Skip to content

Aligning prototype-based networks using cycle consistency to train sample efficient models for domain adaptation

Notifications You must be signed in to change notification settings

alexquach/prototype-based-domain-adaptation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 

Repository files navigation

Protoype-based Domain Adaptation

Extends the paper: Deep Learning for Case-Based Reasoning through Prototypes: A Neural Network that Explains Its Predictions: https://arxiv.org/abs/1710.04806

Important files:

Models:

/src/models/proto_model.py

  • Implements a single ProtoModel similar to the paper above

/src/models/cycle_model.py

  • Implements a pair of ProtoModels (one for the source and one for the target domains) to learn a transition function that maps from the source domain to the target domain

Training Scripts:

/src/scripts/train.py

  • Trains a single ProtoModel

/src/scripts/train_cycle.py

  • Trains a Cycle Model

About

Aligning prototype-based networks using cycle consistency to train sample efficient models for domain adaptation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages