magicpr
commented
2 days ago When running the code with the CPU, the error is reported as follows：
`Traceback (most recent call last):
File "/home/zbin/SEDR-master/SEDR/integration_cpu.py", line 73, in
IndexError: index 10885 is out of bounds for dimension 0 with size 10812`
I solved this problem by taking the smallest value in file tutorial 3、SEDR_module.py and SEDR_model.py
, but I wonder if it will affect the results?
tutorial 3
> sedr_feat, _, _, _ = sedr_net.process()
if sedr_feat.shape[0] != adata.shape[0]:
min_size = min(sedr_feat.shape[0], adata.shape[0])
sedr_feat = sedr_feat[:min_size]
adata = adata[:min_size]
adata.obsm['SEDR'] = sedr_feat
SEDR_module.py
> num_mask_nodes = int(mask_rate * num_nodes)
mask_nodes = perm[: num_mask_nodes]
keep_nodes = perm[num_mask_nodes:]
if x.shape[0] < num_nodes:
padding_size = num_nodes - x.shape[0]
padding = torch.zeros((padding_size, x.shape[1]), device=x.device)
x = torch.cat((x, padding), dim=0)
elif x.shape[0] > num_nodes:
x = x[:num_nodes]
out_x = x.clone()
token_nodes = mask_nodes
SEDR_model.py
> def reconstruction_loss(decoded, x):
if decoded.shape != x.shape:
min_size = min(decoded.shape[0], x.shape[0])
decoded = decoded[:min_size]
x = x[:min_size]
loss_func = torch.nn.MSELoss()
loss_rcn = loss_func(decoded, x)
return loss_rc
def mask_generator(self, N=1):
x = torch.repeat_interleave(self.adj_label.indices()[0], N)
y = torch.concat(list_non_neighbor)
# 确保 x 和 y 的大小一致
if x.size(0) != y.size(0):
min_size = min(x.size(0), y.size(0))
x = x[:min_size]
y = y[:min_size]
indices = torch.stack([x, y])
indices = torch.concat([self.adj_label.indices(), indices], axis=1)
现在运行的结果如下
>enc_mask_token.shape: torch.Size([1, 200])
2024-11-05 12:59:32,202 - harmonypy - INFO - Computing initial centroids with sklearn.KMeans...
2024-11-05 12:59:34,982 - harmonypy - INFO - sklearn.KMeans initialization complete.
2024-11-05 12:59:35,161 - harmonypy - INFO - Iteration 1 of 10
2024-11-05 12:59:39,809 - harmonypy - INFO - Iteration 2 of 10
2024-11-05 12:59:41,276 - harmonypy - INFO - Iteration 3 of 10
2024-11-05 12:59:42,782 - harmonypy - INFO - Converged after 3 iterations
Xuhang01
Hi, I recently had this problem while reproducing the tutorial 3：
100%|████████████████████████████████████████████████████████████████████████| 3/3 [00:08<00:00, 2.83s/it] 0%| | 0/200 [00:00<?, ?it/s]../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [0,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [1,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [2,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [3,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [4,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [5,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [6,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [7,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [8,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [9,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [10,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [11,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [12,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [13,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [14,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. ../aten/src/ATen/native/cuda/IndexKernel.cu:92: operator(): block: [1382,0,0], thread: [15,0,0] Assertionindex >= -sizes[i] && index < sizes[i] && "index out of bounds"failed. 0%| | 0/200 [00:00<?, ?it/s] Traceback (most recent call last): File "/home/zbin/SEDR-master/SEDR/integration.py", line 70, in <module> sedr_net.train_without_dec() File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/SEDR/SEDR_model.py", line 142, in train_without_dec latent_z, mu, logvar, de_feat, _, feat_x, _, loss_self = self.model(self.X, self.adj_norm) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/SEDR/SEDR_module.py", line 176, in forward mu, logvar, feat_x = self.encode(x, adj) ^^^^^^^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/SEDR/SEDR_module.py", line 160, in encode feat_x = self.encoder(x) ^^^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/torch/nn/modules/container.py", line 217, in forward input = module(input) ^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/torch/nn/modules/container.py", line 217, in forward input = module(input) ^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/zbin/anaconda3/envs/sedr_env/lib/python3.11/site-packages/torch/nn/modules/linear.py", line 114, in forward return F.linear(input, self.weight, self.bias) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ RuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when callingcublasCreate(handle)``My GPU use is about close to 100%, so I think the main reason should be the array out of the index, but I don't know why others have not encountered, the relevant data about DLPFC was downloaded from the analyses and the tutorial. Please take a look at how this should be solved?