File size: 2,164 Bytes
f9ed2b4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 | """Evaluate one final binary checkpoint on the fixed WildGen validation split."""
import argparse
import json
from pathlib import Path
import torch
from torch.utils.data import DataLoader
from evaluate_robustness import (BinarySplitDataset, build_model, load_checkpoint, load_test_rows, make_transform, evaluate)
def validate_rows(rows):
real = sum(r["label"] == 0 for r in rows)
fake = sum(r["label"] == 1 for r in rows)
if (real, fake) != (8000, 34200):
raise ValueError(f"Expected 8,000 Real and 34,200 AI validation images; got {real}, {fake}")
if len({r["path"] for r in rows}) != len(rows):
raise ValueError("Repeated validation image paths")
def main():
p = argparse.ArgumentParser(description=__doc__)
p.add_argument("--model-type", choices=["univfd", "cnndetect"], required=True)
p.add_argument("--checkpoint", type=Path, required=True)
p.add_argument("--split-file", type=Path, required=True)
p.add_argument("--output", type=Path, required=True)
p.add_argument("--batch-size", type=int, default=64)
p.add_argument("--num-workers", type=int, default=4)
p.add_argument("--device", default="cuda" if torch.cuda.is_available() else "cpu")
a = p.parse_args()
expected = "fc_weights_final.pth" if a.model_type == "univfd" else "model_final.pth"
if a.checkpoint.name != expected:
p.error(f"Expected {expected}")
rows = load_test_rows(a.split_file)
validate_rows(rows)
device = torch.device(a.device)
model = build_model(a.model_type, a.device).to(device)
load_checkpoint(model, a.model_type, a.checkpoint)
loader = DataLoader(BinarySplitDataset(rows, make_transform(a.model_type)), batch_size=a.batch_size, num_workers=a.num_workers, shuffle=False)
result = {"checkpoint": str(a.checkpoint), "split_file": str(a.split_file), "reporting_policy": "final_epoch", "metrics": evaluate(model, loader, device)}
a.output.parent.mkdir(parents=True, exist_ok=True)
a.output.write_text(json.dumps(result, indent=2), encoding="utf-8")
print(json.dumps(result, indent=2))
if __name__ == "__main__": main()
|