Question about the SCDataset class
Opened this issue · 0 comments
ChaozhongLiu commented
Hi there,
Thanks for the amazing work! When I was reading the source code, I got one question in mind and couldn't find an answer myself. It will great if any of you can help!
So in the SCDataset class, when a single cell is being sampled, why it appends a 0 at the end of the gene expression vector?
class SCDataset(Dataset):
def __init__(self, data):
super().__init__()
self.data = data
def __getitem__(self, index):
rand_start = random.randint(0, self.data.shape[0]-1)
full_seq = self.data[rand_start].toarray()[0]
full_seq[full_seq > (CLASS - 2)] = CLASS - 2
full_seq = torch.from_numpy(full_seq).long()
############################################################
full_seq = torch.cat((full_seq, torch.tensor([0]))).to(device)
############################################################
return full_seq
def __len__(self):
return self.data.shape[0]
Thanks!