.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "beginner/basics/transforms_tutorial.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        Click :ref:`here <sphx_glr_download_beginner_basics_transforms_tutorial.py>`
        to download the full example code

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_beginner_basics_transforms_tutorial.py:


`基础知识 <intro.html>`_ ||
`快速入门 <quickstart_tutorial.html>`_ ||
`张量 <tensorqs_tutorial.html>`_ ||
`数据集与数据加载器 <data_tutorial.html>`_ ||
**Transforms** ||
`构建神经网络 <buildmodel_tutorial.html>`_ ||
`自动微分 <autogradqs_tutorial.html>`_ ||
`优化模型参数 <optimization_tutorial.html>`_ ||
`保存和加载模型 <saveloadrun_tutorial.html>`_

Transforms
==========

数据并不总是以训练机器学习算法所需的最终处理形式呈现。我们使用**transforms**来对数据进行一些处理，使其适用于训练。

所有 TorchVision 数据集都有两个参数 - `transform` 用于修改特征，`target_transform` 用于修改标签 
- 它们接受包含转换逻辑的可调用对象。`torchvision.transforms <https://pytorch.org/vision/stable/transforms.html>`_ 模块提供了几种常用的转换。

FashionMNIST 的特征是以 PIL 图像格式呈现的，标签是整数。对于训练，我们需要将特征转换为归一化的张量，
将标签转换为编码的张量。为了进行这些转换，我们使用了 ``ToTensor`` 和 ``Lambda``。

.. GENERATED FROM PYTHON SOURCE LINES 23-37

.. code-block:: default


    import torch
    from torchvision import datasets
    from torchvision.transforms import ToTensor, Lambda

    ds = datasets.FashionMNIST(
        root="data",
        train=True,
        download=True,
        transform=ToTensor(),
        target_transform=Lambda(lambda y: torch.zeros(
            10, dtype=torch.float).scatter_(0, torch.tensor(y), value=1))
    )


.. GENERATED FROM PYTHON SOURCE LINES 38-43

ToTensor()
-------------------------------

`ToTensor <https://pytorch.org/vision/stable/transforms.html#torchvision.transforms.ToTensor>`_
将 PIL 图像或 NumPy ``ndarray`` 转换为 ``FloatTensor``，并将图像的像素强度值缩放到范围 [0., 1.]。

.. GENERATED FROM PYTHON SOURCE LINES 45-51

Lambda Transforms
-------------------------------

Lambda transforms 应用任何用户定义的 lambda 函数。这里，我们定义一个函数将整数转换为独热编码的张量。
它首先创建一个大小为 10（我们数据集中标签的数量）的零张量，然后调用 `scatter_ <https://pytorch.org/docs/stable/generated/torch.Tensor.scatter_.html>`_，
在由标签 ``y`` 指定的索引上赋值为 ``1``。

.. GENERATED FROM PYTHON SOURCE LINES 51-55

.. code-block:: default


    target_transform = Lambda(lambda y: torch.zeros(
        10, dtype=torch.float).scatter_(dim=0, index=torch.tensor(y), value=1))


.. GENERATED FROM PYTHON SOURCE LINES 56-58

--------------


.. GENERATED FROM PYTHON SOURCE LINES 60-63

延伸阅读
~~~~~~~~~~~~~~~~~
- `torchvision.transforms API <https://pytorch.org/vision/stable/transforms.html>`_


.. rst-class:: sphx-glr-timing

   **Total running time of the script:** ( 0 minutes  0.000 seconds)


.. _sphx_glr_download_beginner_basics_transforms_tutorial.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example


    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: transforms_tutorial.py <transforms_tutorial.py>`

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: transforms_tutorial.ipynb <transforms_tutorial.ipynb>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_