torch.backends.cuda¶

torch.backends.cuda.is_built()[source]¶

Returns whether PyTorch is built with CUDA support. Note that this doesn’t necessarily mean CUDA is available; just that if this PyTorch binary were run a machine with working CUDA drivers and devices, we would be able to use it.

torch.backends.cuda.matmul.allow_tf32¶

A bool that controls whether TensorFloat-32 tensor cores may be used in matrix multiplications on Ampere or newer GPUs. See TensorFloat-32(TF32) on Ampere devices.

torch.backends.cuda.cufft_plan_cache¶

cufft_plan_cache caches the cuFFT plans

size¶

A readonly int that shows the number of plans currently in the cuFFT plan cache.

max_size¶

A int that controls cache capacity of cuFFT plan.

clear()¶

Clears the cuFFT plan cache.

torch.backends.cudnn¶

torch.backends.cudnn.version()[source]¶

Returns the version of cuDNN

torch.backends.cudnn.is_available()[source]¶

Returns a bool indicating if CUDNN is currently available.

torch.backends.cudnn.enabled¶

A bool that controls whether cuDNN is enabled.

torch.backends.cudnn.allow_tf32¶

A bool that controls where TensorFloat-32 tensor cores may be used in cuDNN convolutions on Ampere or newer GPUs. See TensorFloat-32(TF32) on Ampere devices.

torch.backends.cudnn.deterministic¶

A bool that, if True, causes cuDNN to only use deterministic convolution algorithms. See also torch.is_deterministic() and torch.set_deterministic().

torch.backends.cudnn.benchmark¶

A bool that, if True, causes cuDNN to benchmark multiple convolution algorithms and select the fastest.

torch.backends.mkl¶

torch.backends.mkl.is_available()[source]¶

Returns whether PyTorch is built with MKL support.

torch.backends.mkldnn¶

torch.backends.mkldnn.is_available()[source]¶

Returns whether PyTorch is built with MKL-DNN support.

torch.backends.openmp¶

torch.backends.openmp.is_available()[source]¶

Returns whether PyTorch is built with OpenMP support.