open-compass
[Bug] 多卡时，GPU7显存占用比其他卡多30G+
先决条件
问题类型
我修改了代码（配置不视为代码），或者我正在处理我自己的任务/模型/数据集。
环境
{'CUDA available': True,
'CUDA_HOME': '/usr',
'GCC': 'gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0',
'GPU 0,1,2,3,4,5,6,7': 'NVIDIA A100-SXM4-80GB',
'MMEngine': '0.10.4',
'MUSA available': False,
'NVCC': 'Cuda compilation tools, release 11.5, V11.5.119',
'OpenCV': '4.10.0',
'PyTorch': '2.3.0+cu121',
'PyTorch compiling details': 'PyTorch built with:\n'
' - GCC 9.3\n'
' - C++ Version: 201703\n'
' - Intel(R) oneAPI Math Kernel Library Version '
'2023.1-Product Build 20230303 for Intel(R) 64 '
'architecture applications\n'
' - Intel(R) MKL-DNN v3.3.6 (Git Hash '
'86e6af5974177e513fd3fee58425e1063e7f1361)\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: AVX2\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 8.9.2\n'
' - Magma 2.6.1\n'
' - Build settings: BLAS_INFO=mkl, '
'BUILD_TYPE=Release, CUDA_VERSION=12.1, '
'CUDNN_VERSION=8.9.2, '
'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_QNNPACK -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=2.3.0, '
'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': '3.11.9 (main, Apr 19 2024, 16:48:06) [GCC 11.2.0]',
'TorchVision': '0.18.1+cu121',
'numpy_random_seed': 2147483648,
'opencompass': '0.2.5+e0d7808',
'sys.platform': 'linux'}
重现问题 - 代码/配置示例
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python run.py --models hf_xdg-llama-3-8b --datasets ceval_gen cmmlu_gen mmlu_gen --num-gpus 8
我尝试其他的模型也有这个问题
重现问题 - 命令或脚本
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python run.py --models hf_xdg-llama-3-8b --datasets ceval_gen cmmlu_gen mmlu_gen --num-gpus 8
这个模型是我自己定制的
重现问题 - 错误信息
其他信息
感谢你们提供的测评工具，以下是我在使用opencompass时遇到的问题：
当前采用多卡推理，8卡8进程，相当于数据并行，模型不做切分。
预期应该每个卡占用是相近的。 这里的gpu0和gpu7相差非常大。请问如何解决？
另外：
- 另外整体推理速度很慢，能否支持vllm推理
- 能否增加与LLM安全问题相关的测评数据集
期待你的回复和建议。
Hi, we actually has supported the vllm and lmdeploy. Please check the doc.
Thank you for offering the open-compass.
I've tried the lmdeploy. It achieved stability and high infer speed. However, when I set num_gpus = 2 in the configs, it still run on only one gpu. How can I fix this?
Hi, we actually has supported the vllm and lmdeploy. Please check the doc.
Thank you for offering the open-compass. I've tried the lmdeploy. It achieved stability and high infer speed. However, when I set num_gpus = 2 in the configs, it still run on only one gpu. How can I fix this?
你好，我也遇到同样的情况，单机多卡情况还是只用一张卡推理，请问你解决了吗
