spaCy/website/usage/_training/_ner.jade

124 lines
5.8 KiB
Plaintext
Raw Normal View History

2017-10-03 15:26:20 +03:00
//- 💫 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.
2017-10-26 15:44:43 +03:00
+h(3, "example-train-ner") Updating the Named Entity Recognizer
p
| This example shows how to update spaCy's entity recognizer
| with your own examples, starting off with an existing, pre-trained
| model, or from scratch using a blank #[code Language] class. To do
| this, you'll need #[strong example texts] and the
| #[strong character offsets] and #[strong labels] of each entity contained
| in the texts.
2017-10-27 13:30:59 +03:00
+github("spacy", "examples/training/train_ner.py", 500)
2017-10-26 15:44:43 +03:00
+h(4) Step by step guide
+list("numbers")
+item
| #[strong Load the model] you want to start with, or create an
| #[strong empty model] using
| #[+api("spacy#blank") #[code spacy.blank]] with the ID of your
| language. If you're using a blank model, don't forget to add the
| entity recognizer to the pipeline. If you're using an existing model,
| make sure to disable all other pipeline components during training
| using #[+api("language#disable_pipes") #[code nlp.disable_pipes]].
| This way, you'll only be training the entity recognizer.
+item
2017-11-07 02:23:19 +03:00
| #[strong Shuffle and loop over] the examples. For each example,
| #[strong update the model] by calling
| #[+api("language#update") #[code nlp.update]], which steps
2017-10-26 15:44:43 +03:00
| through the words of the input. At each word, it makes a
2017-11-07 02:23:19 +03:00
| #[strong prediction]. It then consults the annotations to see whether
| it was right. If it was wrong, it adjusts its weights so that the
| correct action will score higher next time.
2017-10-26 15:44:43 +03:00
+item
| #[strong Save] the trained model using
| #[+api("language#to_disk") #[code nlp.to_disk]].
+item
| #[strong Test] the model to make sure the entities in the training
| data are recognised correctly.
+h(3, "example-new-entity-type") Training an additional entity type
2017-10-03 15:26:20 +03:00
p
| This script shows how to add a new entity type #[code ANIMAL] to an
| existing pre-trained NER model, or an empty #[code Language] class. To
| keep the example short and simple, only a few sentences are
2017-10-03 15:26:20 +03:00
| 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.
2017-10-27 13:30:59 +03:00
+github("spacy", "examples/training/train_new_entity_type.py", 500)
2017-10-03 15:26:20 +03:00
2017-11-07 02:23:19 +03:00
+aside("Important note", "⚠️")
| If you're using an existing model, make sure to mix in examples of
| #[strong other entity types] that spaCy correctly recognized before.
| Otherwise, your model might learn the new type, but "forget" what it
| previously knew. This is also referred to as the
| #[+a("https://explosion.ai/blog/pseudo-rehearsal-catastrophic-forgetting", true) "catastrophic forgetting" problem].
2017-10-26 15:44:43 +03:00
+h(4) Step by step guide
+list("numbers")
+item
| #[strong Load the model] you want to start with, or create an
| #[strong empty model] using
2017-10-26 15:44:43 +03:00
| #[+api("spacy#blank") #[code spacy.blank]] with the ID of your
| language. If you're using a blank model, don't forget to add the
| entity recognizer to the pipeline. If you're using an existing model,
| make sure to disable all other pipeline components during training
| using #[+api("language#disable_pipes") #[code nlp.disable_pipes]].
| This way, you'll only be training the entity recognizer.
+item
| #[strong Add the new entity label] to the entity recognizer using the
| #[+api("entityrecognizer#add_label") #[code add_label]] method. You
| can access the entity recognizer in the pipeline via
| #[code nlp.get_pipe('ner')].
+item
| #[strong Loop over] the examples and call
| #[+api("language#update") #[code nlp.update]], which steps through
| the words of the input. At each word, it makes a
2017-11-07 02:23:19 +03:00
| #[strong prediction]. It then consults the annotations, to see
| whether it was right. If it was wrong, it adjusts its weights so that
| the correct action will score higher next time.
+item
| #[strong Save] the trained model using
2017-10-26 15:44:43 +03:00
| #[+api("language#to_disk") #[code nlp.to_disk]].
+item
2017-10-26 15:44:43 +03:00
| #[strong Test] the model to make sure the new entity is recognised
| correctly.