gngdb
commented
9 years ago Implemented with two run settings:
"shuffle", set to 1 to activate shuffle, 0 for normal (same as it not being there)."random seed"chooses the random seed to apply before shuffling.
The split used by Pylearn2 for training, validation and test in dense_dataset relies on an invariant order of training image paths from the settings file. Currently these are just sorted to make sure they're always the same. It would be good to use a random seed and then shuffle at this point, from the point of view of the Pylearn2 dataset split. Having such a deterministic split may be causing us problems.