classifier.py

from transformers import pipeline

umbrellaSubjects = [
    "Science, Technology, Engineering, Mathematics",
    "Philosophy",
    "Arts and Humanities",
    "Social Sciences",
    "Languages",
    "Professional Studies"
]
classifier =  pipeline("zero-shot-classification", model="vicgalle/xlm-roberta-large-xnli-anli")

def process_string(string):
    # Remove whitespace from string
    string = string.replace(' ', '')
    # Split string by comma
    string_list = string.split(',')
    # Return list with no whitespace
    return [term.strip() for term in string_list]
def classify(sentences,categories):
    return classifier(process_string(sentences),process_string(categories))