- Sort Score
- Result 10 results
- Languages All
Results 1 - 3 of 3 for CanUseTensorCores (0.17 sec)
-
tensorflow/compiler/mlir/tensorflow/ir/tf_ops_device_helper.h
bool CanUseGpuDevice(const RuntimeDevices &devices); // Returns true if all of the GPUs available at runtime support TensorCores // (NVIDIA compute capability >= 7.0). bool CanUseTensorCores(const RuntimeDevices &devices); // Returns true if operation does not have explicit device placement that would // prevent it from running on GPU device. bool CanUseGpuDevice(Operation *op); } // namespace TF
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri Nov 12 21:57:12 UTC 2021 - 1.4K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/ir/tf_ops_device_helper.cc
return llvm::any_of(devices.device_names(), IsGpuDevice); } // Returns true if all of the GPUs available at runtime support TensorCores // (NVIDIA compute capability >= 7.0). bool CanUseTensorCores(const RuntimeDevices &devices) { auto has_tensor_cores = [&](const DeviceNameUtils::ParsedName &device) { auto md = devices.GetGpuDeviceMetadata(device); return md ? md->getCcMajor() >= 7 : false; };
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Jun 21 08:41:18 UTC 2022 - 2.4K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/ir/tf_ops_a_m.cc
// For f16 data type on devices with Tensor Cores support NHWC data format // is up to ~2x faster. const bool is_f16 = input_ty.getElementType().isF16(); if (is_f16 && CanUseTensorCores(devices)) return "NHWC"; // For f32/f16 data type decision depends on the filter size in spatial // dimensions, for other data types we keep current data format.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 146.7K bytes - Viewed (0)