NLU
Understanding for each token in the sentence.
NLU Object
| KEY | TYPE | DESCRIPTION | CONSTRAINTS |
|---|---|---|---|
| index | int | Index in the initial tokenized sentence | index >= 0 |
| len | int | Number of tokens that have been merged (1 if no merge) | len >= 1 |
| meaning | list of Category Objects | Known meanings for the item | Can be empty. |
| source | string | - | if len is greater than one, will be the merged sources of original items |
| value | dict | Depends on tag. Most values are in the 'scalar' key of this dict |
