[Bug] cannot use GPU when conduct evaluation

[takagi97]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• The bug has not been fixed in the latest version.

Type

I have modified the code (config is not considered code), or I'm working on my own tasks/models/datasets.

Environment

{'CUDA available': True,
'CUDA_HOME': '/usr/local/cuda',
'GCC': 'gcc (GCC) 8.3.1 20191121 (Red Hat 8.3.1-5)',
'GPU 0,1,2,3,4,5,6,7': 'NVIDIA A100-SXM4-40GB',
'MMEngine': '0.10.5',
'MUSA available': False,
'NVCC': 'Cuda compilation tools, release 12.1, V12.1.66',
'OpenCV': '4.10.0',
'PyTorch': '2.4.0+cu121',
'PyTorch compiling details': 'PyTorch built with:\n'
' - GCC 9.3\n'
' - C++ Version: 201703\n'
' - Intel(R) oneAPI Math Kernel Library Version '
'2022.2-Product Build 20220804 for Intel(R) 64 '
'architecture applications\n'
' - Intel(R) MKL-DNN v3.4.2 (Git Hash '
'1137e04ec0b5251ca2b4400a4fd3c667ce843d67)\n'
' - OpenMP 201511 (a.k.a. OpenMP 4.5)\n'
' - LAPACK is enabled (usually provided by '
'MKL)\n'
' - NNPACK is enabled\n'
' - CPU capability usage: AVX512\n'
' - CUDA Runtime 12.1\n'
' - NVCC architecture flags: '
'-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_90,code=sm_90\n'
' - CuDNN 90.1 (built against CUDA 12.4)\n'
' - Magma 2.6.1\n'
' - Build settings: BLAS_INFO=mkl, '
'BUILD_TYPE=Release, CUDA_VERSION=12.1, '
'CUDNN_VERSION=9.1.0, '
'CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, '
'CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 '
'-fabi-version=11 -fvisibility-inlines-hidden '
'-DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO '
'-DLIBKINETO_NOROCTRACER -DUSE_FBGEMM '
'-DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK '
'-DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC '
'-Wall -Wextra -Werror=return-type '
'-Werror=non-virtual-dtor -Werror=bool-operation '
'-Wnarrowing -Wno-missing-field-initializers '
'-Wno-type-limits -Wno-array-bounds '
'-Wno-unknown-pragmas -Wno-unused-parameter '
'-Wno-unused-function -Wno-unused-result '
'-Wno-strict-overflow -Wno-strict-aliasing '
'-Wno-stringop-overflow -Wsuggest-override '
'-Wno-psabi -Wno-error=pedantic '
'-Wno-error=old-style-cast -Wno-missing-braces '
'-fdiagnostics-color=always -faligned-new '
'-Wno-unused-but-set-variable '
'-Wno-maybe-uninitialized -fno-math-errno '
'-fno-trapping-math -Werror=format '
'-Wno-stringop-overflow, LAPACK_INFO=mkl, '
'PERF_WITH_AVX=1, PERF_WITH_AVX2=1, '
'PERF_WITH_AVX512=1, TORCH_VERSION=2.4.0, '
'USE_CUDA=ON, USE_CUDNN=ON, USE_CUSPARSELT=1, '
'USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, '
'USE_GLOG=OFF, USE_GLOO=ON, USE_MKL=ON, '
'USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=1, '
'USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, '
'USE_ROCM_KERNEL_ASSERT=OFF, \n',
'Python': '3.10.14 (main, May 6 2024, 19:42:50) [GCC 11.2.0]',
'TorchVision': '0.19.0+cu121',
'lmdeploy': "not installed:No module named 'lmdeploy'",
'numpy_random_seed': 2147483648,
'opencompass': '0.3.7+f333be1',
'sys.platform': 'linux',
'transformers': '4.46.3'}

Reproduces the problem - code/configuration sample

I have integrated the WMT2024 benchmark (https://www2.statmt.org/wmt24/translation-task.html) into the current project. It uses COMET to conduct evaluation, a pre-trained language model that should be run on GPU. COMET can use GPU to calculate scores when "--debug" is set. However, when "--debug" is not set, COMET cannot access to GPU.

Reproduces the problem - command or script

can access GPUs when calculating COMET scores:
CUDA_VISIBLE_DEVICES=0 python run.py
--datasets wmt_news_gen
--hf-type chat
*--debug *
--work-dir outputs/test_project
--hf-path Llama-3.2-1B-Instruct
cannot access GPUs when calculating COMET scores:
CUDA_VISIBLE_DEVICES=0 python run.py
--datasets wmt_news_gen
--hf-type chat
--work-dir outputs/test_project
--hf-path Llama-3.2-1B-Instruct

Reproduces the problem - error message

can access GPUs when calculating COMET scores:
Starting inference process...
0%| | 0/1 [00:00<?, ?it/s]miniconda3/lib/python3.10/site-packages/transformers/generation/configuration_utils.py:590: UserWarning: do_sample is set to False. However, temperature is set to 0.6 -- this flag is only used in sample-based generation modes. You should set do_sample=True or unset temperature.
warnings.warn(
/miniconda3/lib/python3.10/site-packages/transformers/generation/configuration_utils.py:595: UserWarning: do_sample is set to False. However, top_p is set to 0.9 -- this flag is only used in sample-based generation modes. You should set do_sample=True or unset top_p.
warnings.warn(
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:10<00:00, 10.32s/it]
12/12 11:08:01 - OpenCompass - INFO - Partitioned into 1 tasks.
Encoder model frozen.
Lightning automatically upgraded your loaded checkpoint from v1.8.3.post1 to v2.4.0. To apply the upgrade to your files permanently, run python -m pytorch_lightning.utilities.upgrade_checkpoint ../../original_weights/wmt22-comet-da/checkpoints/model.ckpt
Encoder model frozen.
/miniconda3/lib/python3.10/site-packages/pytorch_lightning/core/saving.py:195: Found keys that are not in the model state dict but in the checkpoint: ['encoder.model.embeddings.position_ids']
GPU available: True (cuda), used: True
TPU available: False, using: 0 TPU cores
HPU available: False, using: 0 HPUs
You are using a CUDA device ('NVIDIA A100-SXM4-40GB') that has Tensor Cores. To properly utilize them, you should set torch.set_float32_matmul_precision('medium' | 'high') which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using tokenizers before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using tokenizers before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Predicting DataLoader 0: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 8.55it/s]

cannot access GPUs when calculating COMET scores:
launch OpenICLInfer[Llama-3.2-1B-Instruct_hf/wmt_2024_news_cs-uk] on GPU 0
12/12 11:51:26 - OpenCompass - INFO - Partitioned into 1 tasks.

0%| | 0/1 [00:00<?, ?it/s]

0%| | 0/1 [00:00<?, ?it/s]
100%|██████████| 1/1 [05:10<00:00, 310.83s/it]
100%|██████████| 1/1 [05:10<00:00, 310.83s/it]
launch OpenICLEval[Llama-3.2-1B-Instruct_hf/wmt_2024_news_cs-uk] on CPU
dataset version metric mode Llama-3.2-1B-Instruct_hf

---

wmt_2024_news_cs-uk 15dad7 BLEU gen 0.57
wmt_2024_news_cs-uk 15dad7 CHRF gen 11.89
wmt_2024_news_cs-uk 15dad7 TER gen 350.76
wmt_2024_news_cs-uk 15dad7 wmt22_comet_da gen 33.30
wmt_2024_news_cs-uk 15dad7 wmt23_cometkiwi_da_xl gen 23.42

Other information

No response