search

AiEson / CrossMatch

CrossMatch: Enhance Semi-Supervised Medical Image Segmentation with Perturbation Strategies and Knowledge Distillation
MIT License
19 stars 0 forks source link

Multi GPUs training problem #1

Closed Leesoon1984 closed 4 months ago

Leesoon1984 commented 4 months ago

Traceback (most recent call last): File "/home/mnt/lee/Med/CrossMatch/ACDC/train_cross_match.py", line 420, in main() File "/home/mnt/lee/Med/CrossMatch/ACDC/train_cross_match.py", line 183, in main for i, ( File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 630, in next data = self._next_data() ^^^^^^^^^^^^^^^^^ File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 1325, in _next_data return self._process_data(data) ^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 1371, in _process_data data.reraise() File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/_utils.py", line 694, in reraise raise exception OSError: Caught OSError in DataLoader worker process 11. Original Traceback (most recent call last): File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/_utils/worker.py", line 308, in _worker_loop data = fetcher.fetch(index) ^^^^^^^^^^^^^^^^^^^^ File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/_utils/fetch.py", line 51, in data = [self.dataset[idx] for idx in possibly_batched_index]


  File "/home/mnt/lee/Med/CrossMatch/ACDC/dataset/acdc.py", line 37, in __getitem__
    sample = h5py.File(os.path.join(self.root, id), 'r')
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/h5py/_hl/files.py", line 562, in __init__
    fid = make_fid(name, mode, userblock_size, fapl, fcpl, swmr=swmr)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/h5py/_hl/files.py", line 235, in make_fid
    fid = h5f.open(name, flags, fapl=fapl)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
  File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
  File "h5py/h5f.pyx", line 102, in h5py.h5f.open
OSError: [Errno 9] Unable to synchronously open file (unable to lock file, errno = 9, error message = 'Bad file descriptor')
AiEson commented 4 months ago

I executed all the processes in the README and found no problems. The possible reason is that you did not clone the repository correctly, causing some files to be damaged.

bash scripts/train.sh 1 12333                                                                                                                                (CrossMatch)
[2024-05-16 17:19:25,385][    INFO] {'T': 1.0,
 'batch_size': 12,
 'conf_thresh': 0.95,
 'config': 'configs/acdc.yaml',
 'crop_size': 256,
 'data_root': './dataset/datasets/ACDC',
 'dataset': 'acdc',
 'drop_rate': 0.5,
 'epochs': 300,
 'eta': 0.3,
 'labeled_id_path': 'splits/acdc/3/labeled.txt',
 'local_rank': 0,
 'lr': 0.01,
 'nclass': 4,
 'ngpus': 1,
 'port': 12333,
 'save_path': 'exp/acdc/train_cross_match_cross_match/unet/3/eta_0.3',
 'unlabeled_id_path': 'splits/acdc/3/unlabeled.txt',
 'use_threshold_relax': False}

[2024-05-16 17:19:25,466][    INFO] Total params: 1.8M

[2024-05-16 17:19:26,427][    INFO] ===========> Epoch: 0, LR: 0.01000, Previous best: 0.00, TH: 0.9500
[2024-05-16 17:19:32,535][    INFO] Iters: 0, Total loss: 0.821, Loss x: 1.175, Loss s: 0.693, Loss KD: 0.595, Mask ratio: 0.000
[2024-05-16 17:19:36,077][    INFO] Iters: 12, Total loss: 0.834, Loss x: 1.011, Loss s: 0.752, Loss KD: 0.738, Mask ratio: 0.001
fish: Job 1, 'bash scripts/train.sh 1 12333' has stopped
Leesoon1984 commented 4 months ago

I executed all the processes in the README and found no problems. The possible reason is that you did not clone the repository correctly, causing some files to be damaged.

bash scripts/train.sh 1 12333                                                                                                                                (CrossMatch)
[2024-05-16 17:19:25,385][    INFO] {'T': 1.0,
 'batch_size': 12,
 'conf_thresh': 0.95,
 'config': 'configs/acdc.yaml',
 'crop_size': 256,
 'data_root': './dataset/datasets/ACDC',
 'dataset': 'acdc',
 'drop_rate': 0.5,
 'epochs': 300,
 'eta': 0.3,
 'labeled_id_path': 'splits/acdc/3/labeled.txt',
 'local_rank': 0,
 'lr': 0.01,
 'nclass': 4,
 'ngpus': 1,
 'port': 12333,
 'save_path': 'exp/acdc/train_cross_match_cross_match/unet/3/eta_0.3',
 'unlabeled_id_path': 'splits/acdc/3/unlabeled.txt',
 'use_threshold_relax': False}

[2024-05-16 17:19:25,466][    INFO] Total params: 1.8M

[2024-05-16 17:19:26,427][    INFO] ===========> Epoch: 0, LR: 0.01000, Previous best: 0.00, TH: 0.9500
[2024-05-16 17:19:32,535][    INFO] Iters: 0, Total loss: 0.821, Loss x: 1.175, Loss s: 0.693, Loss KD: 0.595, Mask ratio: 0.000
[2024-05-16 17:19:36,077][    INFO] Iters: 12, Total loss: 0.834, Loss x: 1.011, Loss s: 0.752, Loss KD: 0.738, Mask ratio: 0.001
fish: Job 1, 'bash scripts/train.sh 1 12333' has stopped

[2024-05-16 16:12:13,033][    INFO] ===========> Epoch: 62, LR: 0.00812, Previous best: 81.34, TH: 0.9500
[2024-05-16 16:12:17,256][    INFO] Iters: 0, Total loss: 0.068, Loss x: 0.038, Loss s: 0.051, Loss KD: 0.115, Mask ratio: 0.977
[2024-05-16 16:12:18,553][    INFO] Iters: 3, Total loss: 0.048, Loss x: 0.040, Loss s: 0.031, Loss KD: 0.075, Mask ratio: 0.979
[2024-05-16 16:12:19,844][    INFO] Iters: 6, Total loss: 0.055, Loss x: 0.041, Loss s: 0.035, Loss KD: 0.089, Mask ratio: 0.978
Traceback (most recent call last):
  File "/home/mnt/lee/Med/CrossMatch/ACDC/train_cross_match.py", line 420, in <module>
    main()
  File "/home/mnt/lee/Med/CrossMatch/ACDC/train_cross_match.py", line 183, in main
    for i, (
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 630, in __next__
    data = self._next_data()
           ^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 1325, in _next_data
    return self._process_data(data)
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 1371, in _process_data
    data.reraise()
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/_utils.py", line 694, in reraise
    raise exception
OSError: Caught OSError in DataLoader worker process 9.
Original Traceback (most recent call last):
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/_utils/worker.py", line 308, in _worker_loop
    data = fetcher.fetch(index)
           ^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/utils/data/_utils/fetch.py", line 51, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
            ~~~~~~~~~~~~^^^^^
  File "/home/mnt/lee/Med/CrossMatch/ACDC/dataset/acdc.py", line 37, in __getitem__
    sample = h5py.File(os.path.join(self.root, id), 'r')
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/h5py/_hl/files.py", line 562, in __init__
    fid = make_fid(name, mode, userblock_size, fapl, fcpl, swmr=swmr)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/h5py/_hl/files.py", line 235, in make_fid
    fid = h5f.open(name, flags, fapl=fapl)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
  File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
  File "h5py/h5f.pyx", line 102, in h5py.h5f.open
OSError: [Errno 9] Unable to synchronously open file (unable to lock file, errno = 9, error message = 'Bad file descriptor')

[2024-05-16 16:12:22,257] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 30 closing signal SIGTERM
[2024-05-16 16:12:22,258] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 32 closing signal SIGTERM
[2024-05-16 16:12:22,258] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 33 closing signal SIGTERM
[2024-05-16 16:12:22,377] torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: 1) local_rank: 1 (pid: 31) of binary: /home/mnt/lee/miniconda3/envs/CrossMatch/bin/python
Traceback (most recent call last):
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/bin/torchrun", line 8, in <module>
    sys.exit(main())
             ^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 346, in wrapper
    return f(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/distributed/run.py", line 806, in main
    run(args)
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/distributed/run.py", line 797, in run
    elastic_launch(
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/distributed/launcher/api.py", line 134, in __call__
    return launch_agent(self._config, self._entrypoint, list(args))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/mnt/lee/miniconda3/envs/CrossMatch/lib/python3.11/site-packages/torch/distributed/launcher/api.py", line 264, in launch_agent
    raise ChildFailedError(
torch.distributed.elastic.multiprocessing.errors.ChildFailedError: 
============================================================
train_cross_match.py FAILED
------------------------------------------------------------
Failures:
  <NO_OTHER_FAILURES>
------------------------------------------------------------
Root Cause (first observed failure):
[0]:
  time      : 2024-05-16_16:12:22
  host      : pt-r5qx9yk4-worker-0.pt-r5qx9yk4.ns-sensetime2023-cb5b8530.svc.cluster.local
  rank      : 1 (local_rank: 1)
  exitcode  : 1 (pid: 31)
  error_file: <N/A>
  traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
============================================================
AiEson commented 4 months ago

Please try running the code below:

import h5py
import glob
import tqdm

ACDC_DIR = '/home/mnt/lee/Med/CrossMatch/ACDC/dataset/datasets/ACDC'

print("Checking the dataset... type: Slices")
for file_name in tqdm.tqdm(glob.glob(f'{ACDC_DIR}/data/slices/*.h5')):
    sample = h5py.File(file_name, 'r')
    img = sample['image'][:]
    mask = sample['label'][:]
print("Finished!")

print("Checking the dataset... type: 3D Images")
for file_name in tqdm.tqdm(glob.glob(f'{ACDC_DIR}/data/*.h5')):
    sample = h5py.File(file_name, 'r')
    img = sample['image'][:]
    mask = sample['label'][:]
print("Finished!")

to check if your dataset can be read correctly.

Leesoon1984 commented 4 months ago

Please try running the code below:

import h5py
import glob
import tqdm

ACDC_DIR = '/home/mnt/lee/Med/CrossMatch/ACDC/dataset/datasets/ACDC'

print("Checking the dataset... type: Slices")
for file_name in tqdm.tqdm(glob.glob(f'{ACDC_DIR}/data/slices/*.h5')):
    sample = h5py.File(file_name, 'r')
    img = sample['image'][:]
    mask = sample['label'][:]
print("Finished!")

print("Checking the dataset... type: 3D Images")
for file_name in tqdm.tqdm(glob.glob(f'{ACDC_DIR}/data/*.h5')):
    sample = h5py.File(file_name, 'r')
    img = sample['image'][:]
    mask = sample['label'][:]
print("Finished!")

to check if your dataset can be read correctly. image

Leesoon1984 commented 4 months ago

https://github.com/h5py/h5py/issues/1101

AiEson commented 4 months ago

h5py/h5py#1101

Thanks for you to provide the solution