Rate this Page
★★★★★
torch.accelerator.device_count#
- torch.accelerator.device_count()[source]#
Return the number of currentaccelerator available.
- Returns
- the number of the currentaccelerator available.
If there is no available accelerators, return 0.
- Return type
Note
This API delegates to the device-specific version ofdevice_count.On CUDA, this API will NOT poison fork if NVML discovery succeeds.Otherwise, it will. For more details, seePoison fork in multiprocessing.
On this page
