\documentclass[10pt,a4paper]{article}

% Packages
\usepackage{fancyhdr}           % For header and footer
\usepackage{multicol}           % Allows multicols in tables
\usepackage{tabularx}           % Intelligent column widths
\usepackage{tabulary}           % Used in header and footer
\usepackage{hhline}             % Border under tables
\usepackage{graphicx}           % For images
\usepackage{xcolor}             % For hex colours
%\usepackage[utf8x]{inputenc}    % For unicode character support
\usepackage[T1]{fontenc}        % Without this we get weird character replacements
\usepackage{colortbl}           % For coloured tables
\usepackage{setspace}           % For line height
\usepackage{lastpage}           % Needed for total page number
\usepackage{seqsplit}           % Splits long words.
%\usepackage{opensans}          % Can't make this work so far. Shame. Would be lovely.
\usepackage[normalem]{ulem}     % For underlining links
% Most of the following are not required for the majority
% of cheat sheets but are needed for some symbol support.
\usepackage{amsmath}            % Symbols
\usepackage{MnSymbol}           % Symbols
\usepackage{wasysym}            % Symbols
%\usepackage[english,german,french,spanish,italian]{babel}              % Languages

% Document Info
\author{furkandurul}
\pdfinfo{
  /Title (google-machine-learning-crash-course.pdf)
  /Creator (Cheatography)
  /Author (furkandurul)
  /Subject (Google Machine Learning Crash Course Cheat Sheet)
}

% Lengths and widths
\addtolength{\textwidth}{6cm}
\addtolength{\textheight}{-1cm}
\addtolength{\hoffset}{-3cm}
\addtolength{\voffset}{-2cm}
\setlength{\tabcolsep}{0.2cm} % Space between columns
\setlength{\headsep}{-12pt} % Reduce space between header and content
\setlength{\headheight}{85pt} % If less, LaTeX automatically increases it
\renewcommand{\footrulewidth}{0pt} % Remove footer line
\renewcommand{\headrulewidth}{0pt} % Remove header line
\renewcommand{\seqinsert}{\ifmmode\allowbreak\else\-\fi} % Hyphens in seqsplit
% This two commands together give roughly
% the right line height in the tables
\renewcommand{\arraystretch}{1.3}
\onehalfspacing

% Commands
\newcommand{\SetRowColor}[1]{\noalign{\gdef\RowColorName{#1}}\rowcolor{\RowColorName}} % Shortcut for row colour
\newcommand{\mymulticolumn}[3]{\multicolumn{#1}{>{\columncolor{\RowColorName}}#2}{#3}} % For coloured multi-cols
\newcolumntype{x}[1]{>{\raggedright}p{#1}} % New column types for ragged-right paragraph columns
\newcommand{\tn}{\tabularnewline} % Required as custom column type in use

% Font and Colours
\definecolor{HeadBackground}{HTML}{333333}
\definecolor{FootBackground}{HTML}{666666}
\definecolor{TextColor}{HTML}{333333}
\definecolor{DarkBackground}{HTML}{A2A2A3}
\definecolor{LightBackground}{HTML}{F3F3F3}
\renewcommand{\familydefault}{\sfdefault}
\color{TextColor}

% Header and Footer
\pagestyle{fancy}
\fancyhead{} % Set header to blank
\fancyfoot{} % Set footer to blank
\fancyhead[L]{
\noindent
\begin{multicols}{3}
\begin{tabulary}{5.8cm}{C}
    \SetRowColor{DarkBackground}
    \vspace{-7pt}
    {\parbox{\dimexpr\textwidth-2\fboxsep\relax}{\noindent
        \hspace*{-6pt}\includegraphics[width=5.8cm]{/web/www.cheatography.com/public/images/cheatography_logo.pdf}}
    }
\end{tabulary}
\columnbreak
\begin{tabulary}{11cm}{L}
    \vspace{-2pt}\large{\bf{\textcolor{DarkBackground}{\textrm{Google Machine Learning Crash Course Cheat Sheet}}}} \\
    \normalsize{by \textcolor{DarkBackground}{furkandurul} via \textcolor{DarkBackground}{\uline{cheatography.com/122404/cs/22715/}}}
\end{tabulary}
\end{multicols}}

\fancyfoot[L]{ \footnotesize
\noindent
\begin{multicols}{3}
\begin{tabulary}{5.8cm}{LL}
  \SetRowColor{FootBackground}
  \mymulticolumn{2}{p{5.377cm}}{\bf\textcolor{white}{Cheatographer}}  \\
  \vspace{-2pt}furkandurul \\
  \uline{cheatography.com/furkandurul} \\
  \end{tabulary}
\vfill
\columnbreak
\begin{tabulary}{5.8cm}{L}
  \SetRowColor{FootBackground}
  \mymulticolumn{1}{p{5.377cm}}{\bf\textcolor{white}{Cheat Sheet}}  \\
   \vspace{-2pt}Not Yet Published.\\
   Updated 8th May, 2020.\\
   Page {\thepage} of \pageref{LastPage}.
\end{tabulary}
\vfill
\columnbreak
\begin{tabulary}{5.8cm}{L}
  \SetRowColor{FootBackground}
  \mymulticolumn{1}{p{5.377cm}}{\bf\textcolor{white}{Sponsor}}  \\
  \SetRowColor{white}
  \vspace{-5pt}
  %\includegraphics[width=48px,height=48px]{dave.jpeg}
  Measure your website readability!\\
  www.readability-score.com
\end{tabulary}
\end{multicols}}




\begin{document}
\raggedright
\raggedcolumns

% Set font size to small. Switch to any value
% from this page to resize cheat sheet text:
% www.emerson.emory.edu/services/latex/latex_169.html
\footnotesize % Small font.

\begin{multicols*}{2}

\begin{tabularx}{8.4cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{8.4cm}}{\bf\textcolor{white}{Machine Learning Terminology}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Label}} is variable we're predicting. Represented by {\bf{y}}.} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Features}} are input variables describing data. Represented by the variables {\bf{\{x`1`,x`2`,…,x`n` \}}}} \tn 
% Row Count 5 (+ 3)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Example}} is a particular instance of data, {\bf{x}} \{\{nl\}\} - {\bf{Labeled example}} has \{features, label\}: (x,y) \{\{nl\}\}~~~Used to train the model. \{\{nl\}\}- {\bf{Unlabeled example}} has \{features,?\}: (x, ?) \{\{nl\}\} ~~~Used for making predictions on new data.} \tn 
% Row Count 11 (+ 6)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Model}} maps examples to predicted labels: y'. Defined by internal parameters, which are learned.} \tn 
% Row Count 14 (+ 3)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Training}} means creating or learning the model. You show the model labeled examples and enable the model to learn the relationship between features and label.} \tn 
% Row Count 18 (+ 4)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Inference}} means applying the trained model to unlabeled examples. You use the trained model to make useful predictions (y').} \tn 
% Row Count 21 (+ 3)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Regression}} model predicts {\emph{continuous values}}. \{\{nl\}\}For example;  What is the value of a house in California?} \tn 
% Row Count 24 (+ 3)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Classification}} model predicts {\emph{discrete values}}. \{\{nl\}\} \{\{nobreak\}\}For example;  Is a given e mail message spam or not spam?} \tn 
% Row Count 27 (+ 3)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Hyperparameters}} are the knobs that programmers tweak in machine learning algorithms.} \tn 
% Row Count 29 (+ 2)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{8.4cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{8.4cm}}{\bf\textcolor{white}{Model and Equation}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{Equation for a model in machine learning;\{\{nl\}\}~~~~y'=b+w`1`x`1`} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{8.4cm}}{y' is the predicted label.\{\{nl\}\}b is the bias, also referred to as w`0`.\{\{nl\}\}w`1` is the weight.\{\{nl\}\}x`1` is a feature (a known input).} \tn 
% Row Count 5 (+ 3)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{Some models have multiple features. For example, a model relies on three features look as follows;\{\{nl\}\}~~~~y'=b+w`1`x`1`+w`2`x`2`+w`3`x`3`} \tn 
% Row Count 9 (+ 4)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{8.4cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{8.4cm}}{\bf\textcolor{white}{Training and Loss}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Training}} a model means learning values for all the weights and bias from labeled examples.} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Loss}} is a number indicating how bad the model's prediction on a single example. The goal of training a model is to find a set of weights and biases that have low loss, on average, across all examples.} \tn 
% Row Count 7 (+ 5)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Mean square error (MSE)}} is the average squared loss per example over the whole dataset.\{\{nl\}\}MSE=1/N ∑`(x,y)∈D` (y-prediction(x))\textasciicircum{}2\textasciicircum{}\{\{nl\}\}~~- x is set of features.\{\{nl\}\}~~- y is example's label.\{\{nl\}\}~~- prediction(x) is function of the weights and bias of features of x.\{\{nl\}\}~~- D is data set containing labeled examples.\{\{nl\}\}~~- N is the number of examples in D.} \tn 
% Row Count 16 (+ 9)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{8.4cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{8.4cm}}{\bf\textcolor{white}{Reducing Loss}}  \tn
\SetRowColor{LightBackground}
\mymulticolumn{1}{p{8.4cm}}{\vspace{1px}\centerline{\includegraphics[width=5.1cm]{/web/www.cheatography.com/public/uploads/furkandurul_1588926085_Capture.JPG}}} \tn 
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{8.4cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{8.4cm}}{\bf\textcolor{white}{Reducing Loss}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{Learning continues iterating until the algorithm discovers the model parameters with the lowest possible loss. Usually, until overall loss stops changing or at least changes extremely slowly. When that happens, we say that the model has {\bf{converged}}.} \tn 
% Row Count 6 (+ 6)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{8.4cm}}{{\bf{Gradient descent}} algorithm calculates the gradient of the loss curve. When there is {\emph{single weight}}, gradient of the loss is the derivative (slope) of the curve, When there are {\emph{multiple weights}}, the gradient is a vector of partial derivatives with respect to the weights.} \tn 
% Row Count 12 (+ 6)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{8.4cm}}{Gradient is a vector, so it has both of the following characteristics; a {\emph{direction}} and a {\emph{magnitude}}\{\{nl\}\}The gradient always points in the direction of steepest increase. The gradient descent algorithm takes a step in the direction of the {\bf{negative gradient}} in order to reduce loss as quickly as possible.} \tn 
% Row Count 19 (+ 7)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{8.4cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{8.4cm}}{\bf\textcolor{white}{Gradient Descent}}  \tn
\SetRowColor{LightBackground}
\mymulticolumn{1}{p{8.4cm}}{\vspace{1px}\centerline{\includegraphics[width=5.1cm]{/web/www.cheatography.com/public/uploads/furkandurul_1588933268_Capture.JPG}}} \tn 
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}


% That's all folks
\end{multicols*}

\end{document}