inference.py

"""
Assign cluster labels to keywords (CSV or list).
"""
import argparse
from pathlib import Path

import pandas as pd
import joblib

from config import MODEL_DIR, KEYWORD_COLUMN


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("--input", default="data/keywords.csv", help="CSV with keyword column")
    parser.add_argument("--output", default="data/clustered.csv", help="Output CSV with cluster column")
    args = parser.parse_args()

    if not Path(args.input).exists():
        raise FileNotFoundError(f"Input file not found: {args.input}. Run from project root or use an absolute path.")
    if not (MODEL_DIR / "encoder.joblib").exists():
        raise FileNotFoundError(f"Run train.py first. No model in {MODEL_DIR}")

    encoder = joblib.load(MODEL_DIR / "encoder.joblib")
    clusterer = joblib.load(MODEL_DIR / "clusterer.joblib")

    df = pd.read_csv(args.input)
    if KEYWORD_COLUMN not in df.columns:
        raise ValueError(f"CSV must have column: {KEYWORD_COLUMN}")

    X = encoder.encode(df[KEYWORD_COLUMN].astype(str).tolist())
    df["cluster"] = clusterer.predict(X)

    Path(args.output).parent.mkdir(parents=True, exist_ok=True)
    df.to_csv(args.output, index=False)
    print(f"Saved to {args.output}")


if __name__ == "__main__":
    main()