I followed the installation instructions and encountered the following error.

[lldhsds]

(opencompass) root@intern-studio-50014188:~/opencompass# python run.py --datasets ceval_gen --hf-path /root/model/Meta-Llama-3-8B-Instruct --tokenizer-path /root/model/Meta-Llama-3-8B-Instruct --tokenizer-kwargs padding_side='left' truncation='left' trust_remote_code=True --model-kwargs trust_remote_code=True device_map='auto' --max-seq-len 2048 --max-out-len 16 --batch-size 4 --num-gpus 1 --debug
05/10 20:43:46 - OpenCompass - INFO - Loading ceval_gen: configs/datasets/ceval/ceval_gen.py
05/10 20:43:46 - OpenCompass - INFO - Loading example: configs/summarizers/example.py
05/10 20:43:47 - OpenCompass - WARNING - SlurmRunner is not used, so the partition argument is ignored.
05/10 20:43:47 - OpenCompass - DEBUG - Modules of opencompass's partitioner registry have been automatically imported from opencompass.partitioners
05/10 20:43:47 - OpenCompass - DEBUG - Get class `SizePartitioner` from "partitioner" registry in "opencompass"
05/10 20:43:47 - OpenCompass - DEBUG - An `SizePartitioner` instance is built from registry, and its implementation can be found in opencompass.partitioners.size
05/10 20:43:47 - OpenCompass - DEBUG - Key eval.runner.task.judge_cfg not found in config, ignored.
05/10 20:43:47 - OpenCompass - DEBUG - Key eval.runner.task.dump_details not found in config, ignored.
05/10 20:43:47 - OpenCompass - DEBUG - Key eval.given_pred not found in config, ignored.
05/10 20:43:47 - OpenCompass - DEBUG - Additional config: {}
05/10 20:43:47 - OpenCompass - INFO - Partitioned into 1 tasks.
05/10 20:43:47 - OpenCompass - DEBUG - Task 0: [opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-college_economics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-accountant,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-tax_accountant,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-physician,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-civil_servant,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-urban_and_rural_planner,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-teacher_qualification,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-college_programming,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-electrical_engineer,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-business_administration,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-art_studies,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-fire_engineer,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-environmental_impact_assessment_engineer,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-education_science,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-professional_tour_guide,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-college_chemistry,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-metrology_engineer,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-mao_zedong_thought,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-law,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-veterinary_medicine,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-modern_chinese_history,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-chinese_language_and_literature,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-legal_professional,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-logic,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-middle_school_history,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-plant_protection,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-clinical_medicine,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-computer_architecture,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-middle_school_biology,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-middle_school_politics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-middle_school_chemistry,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_history,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-computer_network,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-operating_system,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-college_physics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-advanced_mathematics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_physics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_chemistry,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_biology,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-middle_school_mathematics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-middle_school_physics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-marxism,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_politics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_geography,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-ideological_and_moral_cultivation,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_chinese,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-sports_science,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-basic_medicine,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-probability_and_statistics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-high_school_mathematics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-discrete_mathematics,opencompass.models.huggingface.HuggingFace_meta-llama_Meta-Llama-3-8B-Instruct/ceval-middle_school_geography]
05/10 20:43:47 - OpenCompass - DEBUG - Modules of opencompass's runner registry have been automatically imported from opencompass.runners
05/10 20:43:47 - OpenCompass - DEBUG - Get class `LocalRunner` from "runner" registry in "opencompass"
05/10 20:43:47 - OpenCompass - DEBUG - An `LocalRunner` instance is built from registry, and its implementation can be found in opencompass.runners.local
05/10 20:43:47 - OpenCompass - DEBUG - Modules of opencompass's task registry have been automatically imported from opencompass.tasks
05/10 20:43:47 - OpenCompass - DEBUG - Get class `OpenICLInferTask` from "task" registry in "opencompass"
05/10 20:43:47 - OpenCompass - DEBUG - An `OpenICLInferTask` instance is built from registry, and its implementation can be found in opencompass.tasks.openicl_infer
Traceback (most recent call last):
  File "/root/opencompass/run.py", line 4, in <module>
    main()
  File "/root/opencompass/opencompass/cli/main.py", line 302, in main
    runner(tasks)
  File "/root/opencompass/opencompass/runners/base.py", line 38, in __call__
    status = self.launch(tasks)
  File "/root/opencompass/opencompass/runners/local.py", line 88, in launch
    assert len(all_gpu_ids) >= num_gpus
AssertionError

The env Info:

(opencompass) root@intern-studio-50014188:~/opencompass#  nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Tue_Aug_15_22:02:13_PDT_2023
Cuda compilation tools, release 12.2, V12.2.140
Build cuda_12.2.r12.2/compiler.33191640_0

(opencompass) root@intern-studio-50014188:~/opencompass# nvidia-smi 
Fri May 10 21:00:28 2024       
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.54.03              Driver Version: 535.54.03    CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA A100-SXM4-80GB          On  | 00000000:48:00.0 Off |                    0 |
| N/A   34C    P0              65W / 400W |      4MiB / 81920MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
                                                                                         
+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|  No running processes found                                                           |
+---------------------------------------------------------------------------------------+

[JacquelineXu]

检查下python的torch版本对应的cuda，https://pytorch.org/get-started/previous-versions/
我的是cu11.7，当时也报了这个错，然后按照对应版本重装了下就好了。12.2版本的不了解，实在不行只能换下cuda版本了？
验证是否成功的方式，直接python进命令行模式，试一下
import torch
torch.cuda.device_count()
显示1就是好了=.=