<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[无聊的豆奶]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Marryvivien</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Marryvivien]]></copyright><item><title><![CDATA[R 包管理]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/141897392</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/141897392</guid><author>Marryvivien</author><pubDate>Wed, 04 Sep 2024 16:35:31 +0800</pubDate><description><![CDATA[R 包的安装、更新和加载]]></description><category></category></item><item><title><![CDATA[1. 微积分的本质]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/130829517</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/130829517</guid><author>Marryvivien</author><pubDate>Tue, 23 May 2023 16:25:52 +0800</pubDate><description><![CDATA[2023-05-23。]]></description><category></category></item><item><title><![CDATA[TensorFlow 梯度和自动微分]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/127051767</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/127051767</guid><author>Marryvivien</author><pubDate>Mon, 26 Sep 2022 13:19:37 +0800</pubDate><description><![CDATA[自动微分对训练神经网络所用的反向传播算法十分重要。下面介绍 TensorFlow 计算梯度的方法。]]></description><category></category></item><item><title><![CDATA[TensorFlow 变量]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/127051698</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/127051698</guid><author>Marryvivien</author><pubDate>Mon, 26 Sep 2022 13:11:37 +0800</pubDate><description><![CDATA[TensorFlow 使用变量表示程序中共享、持久化的状态。下面介绍如何创建、更新和管理 TensorFlow 变量。]]></description><category></category></item><item><title><![CDATA[matplotlib pyplot 教程]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/127009420</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/127009420</guid><author>Marryvivien</author><pubDate>Fri, 23 Sep 2022 14:02:15 +0800</pubDate><description><![CDATA[matplotlib.pyplot 包含一系列类似 MATLAB 的绘图函数。]]></description><category></category></item><item><title><![CDATA[matplotlib - 反转坐标轴]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/127006514</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/127006514</guid><author>Marryvivien</author><pubDate>Fri, 23 Sep 2022 10:51:39 +0800</pubDate><description><![CDATA[通过翻转坐标轴的范围设置，使坐标轴从小到大的顺序变为从大到小。]]></description><category></category></item><item><title><![CDATA[TensorFlow 张量]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126986810</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126986810</guid><author>Marryvivien</author><pubDate>Thu, 22 Sep 2022 10:25:53 +0800</pubDate><description><![CDATA[张量（Tensor）是具有统一类型（dtype）的多维数组，功能和 NumPy 的np.arrays类似。]]></description><category></category></item><item><title><![CDATA[TensorFlow 基础]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126973399</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126973399</guid><author>Marryvivien</author><pubDate>Wed, 21 Sep 2022 15:19:24 +0800</pubDate><description><![CDATA[TensorFlow 的基础知识简要概述。]]></description><category></category></item><item><title><![CDATA[通过继承定义 Layer 和 Model]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126966212</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126966212</guid><author>Marryvivien</author><pubDate>Wed, 21 Sep 2022 11:03:35 +0800</pubDate><description><![CDATA[如何通过继承定义 Layer 和 Model]]></description><category></category></item><item><title><![CDATA[numpy.testing.assert_allclose]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126965580</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126965580</guid><author>Marryvivien</author><pubDate>Wed, 21 Sep 2022 09:04:39 +0800</pubDate><description><![CDATA[判断两个对象在指定 tolerance 下是否相等，抛出。]]></description><category></category></item><item><title><![CDATA[ReduceLROnPlateau]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126954678</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126954678</guid><author>Marryvivien</author><pubDate>Tue, 20 Sep 2022 15:42:50 +0800</pubDate><description><![CDATA[当 metric 停止改善时，降低学习率（learning rate, LR）。]]></description><category></category></item><item><title><![CDATA[Keras Sequential 模型]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126954442</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126954442</guid><author>Marryvivien</author><pubDate>Tue, 20 Sep 2022 15:35:26 +0800</pubDate><description><![CDATA[TensorFlow Keras Sequential 模型的使用]]></description><category></category></item><item><title><![CDATA[ModelCheckpoint]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126954192</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126954192</guid><author>Marryvivien</author><pubDate>Tue, 20 Sep 2022 15:26:15 +0800</pubDate><description><![CDATA[以指定频率保存 Keras 模型或权重。]]></description><category></category></item><item><title><![CDATA[EarlyStopping]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126954102</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126954102</guid><author>Marryvivien</author><pubDate>Tue, 20 Sep 2022 15:23:49 +0800</pubDate><description><![CDATA[当某个 metric 不再改善时停止训练。]]></description><category></category></item><item><title><![CDATA[TensorFlow - 自定义 callback]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126953717</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126953717</guid><author>Marryvivien</author><pubDate>Tue, 20 Sep 2022 15:15:41 +0800</pubDate><description><![CDATA[callback 是一个强大的工具，可以在训练、评估和推断期间自定义 Keras 模型的行为。例如，可以使用 TensorBoard 可视化训练进度和结果，可以在训练时定期保存模型。下面介绍介绍什么是 callback，它可以做什么，以及如何创建自己的 callback。]]></description><category></category></item><item><title><![CDATA[tf.compat.v1.global_variables]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126871762</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126871762</guid><author>Marryvivien</author><pubDate>Thu, 15 Sep 2022 15:01:36 +0800</pubDate><description><![CDATA[返回 global 变量 list。global 变量在分布式环境中跨机器共享。Variable()构造函数或自动将新变量添加到 graph 集合，而本函数返回该集合的内容。]]></description><category></category></item><item><title><![CDATA[tf.compat.v1.local_variables]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/126870559</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/126870559</guid><author>Marryvivien</author><pubDate>Thu, 15 Sep 2022 14:40:01 +0800</pubDate><description><![CDATA[返回 local 变量 list。local 变量存在于单个进程中，通常不保存到 checkpoint，用于临时保存值。例如，local 变量可用来保存 epoch 数。函数自动将新的变量添加到集合，而本函数返回该集合内容。]]></description><category></category></item><item><title><![CDATA[TensorFlow 基础]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/122046308</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/122046308</guid><author>Marryvivien</author><pubDate>Mon, 20 Dec 2021 17:32:49 +0800</pubDate><description><![CDATA[TensorFlow 基础知识总体概览。]]></description><category></category></item><item><title><![CDATA[CIFAR 数据集]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/121953755</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/121953755</guid><author>Marryvivien</author><pubDate>Wed, 15 Dec 2021 15:36:45 +0800</pubDate><description><![CDATA[简介
CIFAR-10 和 CIFAR-100 是一个包含 8,000 万张微型图像数据集的带标签子集。由 Alex Krizhevsky, Vinod Nair 和 Geoffrey Hinton 收集整理。
CIFAR-10
CIFAR-10 数据集包含 60,000 张 32x32 彩色图片，分为 10 个类别，每个类别 6,000 张。分为训练图片 50,000 张，测试图片 10,000 张。
该数据集分为 5 个训练 batch 和 1 个测试 batch，每个 batch 包含 10,000 ]]></description><category></category></item><item><title><![CDATA[numpy - 数组随机排序]]></title><link>https://blog.youkuaiyun.com/Marryvivien/article/details/113812077</link><guid>https://blog.youkuaiyun.com/Marryvivien/article/details/113812077</guid><author>Marryvivien</author><pubDate>Sun, 14 Feb 2021 23:04:45 +0800</pubDate><description><![CDATA[数组随机排序函数：
random.Generator.permutation(x, axis=0)

随机排列一个序列，或者返回一个排列过的范围。

x: int or array_like

如果 x 为整数，则随机排列 np.arange(x)。如果 x 是数组，则复制数组，并随机排列数组元素。

axis: int, optional

随机排列的轴，默认为 0.
例1，生成指定范围内数值的随机排序：
rng = np.random.default_rng()
a1 = rng.permutation]]></description><category></category></item></channel></rss>