dataset for the result on paper about squeezedet HOT 3 CLOSED

@ByoenghakYim As described in our paper, that result is on a validation dataset that is randomly selected from KITTI's training dataset.

from squeezedet.

@BichenWuUCB seems result on validation and official Test benchmark differs greatly, I randomly split the training set and run the model(squeezedet), after some tune I got close to paper result, which is

but same model with same hyperparams tested on official Test Benchmark only got:

but the result reported on table compares with methods test on official test test, can you explain where is missed here? your feedback is really expected, thank you!

from squeezedet.

@twangnh can you please clarify what exactly what your tuning? I mean did you use different parameters than the one mentioned in the paper? Did you train longer? Something else?

from squeezedet.