概述

我试图将Tensorflow的官方基本word2vec实现转换为使用tf.Estimator.
问题是当使用Tensorflow Estimators时,丢失函数(sampled_softmax_loss或nce_loss)会出错.它在原始实现中完美地运行.

这是Tensorflow的官方基本word2vec实现：

https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/tutorials/word2vec/word2vec_basic.py

以下是我实施此代码的Google Colab笔记本,该代码正常运行.

https://colab.research.google.com/drive/1nTX77dRBHmXx6PEF5pmYpkIVxj_TqT5I

这是Google Colab笔记本,我在其中更改了代码,因此它使用Tensorflow Estimator,它不起作用.

https://colab.research.google.com/drive/1IVDqGwMx6BK5-Bgrw190jqHU6tt3ZR3e

为方便起见,这里是我定义model_fn的Estimator版本的精确代码

batch_size = 128
embedding_size = 128  # Dimension of the embedding vector.
skip_window = 1  # How many words to consider left and right.
num_skips = 2  # How many times to reuse an input to generate a label.
num_sampled = 64  # Number of negative examples to sample.

def my_model( features,labels,mode,params):

    with tf.name_scope('inputs'):
        train_inputs = features
        train_labels = labels

    with tf.name_scope('embeddings'):
        embeddings = tf.Variable(
          tf.random_uniform([vocabulary_size,embedding_size],-1.0,1.0))
        embed = tf.nn.embedding_lookup(embeddings,train_inputs)

    with tf.name_scope('weights'):
        nce_weights = tf.Variable(
          tf.truncated_normal(
              [vocabulary_size,stddev=1.0 / math.sqrt(embedding_size)))
    with tf.name_scope('biases'):
        nce_biases = tf.Variable(tf.zeros([vocabulary_size]))

    with tf.name_scope('loss'):
        loss = tf.reduce_mean(
            tf.nn.nce_loss(
                weights=nce_weights,biases=nce_biases,labels=train_labels,inputs=embed,num_sampled=num_sampled,num_classes=vocabulary_size))

    tf.summary.scalar('loss',loss)

    if mode == "train":
        with tf.name_scope('optimizer'):
            optimizer = tf.train.GradientDescentOptimizer(1.0).minimize(loss)

        return tf.estimator.EstimatorSpec(mode,loss=loss,train_op=optimizer)

这里是我称之为估算和培训的地方

word2vecEstimator = tf.estimator.Estimator(
        model_fn=my_model,params={
            'batch_size': 16,'embedding_size': 10,'num_inputs': 3,'num_sampled': 128,'batch_size': 16
        })

word2vecEstimator.train(
    input_fn=generate_batch,steps=10)

这是我在调用Estimator培训时得到的错误消息：

INFO:tensorflow:Calling model_fn.
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)

  
   python3.6/dist-packages/tensorflow/python/estimator/estimator.py in train(self,input_fn,hooks,steps,max_steps,saving_listeners)
    352 
    353       saving_listeners = _check_listeners_type(saving_listeners)
--> 354       loss = self._train_model(input_fn,saving_listeners)
    355       logging.info('Loss for final step: 
   %s.',loss)
    356       return self

/usr/local/lib/
   python3.6/dist-packages/tensorflow/python/estimator/estimator.py in _train_model(self,saving_listeners)
   1205       return self._train_model_distributed(input_fn,saving_listeners)
   1206     else:
-> 1207       return self._train_model_default(input_fn,saving_listeners)
   1208 
   1209   def _train_model_default(self,saving_listeners):

/usr/local/lib/
   python3.6/dist-packages/tensorflow/python/estimator/estimator.py in _train_model_default(self,saving_listeners)
   1235       worker_hooks.extend(input_hooks)
   1236       estimator_spec = self._call_model_fn(
-> 1237           features,model_fn_lib.ModeKeys.TRAIN,self.con
   fig)
   1238       global_step_tensor = training_util.get_global_step(g)
   1239       return self._train_with_estimator_spec(estimator_spec,worker_hooks,/usr/local/lib/
   python3.6/dist-packages/tensorflow/python/estimator/estimator.py in _call_model_fn(self,features,con
   fig)
   1193 
   1194     logging.info('Calling model_fn.')
-> 1195     model_fn_results = self._model_fn(features=features,**
   kwargs)
   1196     logging.info('Done calling model_fn.')
   1197 


   
    python3.6/dist-packages/tensorflow/python/ops/nn_impl.py in nce_loss(weights,biases,inputs,num_sampled,num_classes,num_true,sampled_values,remove_accidental_hits,partition_strategy,name)
   1246       remove_accidental_hits=remove_accidental_hits,1247       partition_strategy=partition_strategy,-> 1248       name=name)
   1249   sampled_losses = sigmoid_cross_entropy_with_logits(
   1250       labels=labels,logits=logits,name="sampled_losses")

/usr/local/lib/
    python3.6/dist-packages/tensorflow/python/ops/nn_impl.py in _compute_sampled_logits(weights,subtract_log_q,name,seed)
   1029   with ops.name_scope(name,"compute_sampled_logits",1030                       weights + [biases,labels]):
-> 1031     if labels.dtype != dtypes.int64:
   1032       labels = math_ops.cast(labels,dtypes.int64)
   1033     labels_flat = array_ops.reshape(labels,[-1])

TypeError: data type not understood

   
  

编辑：根据请求,这是input_fn的典型输出

print(generate_batch(batch_size = 8,num_skips = 2,skip_window = 1))

(array([3081,3081,12,6,195,195],dtype=int32),array([[5234],[  12],[   6],[3081],[ 195],[   2]],dtype=int32))

word2vecEstimator.train(
    input_fn=generate_batch,steps=10)

使用generate_batch()调用该函数.
但我认为你必须将一些值传递给函数.

总结

以上是编程之家为你收集整理的python – 转换Tensorflow图形以使用Estimator,使用`sampled_softmax_loss`或`nce_loss`在损失函数中获取’TypeError：数据类型不被理解’全部内容，希望文章能够帮你解决python – 转换Tensorflow图形以使用Estimator,使用`sampled_softmax_loss`或`nce_loss`在损失函数中获取’TypeError：数据类型不被理解’所遇到的程序开发问题。

如果您也喜欢它,动动您的小指点个赞吧