becomes part of `outputs[partitions[js]]`.  The slices with `partitions[js] = i`
are placed in `outputs[i]` in lexicographic order of `js`, and the first
dimension of `outputs[i]` is the number of entries in `partitions` equal to `i`.
In detail,

```python
    outputs[i].shape = [sum(partitions == i)] + data.shape[partitions.ndim:]

    outputs[i] = pack([data[js, ...] for js if partitions[js] == i])

    may start to fail. To fix this, either update any inference clients to send
    requests with the actual input and output keys used by the trainer code, or
    conversely, update the trainer code to name the input and output Tensors
    'inputs' and 'outputs', respectively. Signatures using the 'classify' and
    'regress' methods are not affected by this change; they will continue to