eval and test use the same dataset?

[ksmdnl]

I'm still trying to figure it if my assumption that the same dataset for evaluation and testing is being used. Did I miss anything there somewhere? Head ups will be appreciated!

[smallsun1122]

I now have similar doubts, the training process seems to use the 2007VOC and 2012VOC, why the eval.py used in the 2007VOC?

[ksmdnl]

I suppose they were benchmarking or trying to reproduce the original implementation on caffe, which trained the model on trainval set and evaluated it on the test set. I did train my model on both combinations and got, as expected, slightly different performance.