This returns a function outputting features and target based on the dict
of numpy arrays. The dict features has the same keys as the x.
numpy_input_fn(
x,
y = NULL,
batch_size = 128,
num_epochs = 1,
shuffle = NULL,
queue_capacity = 1000,
num_threads = 1
)dict of numpy array object.
numpy array object. NULL if absent.
Integer, size of batches to return.
Integer, number of epochs to iterate over data. If NULL
will run forever.
Boolean, if TRUE shuffles the queue. Avoid shuffle at
prediction time.
Integer, size of queue to accumulate.
Integer, number of threads used for reading and
enqueueing. In order to have predicted and repeatable order of reading and
enqueueing, such as in prediction and evaluation mode, num_threads should
be 1. #'
ValueError: if the shape of y mismatches the shape of
values in x (i.e., values in x have same shape). TypeError: x is not
a dict or shuffle is not bool.
Note that this function is still experimental and should only be used if necessary, e.g. feed in data that's dictionary of numpy arrays.
Other input functions:
input_fn()