update docs for EntityRecognizer.predict

return type was wrongly written as a tuple, changed to syntax.StateClass
This commit is contained in:
Nikhil Saldanha 2020-03-28 18:13:02 +01:00
parent e53232533b
commit d1ddfa1cb7

View File

@ -105,7 +105,7 @@ Apply the pipeline's model to a batch of docs, without modifying them.
| Name | Type | Description |
| ----------- | -------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `docs` | iterable | The documents to predict. |
| **RETURNS** | tuple | A `(scores, tensors)` tuple where `scores` is the model's prediction for each document and `tensors` is the token representations used to predict the scores. Each tensor is an array with one row for each token in the document. |
| **RETURNS** | list | List of `syntax.StateClass` objects. `syntax.StateClass` is a helper class for the parse state (internal). |
## EntityRecognizer.set_annotations {#set_annotations tag="method"}