Docker Swarm
Swarm 与Docker Engine集成的集群管理: 创建一组Docker引擎，可以在其中部署应用程序服务。不需要额外的编排软件来创建或管理swarm
0%
Fat vs Deep Network
发表于：
分类于：
ml
全量空间与样本空间 全量空间(理想状态)： If we can collect all datasets in the universe $ D_{all} $, we can find the best threshold $h^{all}$ $ h^{all} = \text{arg}\min\limits_h L (h, D_{all}) $ 样本空间(现实发生)： We only collect some examples $D_{train}$ from $D_{all}$ $ D_{train} = {(x^1,
Many Factors Affecting Optimization
发表于：
分类于：
ml
机器学习的一般步骤 Model Bias The model is too simple. find a needle in a haystack (大海捞针) but there is no needle Solution: redesign your model to make it more flexible more features more neurons, layers Optimization Issue Large loss not always imply model bias. There is another possibility … A needle is in a haystack…, Just cannot find
Logistic Regression
发表于：
分类于：
ml
Function Set We want to find $P_{w,b}(C_1|x)$ $P_{w, b}(C_1|x) = \sigma(z)$ $ z = w \cdot x + b = \sum\limits_iw_ix_i + b $ function set: $f_{w, b}(x) = P_{w, b}(C_1|x)$ // including all different w and b Goodness of a Function 取minima的 对象写成一个function，这个fun
Using Ghostscript
Ghostscript介绍 Ghostscript 是 PostScript®和可移植文档格式(PDF)文件的解释器。 Ghostscript 由 PostScript解释器层和图形库组成。图
Classification
发表于：
分类于：
ml
Classification: Given options(classes), the function outputs the correct one. Probabilistic Generative Model features and predict target 一共有7个features，其中 Total = HP + Attack + Deffense + SP Atk + Sp Def + Speed predict target: type of pokemon features and predict target How to do Classification 收集Training
Regression
发表于：
分类于：
ml
Regression: Input a vector, the function outputs a scalar. 使用简单模型 预测问题：根据前面的浏览数据，预测后面的浏览量 Function with Unknown Parameters Model: $y = b + wx_1$ $y$(Label): no. of views on 2/26， $x_1$(feature): no. of views on 2/25 $w$(weight) and $b$(bias) are unknown parameters
Mpv快捷键
发表于：
分类于：
blog
Seek backward/forward 5 seconds LEFT and RIGHT Seek backward/forward 5 seconds. Shift+arrow does a 1 second exact seek (see --hr-seek). Seek forward/backward 1 minute UP and DOWN Seek forward/backward 1 minute. Shift+arrow does a 5 second exact seek (see --hr-seek). Seek to the previous/next subtitle Ctrl+LEFT and Ctrl+RIGHT Seek to the previous/next subtitle. Subject to some restrictions and might not always work; see sub-seek command. Adjust subtitle delay Ctrl+Shift+Left and
Docker Command
发表于：
Docker Removing Removing Images by ID or Name docker rmi <id> Docker Image Prune Removes All Dangling Images docker image prune -a Docker System Prune 查找和删除所有未使用的对象 docker system prune -a