Pointnet2.ScanNet
Pointnet2.ScanNet copied to clipboard
What the "CK" means in eval.py?
Hi, @daveredrum I'm happy to meet your wonderful code!
I have a question about 'CK''s meaning in eval.py. I think N is num of points and C is num of class. However, I have no idea about CK. Can you help me? thanks.
In eval.py def forward(args, model, coords, feats):
pred = []
coord_chunk, feat_chunk = torch.split(coords.squeeze(0), args.batch_size, 0), torch.split(feats.squeeze(0), args.batch_size, 0)
assert len(coord_chunk) == len(feat_chunk)
for coord, feat in zip(coord_chunk, feat_chunk):
output = model(torch.cat([coord, feat], dim=2))
pred.append(output)
pred = torch.cat(pred, dim=0).unsqueeze(0) # (1, CK, N, C)