Package | Description |
---|---|
org.bytedeco.pytorch | |
org.bytedeco.pytorch.cuda | |
org.bytedeco.pytorch.global |
Modifier and Type | Method and Description |
---|---|
torch.MemoryFormat |
MemoryFormatOptional.get() |
torch.MemoryFormat |
TensorBase.suggest_memory_format() |
torch.MemoryFormat |
TensorBase.suggest_memory_format(boolean channels_last_strides_exact_match) |
torch.MemoryFormat |
IValue.toMemoryFormat() |
Modifier and Type | Method and Description |
---|---|
Tensor |
Tensor.__dispatch_contiguous(torch.MemoryFormat memory_format) |
Tensor |
Tensor.contiguous(torch.MemoryFormat memory_format) |
TensorBase |
TensorBase.contiguous(torch.MemoryFormat memory_format) |
static LongVector |
TensorType.contiguousStridesOf(long[] in_sizes,
torch.MemoryFormat memory_format) |
static LongVector |
TensorType.contiguousStridesOf(LongArrayRef in_sizes,
torch.MemoryFormat memory_format) |
void |
TensorImpl.empty_tensor_restride_symint(torch.MemoryFormat memory_format) |
void |
TensorImpl.empty_tensor_restride(torch.MemoryFormat memory_format)
Set the strides of the tensor to match memory_format
WARNING: This function doesn't rearrange data and assumes tensor is a
memory contiguous
|
TensorMaybeOwned |
Tensor.expect_contiguous(torch.MemoryFormat memory_format)
Should be used if *this can reasonably be expected to be contiguous and
performance is important.
|
boolean |
TensorImpl.is_contiguous_default(torch.MemoryFormat memory_format) |
boolean |
PyInterpreterVTable.is_contiguous(TensorImpl self,
torch.MemoryFormat arg1) |
boolean |
TensorImpl.is_contiguous(torch.MemoryFormat memory_format)
Whether or not a tensor is laid out in contiguous memory.
|
boolean |
TensorBase.is_contiguous(torch.MemoryFormat memory_format) |
boolean |
TensorImpl.is_strides_like_default(torch.MemoryFormat memory_format) |
boolean |
PyInterpreterVTable.is_strides_like(TensorImpl self,
torch.MemoryFormat arg1) |
boolean |
TensorImpl.is_strides_like(torch.MemoryFormat memory_format) |
MemoryFormatOptional |
MemoryFormatOptional.put(torch.MemoryFormat value) |
Constructor and Description |
---|
IValue(torch.MemoryFormat m) |
MemoryFormatOptional(torch.MemoryFormat value) |
TensorOptions(torch.MemoryFormat memory_format)
Constructs a
TensorOptions object with the given memory format. |
Modifier and Type | Method and Description |
---|---|
void |
TensorDescriptor.set(Tensor t,
torch.MemoryFormat memory_format) |
void |
FilterDescriptor.set(Tensor t,
torch.MemoryFormat memory_format) |
void |
TensorDescriptor.set(Tensor t,
torch.MemoryFormat memory_format,
long pad) |
void |
FilterDescriptor.set(Tensor t,
torch.MemoryFormat memory_format,
long pad) |
Modifier and Type | Method and Description |
---|---|
static torch.MemoryFormat |
torch.get_contiguous_memory_format() |
torch.MemoryFormat |
torch.MemoryFormat.intern() |
static torch.MemoryFormat |
torch.MemoryFormat.valueOf(String name)
Returns the enum constant of this type with the specified name.
|
static torch.MemoryFormat[] |
torch.MemoryFormat.values()
Returns an array containing the constants of this enum type, in
the order they are declared.
|
Modifier and Type | Method and Description |
---|---|
static TensorOptions |
torch.memory_format(torch.MemoryFormat memory_format)
Convenience function that returns a
TensorOptions object with the
memory_format set to the given one. |
static Pointer |
torch.shiftLeft(Pointer stream,
torch.MemoryFormat memory_format) |
Copyright © 2024. All rights reserved.