python – vectorize phrases in iOS with mlmodel from sklearn mannequin?

I made my pure language course of mannequin from python code with sklearn.
Then I transformed the mannequin to mlmodel file.
It is about sentiment evaluation, and predicting code is like beneath:

def sentiment_predictor(enter):
    enter = text_transformation(enter)
    transformed_input = cv.rework(enter)
    prediction = rfc.predict(transformed_input)
    expression_check(prediction)

import coremltools
from sklearn import tree

coreml_model = coremltools.converters.sklearn.convert(rfc,'Sentence','integer')
coreml_model.save('sentiment.mlmodel')

I made mlmodel file from rfc in above code, which is random forest classifier in sklearn.

In python code, I used CountVectorizer ‘s rework to vectorize textual content to make mannequin enter. Additionally, mannequin is skilled with CountVectorizer‘s rework methodology.

So after I use the mannequin in iOS code, rework is required from enter textual content.
How can I vectorize enter textual content in iOS code?