Open Adamtaranto opened 1 month ago
oxli is currently slower than other counters for large genomes. Suggest option to bulk load kmer counts from another counter i.e. Jellyfish via text dump.
Would allow users to load pre-calculated counts for use with oxli's python API.
@ctb what do you reckon?
The other option is for users to use the Python API to manually set the counts i.e. kct[kmer] = count.
kct[kmer] = count
Only advantage of adding a method to load counts from a dump file is that we could make it multithreaded.
oxli is currently slower than other counters for large genomes. Suggest option to bulk load kmer counts from another counter i.e. Jellyfish via text dump.
Would allow users to load pre-calculated counts for use with oxli's python API.
@ctb what do you reckon?