<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Jason24_Zeng的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Jason24_Zeng</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Jason24_Zeng]]></copyright><item><title><![CDATA[【Mac】 iTerm2 使用]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/115430240</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/115430240</guid><author>Jason24_Zeng</author><pubDate>Sun, 04 Apr 2021 16:54:35 +0800</pubDate><description><![CDATA[【Mac】iTerm2 使用iTerm 2 下载iTerm 2 使用操作
iTerm 2 下载
地址：下载链接
配上 tmux 更配哦
iTerm 2 使用操作




操作
快捷键




移动光标至行首
CTRL+A


第一次按时，移动光标至行首；再次按时，回到原有位置
CTRL+X


移动光标至行尾
CTRL+E


光标向左移动一个字符
CTRL+B


光标向左移动一个单词
ESC+B


光标向右移动一个字符
CTRL+F


光标向右移动一个单词
ESC+F


删除光标前一个字符，即退格]]></description><category></category></item><item><title><![CDATA[VSCODE C++入门]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/112725558</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/112725558</guid><author>Jason24_Zeng</author><pubDate>Sat, 16 Jan 2021 22:03:47 +0800</pubDate><description><![CDATA[VSCode C++入门VSCodeVSCode连接远程服务器Terminal ssh连接服务器C++需要安装的插件新建文件编译并run
VSCode
VSCode连接远程服务器
Terminal ssh连接服务器

生成key-gen密钥对
上传公钥到云服务器，不过要是服务器认可的id生成的密钥对才行
先用ssh连接远程服务器看是否成功
然后config Host文件， 没有就touch config文件，有就vim config

Host relay03
  HostName relay03.kuai]]></description><category></category></item><item><title><![CDATA[[SVM] 径向基函数(radial based function)如何将低维向量映射到高维向量的]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109714583</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109714583</guid><author>Jason24_Zeng</author><pubDate>Mon, 16 Nov 2020 10:47:47 +0800</pubDate><description><![CDATA[高斯核函数如何将低维向量映射到高维向量RBF 函数定义RBF 函数性质为什么 RBF 函数是将 x\mathbf{x}x 被映射到无穷维后构造的核函数？
RBF 函数定义
κ(x1,x2)=&lt;ϕ(x1),ϕ(x2)&gt;=exp⁡(−∣∣x1−x2∣∣22σ2)\kappa(\mathbf{x_1}, \mathbf{x_2}) = \left&lt;\phi(\mathbf{x_1}), \phi(\mathbf{x_2})\right&gt;=\exp\left(-\frac{||\mathbf]]></description><category></category></item><item><title><![CDATA[[Mac SSH]如何在Terminal端删除或修改已上传文件夹]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109670055</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109670055</guid><author>Jason24_Zeng</author><pubDate>Fri, 13 Nov 2020 11:02:10 +0800</pubDate><description><![CDATA[如何在Terminal端删除或修改已上传文件夹Step 1: 找到路径Step 2: 修改或删除文件Step 3: Commit to GithubStep 4: Push 修改
Step 1: 找到路径
在Terminal 上 cd 到上传文件时的路径，原因是为了和上传文件所到的仓库取得联系，否则你可能又得git init等一系列连接仓库的操作。

比如我想删除PychamProject里的某一文件，而我之前是直接在这个目录下连接仓库的，那么我就可以操作如下
cd ~/Documents/PycharmP]]></description><category></category></item><item><title><![CDATA[[Mac]如何安装graphviz]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109633285</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109633285</guid><author>Jason24_Zeng</author><pubDate>Wed, 11 Nov 2020 22:55:06 +0800</pubDate><description><![CDATA[如何安装graphviz安装 Homebrew安装Graphviz
安装 Homebrew
method 1：
官网下载，地址：https://brew.sh/
下载指令：
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install.sh)"

method 2:
通过国内镜像进行下载
直接输入下载指令
/bin/bash -c "$(curl -fsSL https://gitee]]></description><category></category></item><item><title><![CDATA[Mac 安装配置JDK14 并安装配置Intellij IDEA]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109525358</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109525358</guid><author>Jason24_Zeng</author><pubDate>Fri, 06 Nov 2020 11:06:33 +0800</pubDate><description><![CDATA[JDK及IDE在Mac OS系统中的安装JDK 下载与安装
JDK 下载与安装
这个简单，几乎不会出问题，我用的Oracle14，官网下载太慢了，将安装包放到了百度网盘，密码是df9w，安装过程没有任何困难，照着指示走就行。

...]]></description><category></category></item><item><title><![CDATA[Mac Git 安装与使用教程]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109515012</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109515012</guid><author>Jason24_Zeng</author><pubDate>Thu, 05 Nov 2020 17:20:13 +0800</pubDate><description><![CDATA[Macbook GitHub 教程1. Git 安装2. 创建ssh keys3. 配置Git4. 验证是否链接5. Clone 项目6. 修改并上传项目
1. Git 安装
由于之前为了安装 jupyter notebook提前安装过了Xcode 及其Command Line Tools，而Xcode其实集成了Git，所以电脑本来就有git的命令，具体安装方式可详见 Xcode中Command Line Tools安装方法
我们可以通过在Terminal上输入git指令来检验是否安装成功，如果安装成功，T]]></description><category></category></item><item><title><![CDATA[Char_Level_CNN_Model 模型介绍]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109236352</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109236352</guid><author>Jason24_Zeng</author><pubDate>Fri, 23 Oct 2020 10:54:53 +0800</pubDate><description><![CDATA[Char_Level_CNN_Model模型介绍模型参数总览逐层介绍Input LayerEmbedding LayerConvolution layers 1D逐层计算参数量：Flatten LayerFull Connected(FC) Layers and Output Layer参数数量预测：Build ModelTrain Model
模型介绍
模型参数总览
_________________________________________________________________
Laye]]></description><category></category></item><item><title><![CDATA[简单理解ConvNet层的Backpropagation]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109120211</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109120211</guid><author>Jason24_Zeng</author><pubDate>Fri, 16 Oct 2020 17:58:30 +0800</pubDate><description><![CDATA[这里只谈一下二维的情况，也就是channel = 1,  strides = 1的情况下，单就这一层的计算.
ConvNet forward propagation
filter’s shape: [height, length, 1]

yi,j=Σn=0lengthΣm=0heightWm,nxi+m,j+ny_{i,j} = \Sigma_{n=0}^{length}\Sigma_{m=0}^{height} W_{m,n}x_{i+m,j+n}yi,j​=Σn=0length​Σm=0height​]]></description><category></category></item><item><title><![CDATA[使用20_newsgroup集做训练集，载入Glove预训练权重训练模型]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109100286</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109100286</guid><author>Jason24_Zeng</author><pubDate>Thu, 15 Oct 2020 17:59:12 +0800</pubDate><description><![CDATA[使用20_newsgroup集做训练集，载入Glove预训练权重训练模型预训练20_newsgroup数据集Load samplePreview file folderDefine the path to 20_newsgroup folderLoad data from all the child folder in 20_newsgroupPreprocess the texts dataImport LibraryTokenizerPad_sequencesPreprocess the labelsI]]></description><category></category></item><item><title><![CDATA[tensorflow.keras的某些函数的使用]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109093249</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109093249</guid><author>Jason24_Zeng</author><pubDate>Thu, 15 Oct 2020 16:30:35 +0800</pubDate><description><![CDATA[Tensorflow.keras common functionTensorflow.keras.preprocessingTokenizersPad_sequences
Tensorflow.keras.preprocessing
Tokenizers
tf.keras.preprocessing.text.Tokenizer(
    num_words=None, filters='!"#$%&amp;()*+,-./:;&lt;=&gt;?@[\\]^_`{|}~\t\n', lower=True,]]></description><category></category></item><item><title><![CDATA[AttributeError: module ‘tensorflow.python.framework.ops‘ has no attribute ‘_TensorLike‘解决方法]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109088801</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109088801</guid><author>Jason24_Zeng</author><pubDate>Thu, 15 Oct 2020 00:26:53 +0800</pubDate><description><![CDATA[贴出问题：
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
&lt;ipython-input-17-c998649dcbf1&gt; in &lt;module&gt;
----&gt; 1 autoencoder, encoder = autoenco]]></description><category></category></item><item><title><![CDATA[Word2Vec skip-gram model]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109076525</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109076525</guid><author>Jason24_Zeng</author><pubDate>Wed, 14 Oct 2020 19:33:55 +0800</pubDate><description><![CDATA[word2vec Skip-Gram model的理解参考文献运用word2vec的目的模型结构1. Feedforward Neural Net Language Model (NNLM)2. Recurrent Neural Net Language Model (RNNLM)3. Parallel Training of Neural Networks4. New Log-Linear Models4.1 CBOW: Continuous Bag-of-Words Model4.2 Continuou]]></description><category></category></item><item><title><![CDATA[Train_word2vec in Convolutional Neural Network for Sentences Classification]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109047044</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109047044</guid><author>Jason24_Zeng</author><pubDate>Tue, 13 Oct 2020 12:03:24 +0800</pubDate><description><![CDATA[Parameters
train_word2vec(sentence_matrix, vocabulary_inv,
				num_features = 300, min_word_count = 1, context = 10)

上面主要涉及了五个参数：

sentences matrix: 一个整数矩阵，就是每句话对应一行，长度不够做pad, 每个单词用vocabulary里的value替代
vocabulary_inv: dict{int: str}用来对应每个value表示哪个单词
num_fe]]></description><category></category></item><item><title><![CDATA[Convolutional Neural Networks for Sentence Classification 代码实现]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109024368</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109024368</guid><author>Jason24_Zeng</author><pubDate>Tue, 13 Oct 2020 00:39:24 +0800</pubDate><description><![CDATA[NLP: Convolutional Neural Networks for Sentence Classification 代码实现预处理数据原始数据及预处理数据结果预处理步骤Load DataLoad from fileSplit by WordsGenerate LabelsPaddingBuild VocabularyMap Sentences and Labels to IndexShuffle Data and Split Train/Test SetWord Embedding（将word 转]]></description><category></category></item><item><title><![CDATA[Convolutional Neural Network for Sentences Classification 笔记]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109016819</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109016819</guid><author>Jason24_Zeng</author><pubDate>Mon, 12 Oct 2020 00:39:14 +0800</pubDate><description><![CDATA[Convolutional Neural Network for Sentences Classfication
文章特点

仅用一个很简单的拥有少数hyperpamater tuning and static vectors 的CNN, 就提高了七个任务中的四个SOA(state of the art)，其中包括sentiment analysis and question classification.
四个model:
2.1 CNN-rand: 基准model, 里面的字随机初始化成向量vector]]></description><category></category></item><item><title><![CDATA[NLP_Transformer_Attention_mechanism]]></title><link>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109006162</link><guid>https://blog.youkuaiyun.com/Jason24_Zeng/article/details/109006162</guid><author>Jason24_Zeng</author><pubDate>Sun, 11 Oct 2020 12:16:12 +0800</pubDate><description><![CDATA[Transformer 之前技术机器局限

RNN: 可捕获长距离依赖性息，无法并行
CNN: 能并行，无法捕获长距离依赖关系（需要通过pooling 或 kernel 去扩张感受野
传统Attention

NLP神经网络中某些层的解释
Attention Mechanism(注意力机制)
注意力模型之前比较重要的需要留意的模型：seq2seq, or RNN encoder-decoder,
http://emnlp2014.org/papers/pdf/EMNLP2014179.pdf

较早火起来是]]></description><category></category></item></channel></rss>