使用tensorflow估计器头api抛出错误

如何解决使用tensorflow估计器头api抛出错误

我使用head api作为 https://www.tensorflow.org/api_docs/python/tf/estimator/Head，但抛出ValueError：trainable_variables不能为None。鉴于没有，这是代码

def _my_dnn_model_fn(features,labels,mode,params,config=None):
  # Optionally your callers can pass head to model_fn as a param.
  head = tf.estimator.MultiClassHead(3)

  feature_layer = tf.keras.layers.DenseFeatures(my_feature_columns)
  inputs = feature_layer(features)

  # Compute logits with tf.keras.layers API
  hidden_layer0 = tf.keras.layers.Dense(
      units=1000,activation="relu")(inputs)
  hidden_layer1 = tf.keras.layers.Dense(
      units=500,activation="relu")(hidden_layer0)
  logits = tf.keras.layers.Dense(
      units=head.logits_dimension,activation=None)(hidden_layer1)
  optimizer = tf.keras.optimizers.Adagrad(lr=0.1)
  return head.create_estimator_spec(
      features=features,labels=labels,mode=mode,logits=logits,optimizer=optimizer)
my_estimator = tf.estimator.Estimator(model_fn=_my_dnn_model_fn)
my_estimator.train(
    input_fn=lambda: input_fn(train,train_y,training=True),steps=5000)

您可以在实例化估算器段

之后运行它

https://colab.research.google.com/github/tensorflow/docs/blob/master/site/en/tutorials/estimator/premade.ipynb，