This release is about being cohesive in function usage, imports, and parallelism.
Parallelism was added to prediction, each read in the sample file spawns a concurrent process.
Also, some cleaning and bugfix was performed along the way.
Upgrades :
- Added multithreading support to prediction
- Moved predict loop to dedicated module
- Added scripts to plot graphs from results
Fixes :
- Removed bug where read threshold inferior to 10kb would crash the software
- Removed bug where building database would fail if a specie had only fragments of length inferior to threshold