| """Compare VLM filter modes on the user's schematic (both recall-boost ON): |
| - whitelist : keep only VLM-labelled 'resistor' (high precision, low recall) |
| - reject-only: drop only confident non-target classes (high recall) |
| Saves annotated PNGs for visual comparison. |
| """ |
| import os |
| import sys |
| import time |
|
|
| import cv2 |
|
|
| sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) |
|
|
| from src.pipeline import PatternDetectionPipeline |
|
|
| DRAWING = os.environ.get("DRAWING", r"D:\Sotatek_Assessment\drawings\1.png") |
| PATTERN = r"D:\Sotatek_Assessment\drawings\test_2.png" |
| OUT = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), |
| "debug_output") |
|
|
|
|
| def run(reject_only: bool, tag: str): |
| print(f"\n########## reject_only={reject_only} (boost+VLM) ##########") |
| pipe = PatternDetectionPipeline(config={ |
| "use_vlm": True, |
| "vlm_recall_boost": True, |
| "vlm_reject_only": reject_only, |
| "vlm_symbol_name": "a resistor (zigzag or plain rectangle)", |
| }) |
| t0 = time.time() |
| result = pipe.detect_auto(PATTERN, DRAWING, return_visualization=True) |
| dt = time.time() - t0 |
| n = result["total_detections"] |
| outpath = os.path.join(OUT, f"mode_{tag}.png") |
| cv2.imwrite(outpath, result["visualization"]) |
| print(f"[CMP] {tag}: {n} detections in {dt:.1f}s -> {outpath}") |
| return n |
|
|
|
|
| def main(): |
| if not os.path.exists(DRAWING) or not os.path.exists(PATTERN): |
| print("MISSING input files"); return |
| n_white = run(False, "whitelist") |
| n_reject = run(True, "rejectonly") |
| print(f"\n========== RESULT ==========") |
| print(f" whitelist : {n_white}") |
| print(f" reject-only : {n_reject}") |
| print(f" recall delta: +{n_reject - n_white}") |
|
|
|
|
| if __name__ == "__main__": |
| main() |
|
|
