interfaces/illframes.py

import gradio as gr

import os
import torch
import numpy as np
import pandas as pd
from transformers import AutoModelForSequenceClassification
from transformers import AutoTokenizer
from huggingface_hub import HfApi

from label_dicts import ILLFRAMES_MIGRATION_LABEL_NAMES, ILLFRAMES_COVID_LABEL_NAMES, ILLFRAMES_WAR_LABEL_NAMES

from .utils import is_disk_full, release_model

HF_TOKEN = os.environ["hf_read"]

languages = [
    "English"
]

domains = {
    "Covid": "covid",
    "Migration": "migration",
    "War": "war"
}


# --- DEBUG ---
import shutil

def convert_size(size):
    for unit in ['B', 'KB', 'MB', 'GB', 'TB', 'PB']:
        if size < 1024:
            return f"{size:.2f} {unit}"
        size /= 1024

def get_disk_space(path="/"):
    total, used, free = shutil.disk_usage(path)
    
    return {
        "Total": convert_size(total),
        "Used": convert_size(used),
        "Free": convert_size(free)
    }

# ---

def check_huggingface_path(checkpoint_path: str):
    try:
        hf_api = HfApi(token=HF_TOKEN)
        hf_api.model_info(checkpoint_path, token=HF_TOKEN)
        return True
    except:
        return False

def build_huggingface_path(domain: str):
    return f"poltextlab/xlm-roberta-large-english-ILLFRAMES-{domain}"

def predict(text, model_id, tokenizer_id, label_names):
    device = torch.device("cpu")

    # Load JIT-traced model
    jit_model_path = f"/data/jit_models/{model_id.replace('/', '_')}.pt"
    model = torch.jit.load(jit_model_path).to(device)
    model.eval()

    # Load tokenizer (still regular HF)
    tokenizer = AutoTokenizer.from_pretrained(tokenizer_id)

    # Tokenize input
    inputs = tokenizer(
        text,
        max_length=256,
        truncation=True,
        padding="do_not_pad",
        return_tensors="pt"
    )
    inputs = {k: v.to(device) for k, v in inputs.items()}

    with torch.no_grad():
        output = model(inputs["input_ids"], inputs["attention_mask"])
        print(output) # debug
        logits = output["logits"]
        
    release_model(model, model_id)

    probs = torch.nn.functional.softmax(logits, dim=1).cpu().numpy().flatten()

    NUMS_DICT = {i: key for i, key in enumerate(sorted(label_names.keys()))}

    output_pred = {f"[{NUMS_DICT[i]}] {label_names[NUMS_DICT[i]]}": probs[i] for i in np.argsort(probs)[::-1]}
    output_info = f'<p style="text-align: center; display: block">Prediction was made using the <a href="https://huggingface.co/{model_id}">{model_id}</a> model.</p>'
    return output_pred, output_info

def predict_illframes(text, language, domain):   
    domain = domains[domain]
    model_id = build_huggingface_path(domain)
    tokenizer_id = "xlm-roberta-large"

    if domain == "migration":
        label_names = ILLFRAMES_MIGRATION_LABEL_NAMES
    elif domain == "covid":
        label_names = ILLFRAMES_COVID_LABEL_NAMES
    elif domain == "war":
        label_names = ILLFRAMES_WAR_LABEL_NAMES

    if is_disk_full():
        os.system('rm -rf /data/models*')
        os.system('rm -r ~/.cache/huggingface/hub')

    return predict(text, model_id, tokenizer_id, label_names)

demo = gr.Interface(
    title="ILLFRAMES Babel Demo",
    fn=predict_illframes,
    inputs=[gr.Textbox(lines=6, label="Input"),
            gr.Dropdown(languages, label="Language", value=languages[0]),
            gr.Dropdown(domains.keys(), label="Domain", value=list(domains.keys())[0])],
    outputs=[gr.Label(num_top_classes=5, label="Output"), gr.Markdown()])