optimize/test more on multi-gpu support in call_mods module
add scipy==1.7.3 in requirements (floating point error in scipy>=1.9->tombo). fix this in deepsignal-plant, although tombo only is the pre-process tool for this tool.
replace np.float/np.int with float/int in extract_features module
bug fixes
enable .gz output in extract/call_mods/call_freq modules
multi-gpu support in call_mods module
update requirements/dependences
make sure results of each read be written together in call_mods' output
make --reference_path not a required input in extract and call_mods module
modify call_freq module for large genomes,
fix bug of extrating contig name from fast5s
add ranger optimizer and modify train module,
fix Queue.qsize() NotImplementedError in macOS partially, however call_mods in CUDA mode in macOS still doesn't work,
add init_model option in train module
change imports in ref_reader,
change requirements,
fix and modify denoise module,
fix MKL_THREADING_LAYER error temporarily,
add --region option,
a combined 5mC model replacing CG/CHG/CHH models
fix bug and optimize call_mods without GPU, add call_freq funciton
Release the first vesrion of deepsignal-plant package