'learning_rate:0' refers to a Tensor which does not exist

[Jarlonyan]

wide_and_deep预测用户收入的代码中，出现这个错误，查了google，没有找到合适的答案。请问这个是什么问题导致的呢？

KeyError: "The name 'learning_rate:0' refers to a Tensor which does not exist. The operation, 'learning_rate', does not exist in the graph."

[aianaconda]

no learning_rate variable in this code，are you sure you use my code?

[Jarlonyan]

我是参考你们的这个代码：link TensorFlow_Engineering_Implementation

然后提示没有learning_rate，我就在这里添加了learning_rate，但是依然报同样的错误

def build_estimator(model_dir, model_type):
    #按照指定的模型，生成估算器对象
    wide_columns, deep_columns = build_model_columns()
    hidden_units = [512, 256, 128]
    run_config = tf.estimator.RunConfig().replace(
        session_config=tf.ConfigProto(device_count={'GPU':0}),
        save_checkpoints_steps=1000
    )

    if model_type == 'wide':
        return tf.estimator.LinearClassifier(model_dir=model_dir, feature_columns=wide_columns, config=run_config)
    elif model_type == 'deep':
        return tf.estimator.DNNClassifier(model_dir=model_dir, feature_columns=deep_columns, hidden_units=hidden_units, config=run_config)
    else:
        return tf.estimator.DNNLinearCombinedClassifier(
                    model_dir=model_dir, \
                    linear_feature_columns=wide_columns, \
                    dnn_feature_columns=deep_columns, \
                    dnn_hidden_units=hidden_units, \
                    dnn_optimizer=tf.train.AdamOptimizer(
                        learning_rate=0.0001
                    ),
                    config=run_config
        )