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
Training a model should also provide a way to know the label name and ordering without knowledge of the dataset. It would also prevent potential errors if we default to use the labels provided inside the artifact directory next to the model weights.
Feature motivation
I am training a model that classifies kanjis with many classes (~5600) and no obvious label ordering so it could be prone to error between a prediction [0.1, ..., 0.5] and knowing that this means:
一: 0.1
...
龠: 0.5
The text was updated successfully, but these errors were encountered:
Feature description
Training a model should also provide a way to know the label name and ordering without knowledge of the dataset. It would also prevent potential errors if we default to use the labels provided inside the artifact directory next to the model weights.
Feature motivation
I am training a model that classifies kanjis with many classes (~5600) and no obvious label ordering so it could be prone to error between a prediction [0.1, ..., 0.5] and knowing that this means:
The text was updated successfully, but these errors were encountered: