"""Test prompt variations for YOLO-World zero-shot detection to maximize recall on micro-components and desoldered pads."""
import os
from pathlib import Path
import torch
from torchvision.ops import nms
import numpy as np
from PIL import Image

cache_dir = Path("data/cache").resolve()
cache_dir.mkdir(parents=True, exist_ok=True)
os.environ["TORCH_HOME"] = str(cache_dir)
os.environ["HF_HOME"] = str(cache_dir)

from ultralytics import YOLOWorld

CLASS_NAMES = [
    "cpu_socket",
    "gold_finger",
    "ic_chip",
    "tantalum_capacitor",
    "forcibly_desoldered_pad",
    "factory_unpopulated_pad",
    "occluded_region",
]

# Alternative prompt sets
PROMPT_SETS = {
    "Original": [
        "CPU socket",
        "gold finger edge connector",
        "microchip",
        "tantalum capacitor",
        "desoldered solder pad",
        "unpopulated solder pad",
        "heatsink shield",
    ],
    "Enriched_Visual": [
        "CPU socket square slot",
        "gold edge connector pins",
        "black IC microchip",
        "yellow SMD capacitor",
        "empty desoldered solder footprint",
        "clean unpopulated solder pad",
        "metal heatsink shield cover",
    ],
    "Short_Direct": [
        "processor socket",
        "gold edge contacts",
        "IC chip",
        "tantalum capacitor",
        "solder pad",
        "empty pad",
        "heatsink",
    ]
}

if __name__ == "__main__":
    images = sorted(list(Path("data/images").glob("*.jpg")))[:50]
    
    for set_name, prompts in PROMPT_SETS.items():
        print(f"\n--- Testing Prompt Set: {set_name} ---")
        model = YOLOWorld("yolov8s-worldv2.pt")
        model.set_classes(prompts)

        counts = {name: 0 for name in CLASS_NAMES}
        total_boxes = 0

        for img_path in images:
            results = model.predict(str(img_path), conf=0.015, verbose=False)
            if results and len(results) > 0 and results[0].boxes is not None:
                cls_ids = results[0].boxes.cls.cpu().numpy().astype(int)
                for cls_id in cls_ids:
                    if 0 <= cls_id < len(CLASS_NAMES):
                        counts[CLASS_NAMES[cls_id]] += 1
                        total_boxes += 1

        print(f"Total boxes ({set_name}): {total_boxes}")
        for k, v in counts.items():
            print(f"  {k:25s}: {v}")