torch.nn.utils.rnn.PackedSequence Class Reference

Inheritance diagram for torch.nn.utils.rnn.PackedSequence:

Public Member Functions
def	__new__ (cls, data, batch_sizes=None, sorted_indices=None, unsorted_indices=None)
def	pin_memory (self)
def	cuda (self, args, kwargs)
def	cpu (self)
def	double (self)
def	float (self)
def	half (self)
def	long (self)
def	int (self)
def	short (self)
def	char (self)
def	byte (self)
def	to (self, args, kwargs)
def	is_cuda (self)
def	is_pinned (self)

Detailed Description

Definition at line 17 of file rnn.py.

Member Function Documentation

def torch.nn.utils.rnn.PackedSequence.cpu

(

self

)

Returns a CPU copy if `self.data` not already on the CPU

Definition at line 100 of file rnn.py.

def torch.nn.utils.rnn.PackedSequence.cuda	(		self,
			args,
			kwargs
	)

Returns a GPU copy if `self.data` not already on the GPU

Definition at line 89 of file rnn.py.

---

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

• torch/nn/utils/rnn.py