app.py

import flask
from flask import request
import os
from dotenv import load_dotenv
load_dotenv()

app = flask.Flask(__name__, template_folder="./")

from transformers import pipeline

classifier = pipeline('text-classification', model="bsenst/classify_services_model")

@app.route('/')
def index():
    return flask.render_template('index.html')

@app.route("/", methods=["POST"])
def predict():
    incoming = request.get_json()
    print(incoming)
    prediction = classifier(incoming["text"])[0]
    print(prediction)
    return prediction

if __name__ == '__main__':
    app.run(host='0.0.0.0',  port=int(os.environ.get('PORT', 7860)))