generate_random_submission.py

import os
import random
import string
import pandas as pd
from datasets import load_dataset

def generate_random_phonemes():
    length = random.randint(5, 20)
    return " ".join(random.choices(string.ascii_lowercase, k=length))

def main():
    ids = []
    # Try reading IDs.txt first
    if os.path.exists("IDs.txt"):
        with open("IDs.txt") as f:
            ids = [l.strip() for l in f if l.strip()]
    
    if not ids:
        print("IDs.txt not found or empty. Fetching from HF (IqraEval/QuranMB.v2 split='test')...")
        token = os.environ.get("HF_TOKEN")
        if not token:
            print("HF_TOKEN not set in environment.")
            return

        try:
            ds = load_dataset("IqraEval/QuranMB.v2", split="test", token=token, streaming=True)
            for item in ds:
                 if "ID" in item: ids.append(str(item["ID"]).strip())
                 elif "id" in item: ids.append(str(item["id"]).strip())
            
            # Save to IDs.txt for future use
            with open("IDs.txt", "w") as f:
                for i in sorted(ids):
                    f.write(f"{i}\n")
            print(f"Cached {len(ids)} IDs to IDs.txt")
        except Exception as e:
            print(f"Error fetching IDs: {e}")
            return

    if not ids:
        print("No IDs found.")
        return

    print(f"Generating submission for {len(ids)} IDs...")
    df = pd.DataFrame({
        "ID": ids,
        "Labels": [generate_random_phonemes() for _ in ids]
    })
    df.to_csv("random_submission.csv", index=False)
    print(f"Generated random_submission.csv with {len(df)} rows.")

if __name__ == "__main__":
    main()