app.py

import gradio as gr 
from transformers import AutoTokenizer, AutoModel
import torch

tokenizer = AutoTokenizer.from_pretrained('intfloat/multilingual-e5-large')
model = AutoModel.from_pretrained('intfloat/multilingual-e5-large')

def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0]
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)

def encode_sentences(sentences):
    encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
    with torch.no_grad():
        model_output = model(**encoded_input)
    sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
    return sentence_embeddings.tolist()

demo = gr.Interface(fn=encode_sentences,
                    inputs="textbox",
                    outputs="text")

if __name__ == "__main__":
    demo.launch()