saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-15 21:00:47 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[docs] Add docs for storage and tensors for quantized Tensor (#51817)

Browse source 
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/51817

Test Plan: Imported from OSS

Reviewed By: supriyar

Differential Revision: D26292464

Pulled By: jerryzh168

fbshipit-source-id: c5992deda4af949de4ea2e40edee8f22bd59b9e1
This commit is contained in:
Jerry Zhang 2021-02-09 13:18:46 -08:00 committed by Facebook GitHub Bot
parent fc314350ad
commit 0ec00c1292
2 changed files with 99 additions and 17 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

76
docs/source/storage.rst
View file

@ -6,7 +6,83 @@ data type.
Every :class:`torch.Tensor` has a corresponding storage of the same data type.
.. autoclass:: torch.DoubleStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.FloatStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.HalfStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.LongStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.IntStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.ShortStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.CharStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.ByteStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.BoolStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.BFloat16Storage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.ComplexDoubleStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.ComplexFloatStorage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.QUInt8Storage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.QInt8Storage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.QInt32Storage
:members:
:undoc-members:
:inherited-members:
.. autoclass:: torch.QUInt4x2Storage
:members:
:undoc-members:
:inherited-members:

40
docs/source/tensors.rst
View file

@ -10,23 +10,27 @@ a single data type.
Torch defines 10 tensor types with CPU and GPU variants which are as follows:
========================== =========================================== ============================= ================================
Data type dtype CPU tensor GPU tensor
========================== =========================================== ============================= ================================
32-bit floating point ``torch.float32`` or ``torch.float`` :class:`torch.FloatTensor` :class:`torch.cuda.FloatTensor`
64-bit floating point ``torch.float64`` or ``torch.double`` :class:`torch.DoubleTensor` :class:`torch.cuda.DoubleTensor`
16-bit floating point [1]_ ``torch.float16`` or ``torch.half`` :class:`torch.HalfTensor` :class:`torch.cuda.HalfTensor`
16-bit floating point [2]_ ``torch.bfloat16`` :class:`torch.BFloat16Tensor` :class:`torch.cuda.BFloat16Tensor`
32-bit complex ``torch.complex32``
64-bit complex ``torch.complex64``
128-bit complex ``torch.complex128`` or ``torch.cdouble``
8-bit integer (unsigned) ``torch.uint8`` :class:`torch.ByteTensor` :class:`torch.cuda.ByteTensor`
8-bit integer (signed) ``torch.int8`` :class:`torch.CharTensor` :class:`torch.cuda.CharTensor`
16-bit integer (signed) ``torch.int16`` or ``torch.short`` :class:`torch.ShortTensor` :class:`torch.cuda.ShortTensor`
32-bit integer (signed) ``torch.int32`` or ``torch.int`` :class:`torch.IntTensor` :class:`torch.cuda.IntTensor`
64-bit integer (signed) ``torch.int64`` or ``torch.long`` :class:`torch.LongTensor` :class:`torch.cuda.LongTensor`
Boolean ``torch.bool`` :class:`torch.BoolTensor` :class:`torch.cuda.BoolTensor`
========================== =========================================== ============================= ================================
======================================= =========================================== ============================= ================================
Data type dtype CPU tensor GPU tensor
======================================= =========================================== ============================= ================================
32-bit floating point ``torch.float32`` or ``torch.float`` :class:`torch.FloatTensor` :class:`torch.cuda.FloatTensor`
64-bit floating point ``torch.float64`` or ``torch.double`` :class:`torch.DoubleTensor` :class:`torch.cuda.DoubleTensor`
16-bit floating point [1]_ ``torch.float16`` or ``torch.half`` :class:`torch.HalfTensor` :class:`torch.cuda.HalfTensor`
16-bit floating point [2]_ ``torch.bfloat16`` :class:`torch.BFloat16Tensor` :class:`torch.cuda.BFloat16Tensor`
32-bit complex ``torch.complex32``
64-bit complex ``torch.complex64``
128-bit complex ``torch.complex128`` or ``torch.cdouble``
8-bit integer (unsigned) ``torch.uint8`` :class:`torch.ByteTensor` :class:`torch.cuda.ByteTensor`
8-bit integer (signed) ``torch.int8`` :class:`torch.CharTensor` :class:`torch.cuda.CharTensor`
16-bit integer (signed) ``torch.int16`` or ``torch.short`` :class:`torch.ShortTensor` :class:`torch.cuda.ShortTensor`
32-bit integer (signed) ``torch.int32`` or ``torch.int`` :class:`torch.IntTensor` :class:`torch.cuda.IntTensor`
64-bit integer (signed) ``torch.int64`` or ``torch.long`` :class:`torch.LongTensor` :class:`torch.cuda.LongTensor`
Boolean ``torch.bool`` :class:`torch.BoolTensor` :class:`torch.cuda.BoolTensor`
quantized 8-bit integer (unsigned) ``torch.quint8`` :class:`torch.ByteTensor` /
quantized 8-bit integer (signed) ``torch.qint8`` :class:`torch.CharTensor` /
quantized 32-bit integer (signed) ``torch.qfint32`` :class:`torch.IntTensor` /
quantized 4-bit integer (unsigned) [3]_ ``torch.quint4x2`` :class:`torch.ByteTensor` /
======================================= =========================================== ============================= ================================
.. [1]
Sometimes referred to as binary16: uses 1 sign, 5 exponent, and 10
@ -35,6 +39,8 @@ Boolean ``torch.bool`` :class:
Sometimes referred to as Brain Floating Point: uses 1 sign, 8 exponent, and 7
significand bits. Useful when range is important, since it has the same
number of exponent bits as ``float32``
.. [3]
quantized 4-bit integer is stored as a 8-bit signed integer. Currently it's only supported in EmbeddingBag operator.
:class:`torch.Tensor` is an alias for the default tensor type (:class:`torch.FloatTensor`).