Medaka

Minor fix release.

Fixed
- Additional fix to handling lowercase reference sequences.
- Fix bug in creation of RLE alignments.
Added
- Update `update_model.py` script.
Changed
- Unify how LabelSchemes store training data.
Removed
- Remove option to select labelling scheme during training.

Minor fix release

Fixed
- Fix regression in medaka stitch and medaka snp speed.
- Remove dill and yaml requirements.
Added
- Handle lowercase letters in reference sequences.

Bugfix and training refactor release

Fixed
- Fix readlink issue on MacOS
- Fix bug where medaka_variant did not call indels by default
- Fix bug in determining when to split contigs
- Make network feature generation 2x faster
Added
- Add smolecule command
- Log use of GPU and cuDNN, noting workaround for RTX cards
Changed
- Store models in git-lfs
- Simplify medaka_variant workflow for speed
- Refactor labelling of training data and storing of models
- Reimplement RLE feature generation
Removed
- Drop support for older basecaller models (guppy<3.0.3)

Documentation release

Added
- Clarify suggested workflows in documentation.

Patch release

Fixed
- Patch import of loading of older models

Model release and development release

Added
- Add support for R10 basecaller
- Add diploid multi-labelling
Changed
- Upgrade to tensorflow 1.14.0