Example
import spacy

from sklearn.pipeline import make_pipeline
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.linear_model import LogisticRegression

from tokenwiser.component import attach_sklearn_categoriser

X = [
    'i really like this post',
    'thanks for that comment',
    'i enjoy this friendly forum',
    'this is a bad post',
    'i dislike this article',
    'this is not well written'
]

y = ['pos', 'pos', 'pos', 'neg', 'neg', 'neg']

# Note that we're training a pipeline here via a single-batch `.fit()` method
pipe = make_pipeline(CountVectorizer(), LogisticRegression()).fit(X, y)

nlp = spacy.load('en_core_web_sm')
# This is where we attach our pre-trained model as a pipeline step.
attach_sklearn_categoriser(nlp, pipe_name='silly_sentiment', estimator=pipe)

Categories pipeline training