app.py

# Zero-Shot Text Classification with Multilingual T5 (mT5)

import streamlit as st
import plotly.graph_objects as go
from mT5Model import runModel

text_1 = """Bilim insanları Botsvana’da Covid-19’un şu ana kadar en çok mutasyona uğramış varyantını tespit etti. \
Resmi olarak B.1.1.529 koduyla bilinen bu varyantı ise “Nu varyantı” adı verildi. Uzmanlar bu varyant içerisinde \
tam 32 farklı mutasyon tespit edildiğini açıklarken, bu virüsün corona virüsü aşılarına karşı daha dirençli olabileceğini duyurdu."""

text_2 = """Argentina beat Australia 2-1 on Saturday and will take on the Netherlands in the World Cup quarterfinals. \
It was a historic night for Lionel Messi as the Argentine superstar took to the pitch for his 1,000th match for club and country. \
He also scored in the match. Messi scored the opening goal in the 35th minute as his low shot in the box beat Australian goalkeeper Mathew Ryan."""

@st.cache(allow_output_mutation=True)
def list2text(label_list):
    labels = ""
    for label in label_list:
        labels = labels + label + ","
    labels = labels[:-1]
    return labels

label_list_1 = ["dünya", "ekonomi", "kültür", "sağlık", "siyaset", "spor", "teknoloji"]
label_list_2 = ["positive", "negative", "neutral"]

hypothesis_1 = "Bu yazı {} konusundadır"
hypothesis_2 = "This text is in {} subject"

st.title("Multilingual Zero-Shot Text Classification with mT5")

model_name = "alan-turing-institute/mt5-large-finetuned-mnli-xtreme-xnli"

st.sidebar.write("For details of used model:")
st.sidebar.write("https://huggingface.co/alan-turing-institute/mt5-large-finetuned-mnli-xtreme-xnli")

st.sidebar.write("For Xtreme XNLI Dataset:")
st.sidebar.write("https://www.tensorflow.org/datasets/catalog/xtreme_xnli")

st.subheader("Select Text, Label List and Hyphothesis")
st.text_area("Text #1", text_1, height=128)
st.text_area("Text #2", text_2, height=128)
st.write(f"Label List #1: {list2text(label_list_1)}")
st.write(f"Label List #2: {list2text(label_list_2)}")
st.write(f"Hypothesis #1: {hypothesis_1}")
st.write(f"Hypothesis #2: {hypothesis_2}")

text = st.radio("Select Text", ("Text #1", "Text #2", "New Text"))
labels = st.radio("Select Label List", ("Label List #1", "Label List #2", "New Label List"))
hypothesis = st.radio("Select Hypothesis", ("Hypothesis #1", "Hypothesis #2", "New Hypothesis"))

if text == "Text #1": sequence_to_classify = text_1
elif text == "Text #2": sequence_to_classify = text_2
elif text == "New Text":
    sequence_to_classify = st.text_area("New Text", value="", height=128)

if labels == "Label List #1": candidate_labels = label_list_1
elif labels == "Label List #2": candidate_labels = label_list_2
elif labels == "New Label List":
    candidate_labels = st.text_area("New Label List (Pls Input as comma-separated)", value="", height=16).split(",")

if hypothesis == "Hypothesis #1": hypothesis_template = hypothesis_1
elif hypothesis == "Hypothesis #2": hypothesis_template = hypothesis_2
elif labels == "New Hypothesis":
    hypothesis_template = st.text_area("Hypothesis Template for NLI (Pls use similar format of examples)", value="", height=16)
        
Run_Button = st.button("Run", key=None)
if Run_Button == True:
    with st.spinner('Model is running...'):
        output = runModel(model_name, sequence_to_classify, candidate_labels, hypothesis_template)
        output_labels = list(output.keys())
        output_scores = list(output.values())

        st.header("Result")
        fig = go.Figure([go.Bar(x=output_labels, y=output_scores)])
        st.plotly_chart(fig, use_container_width=False, sharing="streamlit")
        st.success('Done!')