import streamlit as st
from transformers import pipeline
from textblob import TextBlob
from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity

model = SentenceTransformer('paraphrase-xlm-r-multilingual-v1')

sentences = \
    [
     "I have a bigger house than you", 
     "You have a bigger house than me"
     ]
     
sentence_embeddings = model.encode(sentences)

for sentence, embedding in zip(sentences, sentence_embeddings):
    print("Sentence:", sentence)
    print("Embedding:", embedding)
    print("")
    
print('Similarity between {} and {} is {}'.format(sentences[0],
       sentences[1],
       cosine_similarity(sentence_embeddings[0].reshape(1, -1),
       sentence_embeddings[1].reshape(1, -1))[0][0]))
       
st.title("Similarity text")

st.write('Similarity between {} and {} is {}'.format(sentences[0],
       sentences[1],
       cosine_similarity(sentence_embeddings[0].reshape(1, -1),
       sentence_embeddings[1].reshape(1, -1))[0][0]))