Skip to main content

Infering novel classes for classification problems

Project description

pyopenset

Pyopenset is a package that implements the meyhodology described in the paper:

It allows a easy and intuitive way to add novel classes to a pretrained model without requiring further optimization, New classes will always be added after the last.

So for example if your model has 10 neurons each for a single class ,the new class would be associated with neuron 11.

Note that current implementation only supports softmax activation functions on the last layer.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pyopenset-0.0.12.tar.gz (2.5 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page