**Full Changelog**: https://github.com/dingo-gw/dingo/compare/v0.6.2...v0.7.0
Main changes
* Incorporate Flow Matching Posterior Estimation (https://arxiv.org/abs/2305.17161) and score matching, as alternatives to normalizing flows for modeling the posterior.
* Refactor `PosteriorModel` class into hierarchy of models, covering normalizing flows, continuous flows, flow matching, etc.
* Update to training config file specification:
- Replace `model.type` with `model.posterior_model_type`. Options are `normalizing_flow` (replacing `nsf+embedding`), `flow_matching`, and `score_matching`.
- Replace `model.nsf_kwargs` with `model.posterior_kwargs`.
- Replace `model.embedding_net_kwargs` with `model.embedding_kwargs`.
- Details can be seen in the examples (https://github.com/dingo-gw/dingo/tree/main/examples). The previous configuration file format will continue to work for the time being to accommodate existing trained networks.
Smaller changes
* Add option for early stopping during training.
Note that the documentation still needs to be updated to reflect this update.