tf.estimator.inputs.pandas_input_fn(
x,
y=None,
batch_size=128,
num_epochs=1,
shuffle=None,
queue_capacity=1000,
num_threads=1,
target_column='target'
)
Returns input function that would feed Pandas DataFrame into the model.
Args:
x
: pandasDataFrame
object.y
: pandasSeries
object orDataFrame
.None
if absent.batch_size
: int, size of batches to return.num_epochs
: int, number of epochs to iterate over data. If notNone
, read attempts that would exceed this value will raiseOutOfRangeError
.shuffle
: bool, whether to read the records in random order.queue_capacity
: int, size of the read queue. IfNone
, it will be set roughly to the size ofx
.num_threads
: 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.target_column
: str, name to give the target columny
. This parameter is not used wheny
is aDataFrame
.
Returns:
Function, that has signature of ()->(dict of features
, target
)
Raises:
ValueError
: ifx
already contains a column with the same name asy
, or if the indexes ofx
andy
don't match.ValueError
: if 'shuffle' is not provided or a bool.