[FEA] The NVTabular dataloader + sample weights

[ethem-kinginthenorth]

Is your feature request related to a problem? Please describe.
I wish I could use the NVTabular dataloader to do what TF does with sample weights. Here is the link: https://www.tensorflow.org/guide/keras/train_and_evaluate#sample_weights

Describe the solution you'd like
from TF: 'A "sample weights" array is an array of numbers that specify how much weight each sample in a batch should have in computing the total loss. It is commonly used in imbalanced classification problems (the idea being to give more weight to rarely-seen classes).'

If this feature is already implemented in, I would appreciate a simple example.

Thanks

[benfred]

This should be implemented with #850. @marcromeyn can you work with @bschifferer on an example?

[marcromeyn]

We added a map function to the KerasSequenceLoader. This can be used to pass the sample-weight to the model, an example of this can be found in this test. Let us know if you run into any issues!