These are chat archives for beniz/deepdetect

21st
May 2017
ahoquegh
@ahoquegh
May 21 2017 03:33
service creation mllib bad param: using template while network weights exist, remove 'template' from 'mllib' or would you like to 'finetune' instead ?
alkollo
@alkollo
May 21 2017 09:59
hello, How many pictures per class do we need for a decent training in image training process ? And also is there some doc on meaning of training status (precision, train_loss, mcll, f1, etc), I check on caffe doc but can't find any clue. Thanks by advance.
Emmanuel Benazera
@beniz
May 21 2017 10:02
These are generic deep learning / machine learning matters. Look some generic tutorial and courses out there, there are many.