from apps import mlm, vqa, article import streamlit as st from session import _get_state from multiapp import MultiApp from apps.utils import read_markdown def main(): state = _get_state() st.set_page_config( page_title="Multilingual VQA", layout="wide", initial_sidebar_state="collapsed", page_icon="./misc/mvqa-logo-3-white.png", ) st.title("Multilingual Visual Question Answering") st.write( "[Gunjan Chhablani](https://huggingface.co/gchhablani), [Bhavitvya Malik](https://huggingface.co/bhavitvyamalik)" ) st.sidebar.title("Multilingual VQA") logo = st.sidebar.image("./misc/mvqa-logo-3-white.png") st.sidebar.write("Multilingual VQA addresses the challenge of visual question answering in a multilingual setting. Here, we fuse CLIP Vision transformer into BERT and perform pre-training and fine-tuning on translated versions of Conceptual-12M and VQAv2 datasets. Please use the radio buttons below to navigate.") app = MultiApp(state) app.add_app("Article", article.app) app.add_app("Visual Question Answering", vqa.app) app.add_app("Mask Filling", mlm.app) app.add_app("Examples", mlm.app) app.run() state.sync() if __name__ == "__main__": main()