xs,
            model.ys: res,
        }         
        # 训练
        _, cost,pred = sess.run(
            [model.train_op, model.cost,  model.pred],
            feed_dict=feed_dict)
   
        # plotting
 
        x = xs.reshape(-1,1)
        r = res.reshape(-1, 1)
        p = pred.reshape(-1, 1)
         
        x = range(len(x))
         
        plt.clf()
        plt.plot(x, r, 'r', x, p, 'b--')
        plt.ylim((-1.2, 1.2))
        plt.draw()
        plt.pause(0.3)  # 每 0.3 s 刷新一次
         
        # 打印 cost 结果
        if i % 20 == 0:
            print('cost: ', round(cost, 4))