""" Quick verification for Component 5 training pipeline. Runs a tiny 5-step training smoke test. """ from __future__ import annotations import sys from pathlib import Path # Ensure script imports work from project root. PROJECT_ROOT = Path(__file__).resolve().parents[1] if str(PROJECT_ROOT) not in sys.path: sys.path.insert(0, str(PROJECT_ROOT)) from scripts.train_component5 import main as train_main # noqa: E402 if __name__ == "__main__": try: # Override CLI args for a tiny smoke test. sys.argv = ["verify_component5_training_pipeline.py", "--config", "configs/component5_training_config.yaml"] # Patch config on disk for very short run. import yaml cfg_path = Path("configs/component5_training_config.yaml") cfg = yaml.safe_load(cfg_path.read_text(encoding="utf-8")) cfg["training"]["max_steps"] = 5 cfg["training"]["save_every"] = 5 cfg["training"]["eval_every"] = 5 cfg["training"]["log_every"] = 1 cfg["resume"]["resume_from"] = "none" tmp_cfg = Path("configs/component5_training_config.verify.yaml") tmp_cfg.write_text(yaml.safe_dump(cfg, sort_keys=False), encoding="utf-8") sys.argv = ["verify_component5_training_pipeline.py", "--config", str(tmp_cfg)] train_main() print("") print("Component 5 verification passed.") except Exception as exc: print("Component 5 verification failed.") print(f"What went wrong: {exc}") print("Fix suggestion: ensure CUDA is available and tokenized dataset path is correct.") raise SystemExit(1)