tf.data.experimental.sample_from_datasets(
datasets,
weights=None,
seed=None
)
Defined in tensorflow/python/data/experimental/ops/interleave_ops.py
.
Samples elements at random from the datasets in datasets
.
Args:
datasets
: A list oftf.data.Dataset
objects with compatible structure.weights
: (Optional.) A list oflen(datasets)
floating-point values whereweights[i]
represents the probability with which an element should be sampled fromdatasets[i]
, or atf.data.Dataset
object where each element is such a list. Defaults to a uniform distribution acrossdatasets
.seed
: (Optional.) Atf.int64
scalartf.Tensor
, representing the random seed that will be used to create the distribution. Seetf.set_random_seed
for behavior.
Returns:
A dataset that interleaves elements from datasets
at random, according to
weights
if provided, otherwise with uniform probability.
Raises:
TypeError
: If thedatasets
orweights
arguments have the wrong type.ValueError
: If theweights
argument is specified and does not match the length of thedatasets
element.