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
I would like to use an ONNX optimized version of a model trained using your library. However, for the use case of one-shot classification I need to determine a threshold for classification. Are there any utility functions which can calculate the optimal threshold based on a validation set, and a metric that I want to optimize for?
The text was updated successfully, but these errors were encountered:
Yes, that's exactly what I mean. Ideally, it would take in a model, distance, splits to evaluate, and metric to optimize for. Would you be open to PRs for this? Ofcourse, if you think this would be a good addition to the library.
A PR would be great, assuming there isn't already a library that does this. There are probably other people looking for something similar, so I think it would be a good addition.
Hi,
I would like to use an ONNX optimized version of a model trained using your library. However, for the use case of one-shot classification I need to determine a threshold for classification. Are there any utility functions which can calculate the optimal threshold based on a validation set, and a metric that I want to optimize for?
The text was updated successfully, but these errors were encountered: