NickleDave / visual-search-nets

neural network models of visual search behavior
BSD 3-Clause "New" or "Revised" License
6 stars 2 forks source link

have datasets return dictionaries #64

Closed NickleDave closed 4 years ago

NickleDave commented 4 years ago

because for VOC we want all classes present in addition to largest and random class if required

so for both searchstims and VOC, return dictionary with {'img', 'target'}

additionally for VOC, if largest transform is specified then add 'largest', if 'random' is specified have 'random' ... and so on

also include visual search difficulty score as a key for VOC

NickleDave commented 4 years ago

VOC dict should have