You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If we want to be able to verify our progress we need to have standardized set of products.
Build a list of products (list of barcode):
A first list should be used to be used for training / test sets (we can have a standard split, but this should be also used for cross training with different train / test split)
A second list should be used for validation (it's always the same)
However the sets should be built accounting for the different features present in data. So there should be some balancing in training / test / validation set for data having the different modalities : title / ingredients / nutritional data / images / OCR.
If we want to be able to verify our progress we need to have standardized set of products.
Build a list of products (list of barcode):
However the sets should be built accounting for the different features present in data. So there should be some balancing in training / test / validation set for data having the different modalities : title / ingredients / nutritional data / images / OCR.
depends on
The text was updated successfully, but these errors were encountered: