Why train_input_fn outputs 3 values

[minhhoangbui]

In this function you outputs 3 values [x_train, x_len_train, y_train]:
def train_input_fn(): dataset = tf.data.Dataset.from_tensor_slices((x_train, x_len_train, y_train)) dataset = dataset.shuffle(buffer_size=len(x_train_variable)) dataset = dataset.batch(100) dataset = dataset.map(parser) dataset = dataset.repeat() iterator = dataset.make_one_shot_iterator() return iterator.get_next()

However, in Google document, they say:

The return value must be a two-element tuple organized as follows: :

The first element must be a dict in which each input feature is a key, and then a list of values for the training batch.

The second element is a list of labels for the training batch.

So I don't really understand that how custom Estimator can work with a tuple of 3 values
Thanks in advance

[eisenjulian]

The dataset API has functional-like interface for chaining operation on the input pipeline. In this case we are using the map method with the parser argument that takes the tuple of three arrays into the two-element structure that you describe