In your comment, what does FeatureColumn in tf.learn means ?
Yun jae Choi
1

Thanks, updated n_classes computation to include missing value that categorical processor adds.

FeatureColumn is an abstraction for pre-processing features. They allow to embed, normalize, make crosses and etc features before passing to standard models like DNNClassifier.

You can find a bit more details here — https://www.tensorflow.org/versions/r0.11/tutorials/linear/overview.html#feature-columns-and-transformations

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.