FASTA file (upload .fa/.fasta/.gz):

Sequence type: DNA Protein

encoding (feature tokenization method): k-mer bag-of-words one-hot fixed length

k (k-mer size; only for kmer-bow):

hash_dim (hashed dimension; 0=explicit vocab):

normalize (vector normalization): none l1 l2

epochs (training passes):

batch_size (mini-batch size):

device (compute device): auto cpu cuda