exp end_to_end_subsample_conv_hand_steph_model maj du fichier de log tensorboard + do not recompute repr after training

Transfert learning with batchnorm expe
End to end expe

in benchmark_vgg_tmp now details have changes about time tracking and options
New parameter files
rename benchmark_vgg into benchmark_vgg old

benchmark vgg being replaced by bechmark vgg tmp.
Small change:
remove blank line in benchmarkvgg

Update script benchmark_vgg_multiview.py so it takes --w-after arg.