v1.2.0

Major Features

New INDEL Module and Unified Command-Line Interface:

• Introduced the new mural_indel module for predicting INDEL mutation rates.
• Refactored the existing SNV tool invocation to mural_snv .
• Established unified command patterns between mural_indel and mural_snv with consistent subcommand structures
• Implemented parallel workflows including train, predict, evaluate, transfer, calc_scaling_factor, scale, get_best_model subcommands for both modules

Documentation and CLI enhancements

• Updated all documentation to reflect version 1.2.0 changes and new features
• Add DEVELOPER.md to separate developer-focused parameters from main CLI documentation

Version Management

• Updated the version in MuRaL/_version.py to 1.2.0
• Documented all major features and changes in the CHANGELOG file

v1.1.2

• use a new preprocessing strategy (segmenting the genome and more effective I/O operations)
• modify data loading process
• add support for large input sequence lengths (up to 64,000 bp)
• add a new training control way (without Ray; see train_utils.py, printer_utils.py)
• update documentation

v1.1.1

• fix a bug in calc_kmer_corr
• add more doc for the scaling section
• remove unused code

v1.1.0

• add a new network model (model 3)
• add support for AdamW optimizer
• update documentation

v1.0.1

• remove janggu dependency
• add rtd documentation
• delete unused code

v1.0.0

• update network model (add middle-scale layers)
• improve preprocessing code (numpy implementation; HDF5 related code)
• improve optimization code ('--weight_decay_auto'; '--restart_lr')
• improve documentation

v0.9.1

• clean unnecessary code
• update weight initialization for bias
• add more detailed documentation

v0.9.0

• first public release.