OpenKE.module.model.TransD.TransD Class Reference

Inheritance diagram for OpenKE.module.model.TransD.TransD:

Collaboration diagram for OpenKE.module.model.TransD.TransD:

Public Member Functions
	__init__ (self, ent_tot, rel_tot, dim_e=100, dim_r=100, p_norm=1, norm_flag=True, margin=None, epsilon=None)
	forward (self, data)
	regularization (self, data)
	predict (self, data)
Public Member Functions inherited from OpenKE.module.model.Model.Model
Public Member Functions inherited from OpenKE.module.BaseModule.BaseModule
	load_checkpoint (self, path)
	save_checkpoint (self, path)
	load_parameters (self, path)
	save_parameters (self, path)
	get_parameters (self, mode="numpy", param_dict=None)
	set_parameters (self, parameters)

Public Attributes
	dim_e = dim_e
	dim_r = dim_r
	margin = margin
	epsilon = epsilon
	norm_flag = norm_flag
	p_norm = p_norm
	ent_embeddings = nn.Embedding(self.ent_tot, self.dim_e)
	rel_embeddings = nn.Embedding(self.rel_tot, self.dim_r)
	ent_transfer = nn.Embedding(self.ent_tot, self.dim_e)
	rel_transfer = nn.Embedding(self.rel_tot, self.dim_r)
	ent_embedding_range
	rel_embedding_range
bool	margin = False
bool	margin_flag = True
Public Attributes inherited from OpenKE.module.model.Model.Model
	ent_tot = ent_tot
	rel_tot = rel_tot
Public Attributes inherited from OpenKE.module.BaseModule.BaseModule
	zero_const = nn.Parameter(torch.Tensor([0]))
	pi_const = nn.Parameter(torch.Tensor([3.14159265358979323846]))

Protected Member Functions
	_resize (self, tensor, axis, size)
	_calc (self, h, t, r, mode)
	_transfer (self, e, e_transfer, r_transfer)

Constructor & Destructor Documentation

__init__()

OpenKE.module.model.TransD.TransD.__init__	(		self,
			ent_tot,
			rel_tot,
			dim_e = 100,
			dim_r = 100,
			p_norm = 1,
			norm_flag = True,
			margin = None,
			epsilon = None )

Reimplemented from OpenKE.module.model.Model.Model.

Member Function Documentation

forward()

OpenKE.module.model.TransD.TransD.forward	(		self,
			data )

Reimplemented from OpenKE.module.model.Model.Model.

predict()

OpenKE.module.model.TransD.TransD.predict	(		self,
			data )

Reimplemented from OpenKE.module.model.Model.Model.

Member Data Documentation

ent_embedding_range

OpenKE.module.model.TransD.TransD.ent_embedding_range

Initial value:

=  nn.Parameter(
                torch.Tensor([(self.margin + self.epsilon) / self.dim_e]),
                requires_grad=False,
            )

rel_embedding_range

OpenKE.module.model.TransD.TransD.rel_embedding_range

Initial value:

=  nn.Parameter(
                torch.Tensor([(self.margin + self.epsilon) / self.dim_r]),
                requires_grad=False,
            )

---

The documentation for this class was generated from the following file:

• seed_embeddings/OpenKE/module/model/TransD.py