ModuleNotFoundError: No module named 'torch.utils._import_utils'

[jellyfishovo]

I find a similar issue happens on torch==2.3.0, and it has been solved, but in torch==2.1.0, I try the command “import torch
torch.utils.data.datapipes.utils.common.DILL_AVAILABLE = torch.utils._import_utils.dill_available()
import torchdata” to fix this, it doesn't work.
my platform : torch==2.1.0, PyGCL==0.1.2, dgl==2.3.0+cu118, cuda==12.0
error details:
Traceback (most recent call last):
File "/home/sjd/Projects/New/ICASSP/train.py", line 10, in
from model import MaskedNLLLoss, LSTMModel, GRUModel, Model, MaskedMSELoss, FocalLoss
File "/home/sjd/Projects/New/ICASSP/model.py", line 10, in
from HCL_model import HCLGNN
File "/home/sjd/Projects/New/ICASSP/HCL_model.py", line 11, in
from GCL.models import DualBranchContrast
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/GCL/init.py", line 2, in
import GCL.augmentors
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/GCL/augmentors/init.py", line 3, in
from .rw_sampling import RWSampling
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/GCL/augmentors/rw_sampling.py", line 2, in
from GCL.augmentors.functional import random_walk_subgraph
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/GCL/augmentors/functional.py", line 6, in
from GCL.utils import normalize
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/GCL/utils.py", line 4, in
import dgl
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/dgl/init.py", line 16, in
from . import (
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/dgl/dataloading/init.py", line 13, in
from .dataloader import *
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/dgl/dataloading/dataloader.py", line 27, in
from ..distributed import DistGraph
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/dgl/distributed/init.py", line 5, in
from .dist_graph import DistGraph, DistGraphServer, edge_split, node_split
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/dgl/distributed/dist_graph.py", line 12, in
from .. import backend as F, graphbolt as gb, heterograph_index
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/dgl/graphbolt/init.py", line 39, in
from .base import *
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/dgl/graphbolt/base.py", line 21, in
from torchdata.datapipes.iter import IterDataPipe
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/torchdata/datapipes/init.py", line 11, in
from . import iter, map, utils
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/torchdata/datapipes/iter/init.py", line 79, in
from torchdata.datapipes.iter.util.cacheholder import (
File "/home/sjd/miniconda3/envs/icassp/lib/python3.9/site-packages/torchdata/datapipes/iter/util/cacheholder.py", line 24, in
from torch.utils._import_utils import dill_available
ModuleNotFoundError: No module named 'torch.utils._import_utils'

Process finished with exit code 1

[SarahAlidoost]

@jellyfishovo The error message shows that you are using the dgl package rather than the dgl_ptm package developed in this repository. Since your issue is not related to dgl_ptm package, this is not the correct repository to report it. To help you, here are some links dmlc/dgl#7380 and dmlc/dgl#7387 to relevant issues in dgl repository. You might find helpful insights there on how to fix the error.