Thomas Varsavsky

Feature: Whole volume validation

4

It would be very helpful to do full volume validation during training. This would involve running one of the Aggregators instead of the Samplers mid-training and using the new `niftynet.engine.signal`s...

Feature: Introducing handler_performance

1

We're introducing a `handler_performance.py` which will keep a performance history which other handlers can access to perform performance dependent logic. Future examples include early stopping, reducing LR on plateau and...

Feature request: CSV Reader

1

## This issue is to track progress on new features for the CSVReader The current version is in the `refactoring-data-app` branch. Its intended use is to be called within any...

Bug: A keyword in the wrong .ini section should cause the application to fail, it doesn't

1

If you put in a keyword which is not valid in ANY of the `.ini` sections then NiftyNet will fail with an Error message saying "Unknown keywords in config file"....

Feature proposal: TorchIO hub - A system to store and fetch transform objects for reproducibility

6

**🚀 Feature** Introducing a public **TorchIO hub** where researchers can save the transform object used to randomly sample and augment their data during training with one line of code -...

Nifti files with non-orthonormal directional cosines should be read and the error changed to a warning.

15

Hi all, In this PR - https://github.com/InsightSoftwareConsortium/ITK/pull/1868/files ITK introduced an additional constraint on NiFTI files being read. This has affected users across a variety of technologies which use ITK such...

Weights/Code for CM3Leon

2

## 🚀 Feature Request Congratulations on this great work - Scaling Autoregressive Multi-Modal Models:Pretraining and Instruction Tuning. The paper mentions that `metaseq` was used for the model training. Just wondering...