Rate this Page
★★★★★

torch.cuda.utilization#

torch.cuda.utilization(device=None)[source]#

Return the percent of time over the past sample period during which one ormore kernels was executing on the GPU as given bynvidia-smi.

Parameters

device (torch.device orint,optional) – selected device. Returnsstatistic for the current device, given bycurrent_device(),ifdevice isNone (default).

Return type

int

Warning: Each sample period may be between 1 second and 1/6 second,depending on the product being queried.