error correction fails when ran with batched alignments

[nponts]

Hi! i decided to give a try to Herro, and after repeating errors others did before me (which means a solution was out there) I found my own brand new problem... + an additional question actually...

The problem: when i run herro inference without using already made batched alignments, it runs (still waiting for the result for now); when i run herro inference with already made batched alignments (the read-alns option), it finishes quickly with "processed 0 reads" although parsing occurred fine - what can go wrong?

The question: I work with mostly small haploid fungal genomes; your model was trained on human data. Do you think it still applies?

Thank you!!!

[nponts]

Update: fails no matter what... Always get "processed 0 reads"...

[dominikstanojevic]

Hello,

can you try to compile the tool from this branch (https://github.com/lbcb-sci/herro/tree/decoding-debug) and rerun it? It should give more information about the problem. If it is helpful I will add push it to the main branch.