open-compass
[Bug] 在对DeepSeek-R1-Distill-Qwen-1.5B模型评测livecodebench数据集时，lcb_test_output为什么为0呢？
先决条件
问题类型
我正在使用官方支持的任务/模型/数据集进行评估。
环境
{'CUDA available': True, 'CUDA_HOME': '/usr/local/cuda-11.8', 'GCC': 'gcc (Ubuntu 11.4.0-2ubuntu1~20.04) 11.4.0', 'GPU 0': 'NVIDIA GeForce RTX 3090', 'MMEngine': '0.10.5', 'MUSA available': False, 'NVCC': 'Cuda compilation tools, release 11.8, V11.8.89', 'OpenCV': '4.10.0', 'PyTorch': '2.5.1+cu124', 'PyTorch compiling details': 'PyTorch built with:\n' ' - GCC 9.3\n' ' - C++ Version: 201703\n' ' - Intel(R) oneAPI Math Kernel Library Version ' '2024.2-Product Build 20240605 for Intel(R) 64 ' 'architecture applications\n' ' - Intel(R) MKL-DNN v3.5.3 (Git Hash ' '66f0cb9eb66affd2da3bf5f8d897376f04aae6af)\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 12.4\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\n' ' - Magma 2.6.1\n' ' - Build settings: BLAS_INFO=mkl, ' 'BUILD_TYPE=Release, CUDA_VERSION=12.4, ' '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 -DLIBKINETO_NOXPUPTI=ON ' '-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-strict-overflow -Wno-strict-aliasing ' '-Wno-stringop-overflow -Wsuggest-override ' '-Wno-psabi -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, ' 'TORCH_VERSION=2.5.1, 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': '3.10.15 (main, Oct 3 2024, 07:27:34) [GCC 11.2.0]', 'TorchVision': '0.20.1+cu124', 'lmdeploy': "not installed:No module named 'lmdeploy'", 'numpy_random_seed': 2147483648, 'opencompass': '0.3.9+7f2aeef', 'sys.platform': 'linux', 'transformers': '4.46.3'}
重现问题 - 代码/配置示例
在vllm_qwen2_5_1_5b_instruct.py里面修改代码如下：
from opencompass.models import VLLMwithChatTemplate
models = [
dict(
type=VLLMwithChatTemplate,
# abbr='qwen2.5-1.5b-instruct-vllm',
abbr='DeepSeek-R1-Distill-Qwen-1.5B',
path='/home/lyg/Pth/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B',
# path='/home/lyg/Pth/Qwen/Qwen2.5-Math-1.5B',
model_kwargs=dict(tensor_parallel_size=1, gpu_memory_utilization=0.8),
# max_out_len=4096,
max_out_len=32768,
batch_size=16,
generation_kwargs=dict(temperature=0.6, top_p=0.95),
run_cfg=dict(num_gpus=1),
)
]
重现问题 - 命令或脚本
python run.py --models vllm_qwen2_5_1_5b_instruct --datasets gpqa_gen_4baadb livecodebench_gen
重现问题 - 错误信息
qwen2.5-1.5b-instruct-vllm名称可以忽略不计
其他信息
No response
了解一下，你的vllm版本是哪个呢，低版本的vllm应该不支持DeepSeek-R1-Distill系列模型吧
我使用的是最新的lmdeploy 0.7.1 推理结果也是0
同样问题，看了下predicitons中的内容，看起来reference的内容貌似不对。
{ "0": { "origin_prompt": [ { "role": "HUMAN", "prompt": "### Question:\nYou are given a positive integer N.\r\nIf there are integers x and y such that N=2^x3^y, print Yes; otherwise, print No.\n\nInput\n\nThe input is given from Standard Input in the following format:\nN\n\nOutput\n\nPrint a single line containing Yes if there are integers x and y that satisfy the condition, and No otherwise.\n\nConstraints\n\n\n- 1\\leq N\\leq10^{18}\n- N is an integer.\n\nSample Input 1\n\n324\n\nSample Output 1\n\nYes\r\n\nFor x=2,y=4, we have 2^x3^y=2^23^4=4\\times81=324, so the condition is satisfied.\r\nThus, you should print Yes.\n\nSample Input 2\n\n5\n\nSample Output 2\n\nNo\r\n\nThere are no integers x and y such that 2^x3^y=5.\r\nThus, you should print No.\n\nSample Input 3\n\n32\n\nSample Output 3\n\nYes\r\n\nFor x=5,y=0, we have 2^x3^y=32\\times1=32, so you should print Yes.\n\nSample Input 4\n\n37748736\n\nSample Output 4\n\nYes\n\n### Format: Read the inputs from stdin solve the problem and write the answer to stdout (do not directly test on the sample inputs). Enclose your code within delimiters as follows.\n```python\n# YOUR CODE HERE\n```\n\n### Answer: (use the provided format with backticks)\n\n" } ], "prediction": "\n\n```python\n# Read input\nN = int(input())\n\n# Remove all factors of 2\nx = 0\nwhile N % 2 == 0:\n N = N // 2\n x += 1\n\n# Remove all factors of 3\ny = 0\nwhile N % 3 == 0:\n N = N // 3\n y += 1\n\n# Check if remaining number is 1\nif N == 1:\n print(\"Yes\")\nelse:\n print(\"No\")\n```", "gold": "abc324_b"
