data_juicer.utils.resource_utils module¶

data_juicer.utils.resource_utils.query_cuda_info(query_key)[源代码]¶

data_juicer.utils.resource_utils.get_cpu_utilization()[源代码]¶

data_juicer.utils.resource_utils.query_mem_info(query_key)[源代码]¶

data_juicer.utils.resource_utils.cuda_device_count(cfg=None)[源代码]¶

data_juicer.utils.resource_utils.is_cuda_available()[源代码]¶

data_juicer.utils.resource_utils.cpu_count(cfg=None)[源代码]¶

data_juicer.utils.resource_utils.available_memories(cfg=None) → List[int][源代码]¶: Available memory for each node in MB.

data_juicer.utils.resource_utils.available_gpu_memories(cfg=None) → List[int][源代码]¶: Available gpu memory of each gpu card for each alive node in MB.