dlpy.model.Gpu

class dlpy.model.Gpu(devices=None, use_tensor_rt=False, precision='fp32', use_exclusive=False)

Bases: dlpy.utils.DLPyDict

Gpu parameters object.

Parameters:

devices : list-of-ints, optional

Specifies a list of GPU devices to be used.

use_tensor_rt : bool, optional

Enables using TensorRT for fast inference.
Default: False.

precision : string, optional

Specifies the experimental option to incorporate lower computational precision in forward-backward computations to potentially engage tensor cores.
Valid Values: FP32, FP16
Default: FP32

use_exclusive : bool, optional

Specifies exclusive use of GPU devices.
Default: False

Returns:

Gpu

__init__(devices=None, use_tensor_rt=False, precision='fp32', use_exclusive=False)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__([devices, use_tensor_rt, …])	Initialize self.
clear()
get(k[,d])
items()
keys()
pop(k[,d])	If key is not found, d is returned if given, otherwise KeyError is raised.
popitem()	as a 2-tuple; but raise KeyError if D is empty.
setdefault(k[,d])
update([E, ]**F)	If E present and has a .keys() method, does: for k in E: D[k] = E[k] If E present and lacks .keys() method, does: for (k, v) in E: D[k] = v In either case, this is followed by: for k, v in F.items(): D[k] = v
values()