HERRO is a highly-accurate, haplotype-aware, deep-learning tool for error correction of Nanopore R10.4.1 or R9.4.1 reads (read length of >= 10 kbps is recommended).
Hi
Will you share the code of HERRO model and training script? I'm wondering how to set S in the final part of the model which consists of two classification heads. In addition, the window size is 4096 which is the number of the base of the target read, there may exist a insertion at each position, so the exact length of each window will be lager than 4096. Is this right?
Hi Will you share the code of HERRO model and training script? I'm wondering how to set S in the final part of the model which consists of two classification heads. In addition, the window size is 4096 which is the number of the base of the target read, there may exist a insertion at each position, so the exact length of each window will be lager than 4096. Is this right?