utils.py

import datetime
import json
import random

import streamlit as st
import tokenizers
import torch
from transformers import Pipeline, pipeline


def get_answer(input, context, engine, threshold=0.5):

    answer = engine({"question": input, "context": context})

    if answer["score"] > threshold:
        return answer["answer"]
    else:
        return random.choice(
            [
                "Non so la risposta... prova con un'altra domanda!",
                "Non so rispondere a questa domanda, prova con un'altra!",
            ]
        )


@st.cache
def get_context():

    BIRTHYEAR = 1952
    BIRTHMONTH = 5
    OTHERBIRTHYEAR = 1984
    now = datetime.datetime.now()

    with open("context.json") as f:
        context = (
            json.load(f)["info"]
            .replace("[YEAR]", str(now.year))
            .replace("[TODAY]", f"{datetime.datetime.now():%d-%m-%Y}")
            .replace("[BIRTHYEAR]", str(BIRTHYEAR))
            .replace(
                "[AGE]",
                str((now - datetime.datetime(BIRTHYEAR, BIRTHMONTH, 1)).days // 365),
            )
            .replace("[OTHERAGE]", str(now.year - OTHERBIRTHYEAR))
        )

    return context


@st.cache(
    hash_funcs={
        torch.nn.parameter.Parameter: lambda _: None,
        tokenizers.Tokenizer: lambda _: None,
        tokenizers.AddedToken: lambda _: None,
    },
    allow_output_mutation=True,
    show_spinner=False,
)
def load_engine() -> Pipeline:

    nlp_qa = pipeline(
        "question-answering",
        model="mrm8488/bert-italian-finedtuned-squadv1-it-alfa",
        tokenizer="mrm8488/bert-italian-finedtuned-squadv1-it-alfa",
    )

    return nlp_qa