1. Download the MVTEC-LOCO-AD dataset from the MVTEC website. Then organize the MVTEC-LOCO-AD dataset into MVTEC-AD format. There may be multiple GTs for one image in the MVTEC-LOCO-AD dataset, we only need to take any one of them.
2. Change self.mvtec_folder_path in src/datasets/mvtec.py to your path of MVTEC LOCO AD dataset (In MVTEC AD format).
3. Change the --dateset_base_dir in src/O_evaluation/evaluate_experiment.py -> def parse_arguments() to your own MVTEC LOCO AD dataset path (the original LOCO dataset downloaded, not in MVTEC AD format).
   Change --anomaly_maps_dir to your own path (log_metris folder is already included in this program!) .
4. Run main.py. (The evaluation code was taken from this website.)

• There is generally no randomness in SPADE.
• The results of this project have a small difference in ROCAUC at the image-level from the original paper, but a large difference at the pixel-level.
• I don't know where is wrong, if you have any idea please leave a comment to discuss.

	Pixel-SPRO-AUC (Paper)	Pixel-SPRO-AUC (This Code)	Image-ROC-AUC (Paper)	Image-ROC-AUC (This Code)
Breakfast Box	0.372	0.143	-	0.768
Screw Bag	0.331	0.421	-	0.532
Pushpins	0.234	0.251	-	0.569
Splicing Connectors	0.516	0.598	-	0.778
Juice Bottle	0.804	0.587	-	0.88
Mean	0.451	0.4	0.689	0.7054

Reference:https://github.com/byungjae89/SPADE-pytorch