<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Coder]]></title><description><![CDATA[]]></description><link>https://blog.csdn.net/AnimateX</link><language>zh-cn</language><generator>https://blog.csdn.net/</generator><copyright><![CDATA[Copyright &copy; AnimateX]]></copyright><item><title><![CDATA[[PyQt] Pycharm 配置 PyQt 开发环境]]></title><link>https://blog.csdn.net/AnimateX/article/details/127108043</link><guid>https://blog.csdn.net/AnimateX/article/details/127108043</guid><author>AnimateX</author><pubDate>Thu, 29 Sep 2022 15:41:08 +0800</pubDate><description><![CDATA[pycharm 配置pyqt5 external tool]]></description><category></category></item><item><title><![CDATA[[valgrind] 安装与使用]]></title><link>https://blog.csdn.net/AnimateX/article/details/126040604</link><guid>https://blog.csdn.net/AnimateX/article/details/126040604</guid><author>AnimateX</author><pubDate>Thu, 28 Jul 2022 17:54:35 +0800</pubDate><description><![CDATA[valgrind 安装与使用]]></description><category></category></item><item><title><![CDATA[C99 designator ‘name’ outside aggregate initializer]]></title><link>https://blog.csdn.net/AnimateX/article/details/125777499</link><guid>https://blog.csdn.net/AnimateX/article/details/125777499</guid><author>AnimateX</author><pubDate>Thu, 14 Jul 2022 09:34:56 +0800</pubDate><description><![CDATA[C99 designator ‘xxx’ outside aggregate initializer]]></description><category></category></item><item><title><![CDATA[彻底卸载VSC]]></title><link>https://blog.csdn.net/AnimateX/article/details/125095973</link><guid>https://blog.csdn.net/AnimateX/article/details/125095973</guid><author>AnimateX</author><pubDate>Thu, 02 Jun 2022 10:21:05 +0800</pubDate><description><![CDATA[完整卸载vscode，包含插件和个人信息]]></description><category></category></item><item><title><![CDATA[CMD打开闪退问题修复]]></title><link>https://blog.csdn.net/AnimateX/article/details/125095851</link><guid>https://blog.csdn.net/AnimateX/article/details/125095851</guid><author>AnimateX</author><pubDate>Thu, 02 Jun 2022 10:16:54 +0800</pubDate><description><![CDATA[cmd闪退修复
VScode remote ssh 连接失败]]></description><category></category></item><item><title><![CDATA[Source Code Pro l l 显示异常修复]]></title><link>https://blog.csdn.net/AnimateX/article/details/123048378</link><guid>https://blog.csdn.net/AnimateX/article/details/123048378</guid><author>AnimateX</author><pubDate>Mon, 21 Feb 2022 15:28:28 +0800</pubDate><description><![CDATA[VSC 中等宽字体显示 l l异常修复
"editor.fontLigatures": "'ccmp' off",


]]></description><category></category></item><item><title><![CDATA[PyTorch 深度学习实践-06-[Logistic Regression]]]></title><link>https://blog.csdn.net/AnimateX/article/details/122184414</link><guid>https://blog.csdn.net/AnimateX/article/details/122184414</guid><author>AnimateX</author><pubDate>Tue, 28 Dec 2021 07:54:49 +0800</pubDate><description><![CDATA[Date: 2021-12-27
Repositity:  Gitee
本节引出分类问题，使用模型为Logistic regression。使用数据集为：MNIST手写数字数据集。
0. MNIST Dataset
关于MNIST数据集有：

训练集大小：60000张手写样本；
测试集大小：10000张手些样本；
共计10个类别：{0, 1, 2, 3, 4, 5, 6, 7, 8, 9}

因此我们需要区分出来 y∈{0,&nbsp;1,&nbsp;2,&nbsp;3,&nbsp;4,&nbsp;5,6,]]></description><category></category></item><item><title><![CDATA[PyTorch 深度学习实践-05-[Linear Regression with PyTorch]]]></title><link>https://blog.csdn.net/AnimateX/article/details/122161064</link><guid>https://blog.csdn.net/AnimateX/article/details/122161064</guid><author>AnimateX</author><pubDate>Sun, 26 Dec 2021 21:21:38 +0800</pubDate><description><![CDATA[Date: 2021-12-20
Repositity:  Gitee

0. 回顾
前述模型为： y^=x∗ω+b\hat{y} = x * \omega + by^​=x∗ω+b ，损失函数为MSE: loss=(x∗ω−y)2loss = (x*\omega - y)^2loss=(x∗ω−y)2 ，优化器为梯度下降GD。
for epoch in range(100):
    # Use Tensor: x_data, y_dta
    for x, y in zip(x_data, y_da.]]></description><category></category></item><item><title><![CDATA[PyTorch 深度学习实践-04-[Back Propagation]]]></title><link>https://blog.csdn.net/AnimateX/article/details/122075999</link><guid>https://blog.csdn.net/AnimateX/article/details/122075999</guid><author>AnimateX</author><pubDate>Wed, 22 Dec 2021 00:01:41 +0800</pubDate><description><![CDATA[Date: 2021-12-20
Repositity:  Gitee

0. 前言

Reference: WIKI

反向传播（英语：Back Propagation，缩写为BP）是“误差反向传播”的简称，是一种与最优化方法（如梯度下降法）结合使用的，用来训练人工神经网络的常见方法。该方法对网络中所有权重计算损失函数的梯度。这个梯度会回馈给最佳化方法，用来更新权值以最小化损失函数。
反向传播要求对每个输入值得到已知输出，来计算损失函数梯度。因此，它通常被认为是一种监督式学习方法，虽然它也用在一些无监督.]]></description><category></category></item><item><title><![CDATA[Pytorch深度学习实践-03-[Gradient Descent]]]></title><link>https://blog.csdn.net/AnimateX/article/details/122027277</link><guid>https://blog.csdn.net/AnimateX/article/details/122027277</guid><author>AnimateX</author><pubDate>Sun, 19 Dec 2021 18:06:50 +0800</pubDate><description><![CDATA[梯度下降]]></description><category></category></item><item><title><![CDATA[Pytorch 深度学习实践-02-[Linear Model]]]></title><link>https://blog.csdn.net/AnimateX/article/details/122016419</link><guid>https://blog.csdn.net/AnimateX/article/details/122016419</guid><author>AnimateX</author><pubDate>Sat, 18 Dec 2021 20:14:50 +0800</pubDate><description><![CDATA[Date: 2021-12-17
Repositity:  Gitee

0. 前言
通常深度学习模型的训练流程可以简单分为以下4个部分：

准备数据集（包含预处理）
选择或设计模型（包含loss和优化器）
训练（找到最佳的模型参数）
推理（部署）

接下来从基础的线性模型入手展开。
1. 数据集注意事项
这里以学习时间和成绩作为样本，考虑篇幅这里仅用文字描述。
""" Dataset """
# hours
[1, 2, 3, 4]
# points
[2, 4, 6, ?]

如上所示，这里给出数据集.]]></description><category></category></item><item><title><![CDATA[Ubuntu 查看CPU信息]]></title><link>https://blog.csdn.net/AnimateX/article/details/121856286</link><guid>https://blog.csdn.net/AnimateX/article/details/121856286</guid><author>AnimateX</author><pubDate>Fri, 10 Dec 2021 14:49:07 +0800</pubDate><description><![CDATA[# see all cpu-core info
cat /proc/cpuinfo

# see cpu num
more /proc/cpuinfo | grep "physical id" | uniq | wc -l
cat /proc/cpuinfo | grep "physical id" | sort | uniq | wc -l

# see core num
more /proc/cpuinfo | grep "physical id" | grep "0" | wc -l
cat /pro]]></description><category></category></item><item><title><![CDATA[loss曲线本地动态显示并保存成gif]]></title><link>https://blog.csdn.net/AnimateX/article/details/121844868</link><guid>https://blog.csdn.net/AnimateX/article/details/121844868</guid><author>AnimateX</author><pubDate>Thu, 09 Dec 2021 21:52:35 +0800</pubDate><description><![CDATA[import os
import torch
import imageio
import numpy as np
import matplotlib.pyplot as plt
from tqdm import tqdm
from natsort import natsorted

Vector = [torch.Tensor, torch.Tensor]


def load_diabetes_data(csv_file_path: str, delim: str, data_type=np.float3]]></description><category></category></item><item><title><![CDATA[VSCode RemoteSSH 过程试图写入的管道不存在问题 解决]]></title><link>https://blog.csdn.net/AnimateX/article/details/121658165</link><guid>https://blog.csdn.net/AnimateX/article/details/121658165</guid><author>AnimateX</author><pubDate>Wed, 01 Dec 2021 16:34:27 +0800</pubDate><description><![CDATA[baron@172.168.x.x: Permission denied (publickey,password).
 过程试图写入的管道不存在问题解决

然后我依次做了以下行为，当然初最后一项都是没有用的！
1. 修改系统OpenSSH Path
屏蔽win下的OpenSSH路径，使用Git bash的Openssh.exe，对应修改环境变量
	C:\Program Files\Git\usr\bin

搞定后测试，没屌用。
2. 清理known_hosts对应IP内容
清理目标服务器对应的known_h]]></description><category></category></item><item><title><![CDATA[Python 数据结构之字典]]></title><link>https://blog.csdn.net/AnimateX/article/details/120632788</link><guid>https://blog.csdn.net/AnimateX/article/details/120632788</guid><author>AnimateX</author><pubDate>Thu, 07 Oct 2021 11:04:36 +0800</pubDate><description><![CDATA[Python-dictionary

[Reference]
[1] https://docs.python.org/3/library/stdtypes.html#typesmapping

1. dict 格式和特性

A mapping object maps hashable values to arbitrary objects. Mappings are mutable objects. There is currently only one standard mapping type, the]]></description><category></category></item><item><title><![CDATA[PyTorch TORCH.AS_STRIDED]]></title><link>https://blog.csdn.net/AnimateX/article/details/120596435</link><guid>https://blog.csdn.net/AnimateX/article/details/120596435</guid><author>AnimateX</author><pubDate>Sun, 03 Oct 2021 17:18:26 +0800</pubDate><description><![CDATA[torch.as_strided(input, size, stride, storage_offset=0)→ Tensor

1. 官方文档
Create a view of an existing torch.Tensor input with specified size, stride and storage_offset.

parameters


input (Tensor) – the input tensor.
size (tuple or ints) – the shape of t.]]></description><category></category></item><item><title><![CDATA[Pytorch torch.sparse_coo_tensor()]]></title><link>https://blog.csdn.net/AnimateX/article/details/120512130</link><guid>https://blog.csdn.net/AnimateX/article/details/120512130</guid><author>AnimateX</author><pubDate>Mon, 27 Sep 2021 17:03:39 +0800</pubDate><description><![CDATA[Pytorch 构造稀疏 Tensor

torch.sparse_coo_tensor(indices, values,
					    size=None, *,
					    dtype=None,
					    device=None,
					    requires_grad=False)


Constructs a sparse tensor in COO(rdinate) format with specified values at the given indices.
T.]]></description><category></category></item><item><title><![CDATA[python map与reduce]]></title><link>https://blog.csdn.net/AnimateX/article/details/120423186</link><guid>https://blog.csdn.net/AnimateX/article/details/120423186</guid><author>AnimateX</author><pubDate>Wed, 22 Sep 2021 21:13:26 +0800</pubDate><description><![CDATA[函数式编程之map、reduce

1 定义
map(function, iterable, ...)：返回一个迭代器，该迭代器将函数应用于可迭代的每个项目，产生结果。 如果传递了额外的可迭代参数，则函数必须采用多参数且并行应用于所有可迭代项中的项目。
示例：
&gt;&gt;&gt; list(map(lambda x: x + 2, list(range(10))))
[2, 3, 4, 5, 6, 7, 8, 9, 10, 11]

functools.reduce(function, iterab.]]></description><category></category></item><item><title><![CDATA[C语言 字符串数组笔记]]></title><link>https://blog.csdn.net/AnimateX/article/details/118771439</link><guid>https://blog.csdn.net/AnimateX/article/details/118771439</guid><author>AnimateX</author><pubDate>Thu, 15 Jul 2021 23:29:34 +0800</pubDate><description><![CDATA[温故而知新，实则菜狗重新看下

1. 引言
字符串数组通常有两类，一种为长度固定了（下图左）；一种长度不固定（下图右）。如下图所示：

2. 使用
这里将显示月份的switch-case替换成字符串数组。
#include &lt;stdio.h&gt;

int main(void)
{
    const char *month_name[] = {
        "January",
        "February",
        "March",
        "April",
   .]]></description><category></category></item><item><title><![CDATA[QGraphicsPixmapItem setOffset踩坑说明]]></title><link>https://blog.csdn.net/AnimateX/article/details/118754290</link><guid>https://blog.csdn.net/AnimateX/article/details/118754290</guid><author>AnimateX</author><pubDate>Thu, 15 Jul 2021 11:00:13 +0800</pubDate><description><![CDATA[背景：在做图像关键点标注工具时，希望同一scene重载文件夹时，图元回归中心（拖拽缩放导致的偏移）。
解决：获取始末坐标；计算x方向和y方向的坐标偏移，然后再重载图元时，反向设置图元偏移即可。

1. bug说明
图元在第一次重载，设置Offset后，画面显示回归正常。但是此时我去跟踪图元的鼠标事件的坐标变化时发现如下情况：

初始，我选择拖动位置为图元的原点（左上角），起始状态均为0.0
然后，拖动图元结束，拖动变化为：x: 271.0, y: 99.0
然后，重新载入其他图片文件夹，并设置Offset.]]></description><category></category></item></channel></rss>