Quick Start


Load & Predict 1 liner

The johnsnowlabs library provides 2 simple methods with which most NLP tasks can be solved while achieving state-of-the-art results.
The load and predict method.

when building a load&predict based model you will follow these steps:

  1. Pick a model/pipeline/component you want to create from the Namespace
  2. Call the model = nlp.load(component) method which will return an auto-completed pipeline
  3. Call model.predict('that was easy') on some String input

These 3 steps can be boiled down to just 1 line

from johnsnowlabs import nlp
nlp.load('sentiment').predict('How does this witchcraft work?')

jsl.load() defines 18 components types usable in 1-liners, some can be prefixed with .train for training models

Any of the actions for the component types can be passed as a string to nlp.load() and will return you the default model for that component type for the English language. You can further specify your model selection by placing a ‘.’ behind your component selection.
After the ‘.’ you can specify the model you want via specifying a dataset or model version.
See the Models Hub, the Components Namespace and The load function for more infos.

Component type nlp.load() base
Named Entity Recognition(NER) nlp.load('ner')
Part of Speech (POS) nlp.load('pos')
Classifiers nlp.load('classify')
Word embeddings nlp.load('embed')
Sentence embeddings nlp.load('embed_sentence')
Chunk embeddings nlp.load('embed_chunk')
Labeled dependency parsers nlp.load('dep')
Unlabeled dependency parsers nlp.load('dep.untyped')
Legitimatizes nlp.load('lemma')
Matchers nlp.load('match')
Normalizers nlp.load('norm')
Sentence detectors nlp.load('sentence_detector')
Chunkers nlp.load('chunk')
Spell checkers nlp.load('spell')
Stemmers nlp.load('stem')
Stopwords cleaners nlp.load('stopwords')
Cleaner nlp.load('clean')
N-Grams nlp.load('ngram')
Tokenizers nlp.load('tokenize')

Annotator & PretrainedPipeline based pipelines

You can create Annotator & PretrainedPipeline based pipelines using all the classes attached to the nlp module.

nlp.PretrainedPipeline('pipe_name') gives access to Pretrained Pipelines

from johnsnowlabs import nlp
from pprint import pprint

explain_document_pipeline = nlp.PretrainedPipeline("explain_document_ml")
annotations = explain_document_pipeline.annotate("We are very happy about SparkNLP")

  'stem': ['we', 'ar', 'veri', 'happi', 'about', 'sparknlp'],
  'checked': ['We', 'are', 'very', 'happy', 'about', 'SparkNLP'],
  'lemma': ['We', 'be', 'very', 'happy', 'about', 'SparkNLP'],
  'document': ['We are very happy about SparkNLP'],
  'pos': ['PRP', 'VBP', 'RB', 'JJ', 'IN', 'NNP'],
  'token': ['We', 'are', 'very', 'happy', 'about', 'SparkNLP'],
  'sentence': ['We are very happy about SparkNLP']

Custom Pipes

Alternatively you can compose Annotators into a pipeline which offers the highest degree of customization

from johnsnowlabs import nlp
spark = nlp.start(nlp=False)
pipe = nlp.Pipeline(stages=
spark_df = spark.createDataFrame([['Hello NLP World']]).toDF("text")
Last updated