mirror of https://github.com/explosion/spaCy.git
62 lines
3.2 KiB
Plaintext
62 lines
3.2 KiB
Plaintext
|
//- 💫 DOCS > USAGE > TRAINING > NER
|
||
|
|
|
||
|
|
p
|
||
|
|
| All #[+a("/models") spaCy models] support online learning, so
|
||
|
|
| you can update a pre-trained model with new examples. To update the
|
||
|
|
| model, you first need to create an instance of
|
||
|
|
| #[+api("goldparse") #[code GoldParse]], with the entity labels
|
||
|
|
| you want to learn. You'll usually need to provide many examples to
|
||
|
|
| meaningfully improve the system — a few hundred is a good start, although
|
||
|
|
| more is better.
|
||
|
|
|
||
|
|
p
|
||
|
|
| You should avoid iterating over the same few examples multiple times, or
|
||
|
|
| the model is likely to "forget" how to annotate other examples. If you
|
||
|
|
| iterate over the same few examples, you're effectively changing the loss
|
||
|
|
| function. The optimizer will find a way to minimize the loss on your
|
||
|
|
| examples, without regard for the consequences on the examples it's no
|
||
|
|
| longer paying attention to. One way to avoid this
|
||
|
|
| #[+a("https://explosion.ai/blog/pseudo-rehearsal-catastrophic-forgetting", true) "catastrophic forgetting" problem]
|
||
|
|
| is to "remind"
|
||
|
|
| the model of other examples by augmenting your annotations with sentences
|
||
|
|
| annotated with entities automatically recognised by the original model.
|
||
|
|
| Ultimately, this is an empirical process: you'll need to
|
||
|
|
| #[strong experiment on your own data] to find a solution that works best
|
||
|
|
| for you.
|
||
|
|
|
||
|
|
+h(3, "example-new-entity-type") Example: Training an additional entity type
|
||
|
|
|
||
|
|
p
|
||
|
|
| This script shows how to add a new entity type to an existing pre-trained
|
||
|
|
| NER model. To keep the example short and simple, only a few sentences are
|
||
|
|
| provided as examples. In practice, you'll need many more — a few hundred
|
||
|
|
| would be a good start. You will also likely need to mix in examples of
|
||
|
|
| other entity types, which might be obtained by running the entity
|
||
|
|
| recognizer over unlabelled sentences, and adding their annotations to the
|
||
|
|
| training set.
|
||
|
|
|
||
|
|
p
|
||
|
|
| The actual training is performed by looping over the examples, and
|
||
|
|
| calling #[+api("language#update") #[code nlp.update()]]. The
|
||
|
|
| #[code update] method steps through the words of the input. At each word,
|
||
|
|
| it makes a prediction. It then consults the annotations provided on the
|
||
|
|
| #[+api("goldparse") #[code GoldParse]] instance, to see whether it was
|
||
|
|
| right. If it was wrong, it adjusts its weights so that the correct
|
||
|
|
| action will score higher next time.
|
||
|
|
|
||
|
|
+github("spacy", "examples/training/train_new_entity_type.py")
|
||
|
|
|
||
|
|
+h(3, "example-ner-from-scratch") Example: Training an NER system from scratch
|
||
|
|
|
||
|
|
p
|
||
|
|
| This example is written to be self-contained and reasonably transparent.
|
||
|
|
| To achieve that, it duplicates some of spaCy's internal functionality.
|
||
|
|
| Specifically, in this example, we don't use spaCy's built-in
|
||
|
|
| #[+api("language") #[code Language]] class to wire together the
|
||
|
|
| #[+api("vocab") #[code Vocab]], #[+api("tokenizer") #[code Tokenizer]]
|
||
|
|
| and #[+api("entityrecognizer") #[code EntityRecognizer]]. Instead, we
|
||
|
|
| write our own simle #[code Pipeline] class, so that it's easier to see
|
||
|
|
| how the pieces interact.
|
||
|
|
|
||
|
|
+github("spacy", "examples/training/train_ner_standalone.py")
|