我正在使用Tensorflow Estimator API来训练一些模型，但是发现我必须为每个功能使用tf.feature_column
我的数据集由以下列组成：
[＆＃39; age＆＃39;，＆＃39; workclass＆＃39;，＆＃39; education＆＃39;，＆＃39; education_num＆＃39;，＆＃39; marital_status＆＃39;， ＆＃39;职业＆＃39;，＆＃39;关系＆＃39;，＆＃39;种族＆＃39;，＆＃39;性别＆＃39;，＆＃39; capital_gain＆＃39;， ＆＃39; capital_loss＆＃39;，＆＃39; hours_per_week＆＃39;，＆＃39; native_country＆＃39;，＆＃39; income_bracket＆＃39;]
为什么我必须像这样做
gender = tf.feature_column.categorical_column_with_hash_bucket("gender", hash_bucket_size=10)
occupation = tf.feature_column.categorical_column_with_hash_bucket("occupation", hash_bucket_size=1000)
marital_status = tf.feature_column.categorical_column_with_hash_bucket("marital_status", hash_bucket_size=1000)
relationship = tf.feature_column.categorical_column_with_hash_bucket("relationship", hash_bucket_size=1000)
education = tf.feature_column.categorical_column_with_hash_bucket("education", hash_bucket_size=1000)
workclass = tf.feature_column.categorical_column_with_hash_bucket("workclass", hash_bucket_size=1000)
native_country = tf.feature_column.categorical_column_with_hash_bucket("native_country", hash_bucket_size=1000)
..等，似乎不是处理功能预处理的最佳方式。有没有更好的方法，或者如果我们去tf.estimator