We made some effort to make this easy to extend and we are very happy that you are interested in cooking your own CL model using PyContinual. In order to achieve this, you only need to review the following files: ./networks
, ./approaches
, and impoty_classification.py
- Build your network architecture in
./networks
- Define your training strategy in
./approaches
- (Optional) add you new dataloader in
./dataloader
- Finally, add the import statement in the
impoty_classification.py
- You are now all set! You are more than welcomed to create a PR for your model to make PyContinual "continually" growing!