tf.contrib.training.resample_at_rate( inputs, rates, scope=None, seed=None, back_prop=False )
See the guide: Training (contrib) > Online data resampling
inputs tensors, stochastically resamples each at a given rate.
For example, if the inputs are
[[a1, a2], [b1, b2]] and the rates
[3, 1], then the return value may look like
a2, a1, a1], [b1, b2, b1, b1]]. However, many other outputs are
possible, since this is stochastic -- averaged over many repeated
calls, each set of inputs should appear in the output
the number of invocations.
inputs: A list of tensors, each of which has a shape of
rates: A tensor of shape
[batch_size]containing the resampling rates for each input.
scope: Scope for the op.
seed: Random seed to use.
back_prop: Whether to allow back-propagation through this op.
Selections from the input tensors.