Evaluate performance on VOT

Hello,
I really liked you paper and i am interested in doing some experiments on the code you provided.
but I have faced a problem while doing the evaluation:

python ../../tools/eval.py --dataset VOT2016 --tracker_prefix C --result_dir ./test/VOT2016
loading VOT2016: 100%|█████████████████████████████████████| 60/60 [00:00<00:00, 2127.27it/s, wiper]
eval ar: 100%|████████████████████████████████████████████████████████| 2/2 [00:01<00:00, 1.65it/s]
eval eao: 0%| | 0/2 [00:00

Start with tools/eval.py and reproduce the VOT2016 command shown in the report. Trace the EAO evaluation worker and the multiprocessing traceback involving the missing 'loc' argument. Done means the VOT2016 evaluation completes and produces EAO results without the exception.

Written by the indexing model from the issue text.