scripts/prepare_data.py

"""CLI: download FiNER-139, align labels, split into train/val/test.

Usage:
    python scripts/prepare_data.py
    python scripts/prepare_data.py --seed 123
"""

from __future__ import annotations

import argparse
import logging
import sys
from pathlib import Path

sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "src"))

from rich.console import Console
from rich.table import Table
from collections import Counter

from finner.data.load import load_all
from finner.data.split import split_and_save
from finner.labels import ENTITY_TYPES

logging.basicConfig(level=logging.INFO, format="%(levelname)s %(message)s")
console = Console()


def label_distribution(examples: list[dict]) -> dict:
    counts = Counter()
    for ex in examples:
        for tag in ex["ner_tags"]:
            if tag != "O":
                counts[tag.split("-", 1)[1]] += 1
    return dict(counts)


def main():
    parser = argparse.ArgumentParser(description="Prepare FinNER training data")
    parser.add_argument("--seed", type=int, default=42)
    args = parser.parse_args()

    console.print("[bold cyan]Step 1: Loading FiNER-139...[/bold cyan]")
    examples = load_all()
    console.print(f"Loaded [bold]{len(examples)}[/bold] examples total.")

    console.print("\n[bold cyan]Step 2: Splitting (stratified, 70/15/15)...[/bold cyan]")
    train, val, test = split_and_save(examples, seed=args.seed)

    # Print distribution stats
    table = Table(title="Label distribution by split", show_header=True)
    table.add_column("Entity Type", style="cyan")
    table.add_column("Train", justify="right")
    table.add_column("Val", justify="right")
    table.add_column("Test", justify="right")

    train_dist = label_distribution(train)
    val_dist = label_distribution(val)
    test_dist = label_distribution(test)

    for etype in ENTITY_TYPES:
        table.add_row(
            etype,
            str(train_dist.get(etype, 0)),
            str(val_dist.get(etype, 0)),
            str(test_dist.get(etype, 0)),
        )
    console.print(table)

    console.print(f"\n[bold green]✓ Done.[/bold green] Splits saved to data/splits/")
    console.print(f"  train: {len(train):,}  val: {len(val):,}  test: {len(test):,}")
    console.print("\n[yellow]Note: O-class dominance is expected and will be addressed in Phase 3.[/yellow]")


if __name__ == "__main__":
    main()