Added Multilingual Model support

app.py

@@ -1,9 +1,10 @@
 import streamlit as st
-from hf_model import classifier_zero,load_model
+from hf_model import classifier_zero,load_model,load_model_multil
 from utils import plot_result,examples_load
 import json
 
-classifier=load_model()
+classifier_model0=load_model()
+classifier_model1=load_model_multil()
 ex_text,ex_labels=examples_load()
 
 # with open("examples.json") as f:
@@ -13,12 +14,15 @@ if __name__ == '__main__':
     st.header("Zero Shot Classification")
     st.write("This app allows you to classify any text into any categories you are interested in.")
 
+
     with st.form(key='my_form'):
         text_input = st.text_area("Input Text",ex_text)
         labels = st.text_input('Possible topics (separated by `,`)',ex_labels, max_chars=1000)
         labels = list(set([x.strip() for x in labels.strip().split(',') if len(x.strip()) > 0]))
         radio = st.radio("Select Multiclass",('Only one topic can be corect at a time','Multiple topics can be correct at a time'),)
+        radio1 = st.radio("Select Model",('Using English Language model','Using Multilingual Model'),)
         multi_class= True if radio=="Multiple topics can be correct at a time" else False
+        classifier = classifier_model1 if radio1=="Using Multilingual Model" else classifier_model0
         submit_button = st.form_submit_button(label='Submit')
 
     if submit_button:

hf_model.py

@@ -2,13 +2,19 @@ from transformers import AutoTokenizer, AutoModelForSequenceClassification,pipel
 import torch
 
 def load_model():
-
     model_name = "MoritzLaurer/DeBERTa-v3-base-mnli"
     tokenizer = AutoTokenizer.from_pretrained(model_name)
     model = AutoModelForSequenceClassification.from_pretrained(model_name)
     classifier = pipeline(task='zero-shot-classification', model=model, tokenizer=tokenizer, framework='pt')
     return classifier
 
+def load_model_multil():
+    model_name = "joeddav/xlm-roberta-large-xnli"
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    model = AutoModelForSequenceClassification.from_pretrained(model_name)
+    classifier = pipeline(task='zero-shot-classification', model=model, tokenizer=tokenizer, framework='pt')
+    return classifier
+
 def classifier_zero(classifier,sequence:str,labels:list,multi_class:bool):
     outputs=classifier(sequence, labels,multi_label=multi_class)
     return outputs['labels'], outputs['scores']