app.py

import time
import torch

import psutil
import streamlit as st

from generator import GeneratorFactory
from langdetect import detect
from default_texts import default_texts

device = torch.cuda.device_count() - 1

TRANSLATION_EN_TO_NL = "translation_en_to_nl"
TRANSLATION_NL_TO_EN = "translation_nl_to_en"

GENERATOR_LIST = [
    {
        "model_name": "yhavinga/t5-small-24L-ccmatrix-multi",
        "desc": "T5 small nl24 ccmatrix en->nl",
        "task": TRANSLATION_EN_TO_NL,
        "split_sentences": True,
    },
    {
        "model_name": "yhavinga/t5-small-24L-ccmatrix-multi",
        "desc": "T5 small nl24 ccmatrix nl-en",
        "task": TRANSLATION_NL_TO_EN,
        "split_sentences": True,
    },
    {
        "model_name": "Helsinki-NLP/opus-mt-en-nl",
        "desc": "Opus MT en->nl",
        "task": TRANSLATION_EN_TO_NL,
        "split_sentences": True,
    },
    # {
    #     "model_name": "yhavinga/longt5-local-eff-large-nl8-voc8k-ddwn-512beta-512l-nedd-256ccmatrix-en-nl",
    #     "desc": "longT5 large nl8 256cc/512beta/512l en->nl",
    #     "task": TRANSLATION_EN_TO_NL,
    #     "split_sentences": False,
    # },
    # {
    #     "model_name": "yhavinga/byt5-small-ccmatrix-en-nl",
    #     "desc": "ByT5 small ccmatrix en->nl",
    #     "task": TRANSLATION_EN_TO_NL,
    #     "split_sentences": True,
    # },
    # {
    #     "model_name": "yhavinga/t5-eff-large-8l-nedd-en-nl",
    #     "desc": "T5 eff large nl8 en->nl",
    #     "task": TRANSLATION_EN_TO_NL,
    #     "split_sentences": True,
    # },
    # {
    #     "model_name": "yhavinga/t5-base-36L-ccmatrix-multi",
    #     "desc": "T5 base nl36 ccmatrix en->nl",
    #     "task": TRANSLATION_EN_TO_NL,
    #     "split_sentences": True,
    # },
    # {
    #     "model_name": "yhavinga/longt5-local-eff-large-nl8-voc8k-ddwn-512beta-512-nedd-en-nl",
    #     "desc": "longT5 large nl8 512beta/512l en->nl",
    #     "task": TRANSLATION_EN_TO_NL,
    #     "split_sentences": False,
    # },
    # {
    #     "model_name": "yhavinga/t5-base-36L-nedd-x-en-nl-300",
    #     "desc": "T5 base 36L nedd en->nl 300",
    #     "task": TRANSLATION_EN_TO_NL,
    #     "split_sentences": True,
    # },
    # {
    #     "model_name": "yhavinga/long-t5-local-small-ccmatrix-en-nl",
    #     "desc": "longT5 small ccmatrix en->nl",
    #     "task": TRANSLATION_EN_TO_NL,
    #     "split_sentences": True,
    # },
]


def main():
    st.set_page_config(  # Alternate names: setup_page, page, layout
        page_title="Babel",  # String or None. Strings get appended with "• Streamlit".
        layout="wide",  # Can be "centered" or "wide". In the future also "dashboard", etc.
        initial_sidebar_state="expanded",  # Can be "auto", "expanded", "collapsed"
        page_icon="📚",  # String, anything supported by st.image, or None.
    )

    if "generators" not in st.session_state:
        st.session_state["generators"] = GeneratorFactory(GENERATOR_LIST)
    generators = st.session_state["generators"]

    with open("style.css") as f:
        st.markdown(f"<style>{f.read()}</style>", unsafe_allow_html=True)
    st.sidebar.image("babel.png", width=200)
    st.sidebar.markdown(
        """# Babel
    Vertaal van en naar Engels"""
    )
    st.sidebar.title("Parameters:")

    default_text = st.sidebar.radio(
        "Change default text",
        tuple(default_texts.keys()),
        index=0,
    )

    if default_text or "prompt_box" not in st.session_state:
        st.session_state["prompt_box"] = default_texts[default_text]["text"]

    text_area = st.text_area("Enter text", st.session_state.prompt_box, height=300)
    st.session_state["text"] = text_area
    num_beams = st.sidebar.number_input("Num beams", min_value=1, max_value=10, value=1)
    num_beam_groups = st.sidebar.number_input(
        "Num beam groups", min_value=1, max_value=10, value=1
    )
    length_penalty = st.sidebar.number_input(
        "Length penalty", min_value=0.0, max_value=2.0, value=1.2, step=0.1
    )

    st.sidebar.markdown(
        """For an explanation of the parameters, head over to the [Huggingface blog post about text generation](https://huggingface.co/blog/how-to-generate)
and the [Huggingface text generation interface doc](https://huggingface.co/transformers/main_classes/model.html?highlight=generate#transformers.generation_utils.GenerationMixin.generate).
"""
    )

    params = {
        "num_beams": num_beams,
        "num_beam_groups": num_beam_groups,
        "length_penalty": length_penalty,
        "early_stopping": True,
    }

    if st.button("Run"):
        memory = psutil.virtual_memory()

        language = detect(st.session_state.text)
        if language == "en":
            task = TRANSLATION_EN_TO_NL
        elif language == "nl":
            task = TRANSLATION_NL_TO_EN
        else:
            st.error(f"Language {language} not supported")
            return

        # Num beam groups should be a divisor of num beams
        if num_beams % num_beam_groups != 0:
            st.error("Num beams should be a multiple of num beam groups")
            return

        for generator in generators.filter(task=task):
            st.markdown(f"🧮 **Model `{generator}`**")
            time_start = time.time()
            result, params_used = generator.generate(
                text=st.session_state.text, **params
            )
            time_end = time.time()
            time_diff = time_end - time_start

            st.write(result.replace("\n", "  \n"))
            text_line = ", ".join([f"{k}={v}" for k, v in params_used.items()])
            st.markdown(f"    🕙 *generated in {time_diff:.2f}s, `{text_line}`*")

        st.write(
            f"""
        ---
        *Memory: {memory.total / 10**9:.2f}GB, used: {memory.percent}%, available: {memory.available / 10**9:.2f}GB*
        """
        )


if __name__ == "__main__":
    main()