+++ /dev/null
-
-% Default to the notebook output style
-
-
-
-
-% Inherit from the specified cell style.
-
-
-
-
-
-\documentclass[11pt]{article}
-
-
-
- \usepackage[T1]{fontenc}
- % Nicer default font (+ math font) than Computer Modern for most use cases
- \usepackage{mathpazo}
-
- % Basic figure setup, for now with no caption control since it's done
- % automatically by Pandoc (which extracts  syntax from Markdown).
- \usepackage{graphicx}
- % We will generate all images so they have a width \maxwidth. This means
- % that they will get their normal width if they fit onto the page, but
- % are scaled down if they would overflow the margins.
- \makeatletter
- \def\maxwidth{\ifdim\Gin@nat@width>\linewidth\linewidth
- \else\Gin@nat@width\fi}
- \makeatother
- \let\Oldincludegraphics\includegraphics
- % Set max figure width to be 80% of text width, for now hardcoded.
- \renewcommand{\includegraphics}[1]{\Oldincludegraphics[width=.8\maxwidth]{#1}}
- % Ensure that by default, figures have no caption (until we provide a
- % proper Figure object with a Caption API and a way to capture that
- % in the conversion process - todo).
- \usepackage{caption}
- \DeclareCaptionLabelFormat{nolabel}{}
- \captionsetup{labelformat=nolabel}
-
- \usepackage{adjustbox} % Used to constrain images to a maximum size
- \usepackage{xcolor} % Allow colors to be defined
- \usepackage{enumerate} % Needed for markdown enumerations to work
- \usepackage{geometry} % Used to adjust the document margins
- \usepackage{amsmath} % Equations
- \usepackage{amssymb} % Equations
- \usepackage{textcomp} % defines textquotesingle
- % Hack from http://tex.stackexchange.com/a/47451/13684:
- \AtBeginDocument{%
- \def\PYZsq{\textquotesingle}% Upright quotes in Pygmentized code
- }
- \usepackage{upquote} % Upright quotes for verbatim code
- \usepackage{eurosym} % defines \euro
- \usepackage[mathletters]{ucs} % Extended unicode (utf-8) support
- \usepackage[utf8x]{inputenc} % Allow utf-8 characters in the tex document
- \usepackage{fancyvrb} % verbatim replacement that allows latex
- \usepackage{grffile} % extends the file name processing of package graphics
- % to support a larger range
- % The hyperref package gives us a pdf with properly built
- % internal navigation ('pdf bookmarks' for the table of contents,
- % internal cross-reference links, web links for URLs, etc.)
- \usepackage{hyperref}
- \usepackage{longtable} % longtable support required by pandoc >1.10
- \usepackage{booktabs} % table support for pandoc > 1.12.2
- \usepackage[inline]{enumitem} % IRkernel/repr support (it uses the enumerate* environment)
- \usepackage[normalem]{ulem} % ulem is needed to support strikethroughs (\sout)
- % normalem makes italics be italics, not underlines
-
-
-
-
- % Colors for the hyperref package
- \definecolor{urlcolor}{rgb}{0,.145,.698}
- \definecolor{linkcolor}{rgb}{.71,0.21,0.01}
- \definecolor{citecolor}{rgb}{.12,.54,.11}
-
- % ANSI colors
- \definecolor{ansi-black}{HTML}{3E424D}
- \definecolor{ansi-black-intense}{HTML}{282C36}
- \definecolor{ansi-red}{HTML}{E75C58}
- \definecolor{ansi-red-intense}{HTML}{B22B31}
- \definecolor{ansi-green}{HTML}{00A250}
- \definecolor{ansi-green-intense}{HTML}{007427}
- \definecolor{ansi-yellow}{HTML}{DDB62B}
- \definecolor{ansi-yellow-intense}{HTML}{B27D12}
- \definecolor{ansi-blue}{HTML}{208FFB}
- \definecolor{ansi-blue-intense}{HTML}{0065CA}
- \definecolor{ansi-magenta}{HTML}{D160C4}
- \definecolor{ansi-magenta-intense}{HTML}{A03196}
- \definecolor{ansi-cyan}{HTML}{60C6C8}
- \definecolor{ansi-cyan-intense}{HTML}{258F8F}
- \definecolor{ansi-white}{HTML}{C5C1B4}
- \definecolor{ansi-white-intense}{HTML}{A1A6B2}
-
- % commands and environments needed by pandoc snippets
- % extracted from the output of `pandoc -s`
- \providecommand{\tightlist}{%
- \setlength{\itemsep}{0pt}\setlength{\parskip}{0pt}}
- \DefineVerbatimEnvironment{Highlighting}{Verbatim}{commandchars=\\\{\}}
- % Add ',fontsize=\small' for more characters per line
- \newenvironment{Shaded}{}{}
- \newcommand{\KeywordTok}[1]{\textcolor[rgb]{0.00,0.44,0.13}{\textbf{{#1}}}}
- \newcommand{\DataTypeTok}[1]{\textcolor[rgb]{0.56,0.13,0.00}{{#1}}}
- \newcommand{\DecValTok}[1]{\textcolor[rgb]{0.25,0.63,0.44}{{#1}}}
- \newcommand{\BaseNTok}[1]{\textcolor[rgb]{0.25,0.63,0.44}{{#1}}}
- \newcommand{\FloatTok}[1]{\textcolor[rgb]{0.25,0.63,0.44}{{#1}}}
- \newcommand{\CharTok}[1]{\textcolor[rgb]{0.25,0.44,0.63}{{#1}}}
- \newcommand{\StringTok}[1]{\textcolor[rgb]{0.25,0.44,0.63}{{#1}}}
- \newcommand{\CommentTok}[1]{\textcolor[rgb]{0.38,0.63,0.69}{\textit{{#1}}}}
- \newcommand{\OtherTok}[1]{\textcolor[rgb]{0.00,0.44,0.13}{{#1}}}
- \newcommand{\AlertTok}[1]{\textcolor[rgb]{1.00,0.00,0.00}{\textbf{{#1}}}}
- \newcommand{\FunctionTok}[1]{\textcolor[rgb]{0.02,0.16,0.49}{{#1}}}
- \newcommand{\RegionMarkerTok}[1]{{#1}}
- \newcommand{\ErrorTok}[1]{\textcolor[rgb]{1.00,0.00,0.00}{\textbf{{#1}}}}
- \newcommand{\NormalTok}[1]{{#1}}
-
- % Additional commands for more recent versions of Pandoc
- \newcommand{\ConstantTok}[1]{\textcolor[rgb]{0.53,0.00,0.00}{{#1}}}
- \newcommand{\SpecialCharTok}[1]{\textcolor[rgb]{0.25,0.44,0.63}{{#1}}}
- \newcommand{\VerbatimStringTok}[1]{\textcolor[rgb]{0.25,0.44,0.63}{{#1}}}
- \newcommand{\SpecialStringTok}[1]{\textcolor[rgb]{0.73,0.40,0.53}{{#1}}}
- \newcommand{\ImportTok}[1]{{#1}}
- \newcommand{\DocumentationTok}[1]{\textcolor[rgb]{0.73,0.13,0.13}{\textit{{#1}}}}
- \newcommand{\AnnotationTok}[1]{\textcolor[rgb]{0.38,0.63,0.69}{\textbf{\textit{{#1}}}}}
- \newcommand{\CommentVarTok}[1]{\textcolor[rgb]{0.38,0.63,0.69}{\textbf{\textit{{#1}}}}}
- \newcommand{\VariableTok}[1]{\textcolor[rgb]{0.10,0.09,0.49}{{#1}}}
- \newcommand{\ControlFlowTok}[1]{\textcolor[rgb]{0.00,0.44,0.13}{\textbf{{#1}}}}
- \newcommand{\OperatorTok}[1]{\textcolor[rgb]{0.40,0.40,0.40}{{#1}}}
- \newcommand{\BuiltInTok}[1]{{#1}}
- \newcommand{\ExtensionTok}[1]{{#1}}
- \newcommand{\PreprocessorTok}[1]{\textcolor[rgb]{0.74,0.48,0.00}{{#1}}}
- \newcommand{\AttributeTok}[1]{\textcolor[rgb]{0.49,0.56,0.16}{{#1}}}
- \newcommand{\InformationTok}[1]{\textcolor[rgb]{0.38,0.63,0.69}{\textbf{\textit{{#1}}}}}
- \newcommand{\WarningTok}[1]{\textcolor[rgb]{0.38,0.63,0.69}{\textbf{\textit{{#1}}}}}
-
-
- % Define a nice break command that doesn't care if a line doesn't already
- % exist.
- \def\br{\hspace*{\fill} \\* }
- % Math Jax compatability definitions
- \def\gt{>}
- \def\lt{<}
- % Document parameters
- \title{report\_P7\_H17}
-
-
-
-
- % Pygments definitions
-
-\makeatletter
-\def\PY@reset{\let\PY@it=\relax \let\PY@bf=\relax%
- \let\PY@ul=\relax \let\PY@tc=\relax%
- \let\PY@bc=\relax \let\PY@ff=\relax}
-\def\PY@tok#1{\csname PY@tok@#1\endcsname}
-\def\PY@toks#1+{\ifx\relax#1\empty\else%
- \PY@tok{#1}\expandafter\PY@toks\fi}
-\def\PY@do#1{\PY@bc{\PY@tc{\PY@ul{%
- \PY@it{\PY@bf{\PY@ff{#1}}}}}}}
-\def\PY#1#2{\PY@reset\PY@toks#1+\relax+\PY@do{#2}}
-
-\expandafter\def\csname PY@tok@w\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.73,0.73}{##1}}}
-\expandafter\def\csname PY@tok@c\endcsname{\let\PY@it=\textit\def\PY@tc##1{\textcolor[rgb]{0.25,0.50,0.50}{##1}}}
-\expandafter\def\csname PY@tok@cp\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.74,0.48,0.00}{##1}}}
-\expandafter\def\csname PY@tok@k\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@kp\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@kt\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.69,0.00,0.25}{##1}}}
-\expandafter\def\csname PY@tok@o\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@ow\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.67,0.13,1.00}{##1}}}
-\expandafter\def\csname PY@tok@nb\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@nf\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.00,1.00}{##1}}}
-\expandafter\def\csname PY@tok@nc\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.00,1.00}{##1}}}
-\expandafter\def\csname PY@tok@nn\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.00,1.00}{##1}}}
-\expandafter\def\csname PY@tok@ne\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.82,0.25,0.23}{##1}}}
-\expandafter\def\csname PY@tok@nv\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.10,0.09,0.49}{##1}}}
-\expandafter\def\csname PY@tok@no\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.53,0.00,0.00}{##1}}}
-\expandafter\def\csname PY@tok@nl\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.63,0.63,0.00}{##1}}}
-\expandafter\def\csname PY@tok@ni\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.60,0.60,0.60}{##1}}}
-\expandafter\def\csname PY@tok@na\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.49,0.56,0.16}{##1}}}
-\expandafter\def\csname PY@tok@nt\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@nd\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.67,0.13,1.00}{##1}}}
-\expandafter\def\csname PY@tok@s\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@sd\endcsname{\let\PY@it=\textit\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@si\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.73,0.40,0.53}{##1}}}
-\expandafter\def\csname PY@tok@se\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.73,0.40,0.13}{##1}}}
-\expandafter\def\csname PY@tok@sr\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.40,0.53}{##1}}}
-\expandafter\def\csname PY@tok@ss\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.10,0.09,0.49}{##1}}}
-\expandafter\def\csname PY@tok@sx\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@m\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@gh\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.00,0.50}{##1}}}
-\expandafter\def\csname PY@tok@gu\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.50,0.00,0.50}{##1}}}
-\expandafter\def\csname PY@tok@gd\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.63,0.00,0.00}{##1}}}
-\expandafter\def\csname PY@tok@gi\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.63,0.00}{##1}}}
-\expandafter\def\csname PY@tok@gr\endcsname{\def\PY@tc##1{\textcolor[rgb]{1.00,0.00,0.00}{##1}}}
-\expandafter\def\csname PY@tok@ge\endcsname{\let\PY@it=\textit}
-\expandafter\def\csname PY@tok@gs\endcsname{\let\PY@bf=\textbf}
-\expandafter\def\csname PY@tok@gp\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.00,0.50}{##1}}}
-\expandafter\def\csname PY@tok@go\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.53,0.53,0.53}{##1}}}
-\expandafter\def\csname PY@tok@gt\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.27,0.87}{##1}}}
-\expandafter\def\csname PY@tok@err\endcsname{\def\PY@bc##1{\setlength{\fboxsep}{0pt}\fcolorbox[rgb]{1.00,0.00,0.00}{1,1,1}{\strut ##1}}}
-\expandafter\def\csname PY@tok@kc\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@kd\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@kn\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@kr\endcsname{\let\PY@bf=\textbf\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@bp\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.50,0.00}{##1}}}
-\expandafter\def\csname PY@tok@fm\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.00,0.00,1.00}{##1}}}
-\expandafter\def\csname PY@tok@vc\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.10,0.09,0.49}{##1}}}
-\expandafter\def\csname PY@tok@vg\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.10,0.09,0.49}{##1}}}
-\expandafter\def\csname PY@tok@vi\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.10,0.09,0.49}{##1}}}
-\expandafter\def\csname PY@tok@vm\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.10,0.09,0.49}{##1}}}
-\expandafter\def\csname PY@tok@sa\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@sb\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@sc\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@dl\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@s2\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@sh\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@s1\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.73,0.13,0.13}{##1}}}
-\expandafter\def\csname PY@tok@mb\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@mf\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@mh\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@mi\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@il\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@mo\endcsname{\def\PY@tc##1{\textcolor[rgb]{0.40,0.40,0.40}{##1}}}
-\expandafter\def\csname PY@tok@ch\endcsname{\let\PY@it=\textit\def\PY@tc##1{\textcolor[rgb]{0.25,0.50,0.50}{##1}}}
-\expandafter\def\csname PY@tok@cm\endcsname{\let\PY@it=\textit\def\PY@tc##1{\textcolor[rgb]{0.25,0.50,0.50}{##1}}}
-\expandafter\def\csname PY@tok@cpf\endcsname{\let\PY@it=\textit\def\PY@tc##1{\textcolor[rgb]{0.25,0.50,0.50}{##1}}}
-\expandafter\def\csname PY@tok@c1\endcsname{\let\PY@it=\textit\def\PY@tc##1{\textcolor[rgb]{0.25,0.50,0.50}{##1}}}
-\expandafter\def\csname PY@tok@cs\endcsname{\let\PY@it=\textit\def\PY@tc##1{\textcolor[rgb]{0.25,0.50,0.50}{##1}}}
-
-\def\PYZbs{\char`\\}
-\def\PYZus{\char`\_}
-\def\PYZob{\char`\{}
-\def\PYZcb{\char`\}}
-\def\PYZca{\char`\^}
-\def\PYZam{\char`\&}
-\def\PYZlt{\char`\<}
-\def\PYZgt{\char`\>}
-\def\PYZsh{\char`\#}
-\def\PYZpc{\char`\%}
-\def\PYZdl{\char`\$}
-\def\PYZhy{\char`\-}
-\def\PYZsq{\char`\'}
-\def\PYZdq{\char`\"}
-\def\PYZti{\char`\~}
-% for compatibility with earlier versions
-\def\PYZat{@}
-\def\PYZlb{[}
-\def\PYZrb{]}
-\makeatother
-
-
- % Exact colors from NB
- \definecolor{incolor}{rgb}{0.0, 0.0, 0.5}
- \definecolor{outcolor}{rgb}{0.545, 0.0, 0.0}
-
-
-
-
- % Prevent overflowing lines due to hard-to-break entities
- \sloppy
- % Setup hyperref package
- \hypersetup{
- breaklinks=true, % so long urls are correctly broken across lines
- colorlinks=true,
- urlcolor=urlcolor,
- linkcolor=linkcolor,
- citecolor=citecolor,
- }
- % Slightly bigger margins than the latex defaults
-
- \geometry{verbose,tmargin=1in,bmargin=1in,lmargin=1in,rmargin=1in}
-
-
-\graphicspath{{./figs/}}
-
- \begin{document}
-
-
- \maketitle
-
-
-
-
- \subsection*{Introduction}
-
-Cette partie montre les résultats obtenus via des variantes de
-l'algorithme décrit à la section 2, en utilisant le package présenté à
-la section 3. Cet algorithme est systématiquement comparé à deux
-approches naïves : * la moyenne des lendemains des jours "similaires"
-dans tout le passé, c'est-à-dire prédiction = moyenne de tous les mardis
-passé si le jour courant est un lndi par exemple. * la persistence,
-reproduisant le jour courant ou allant chercher le lendemain de la
-dernière journée "similaire" (même principe que ci-dessus ; argument
-"same\_day").
-
-Concernant l'algorithme principal à voisins, trois variantes sont
-étudiées dans cette partie : * avec simtype="mix" et raccordement
-"Neighbors" dans le cas "non local", i.e. on va chercher des voisins
-n'importe où du moment qu'ils correspondent au premier élément d'un
-couple de deux jours consécutifs sans valeurs manquantes. * avec
-simtype="endo" + raccordement "Neighbors" puis simtype="none" +
-raccordement "Zero" (sans ajustement) dans le cas "local" : voisins de
-même niveau de pollution et même saison.
-
-Pour chaque période retenue -\/- chauffage, épandage, semaine non
-polluée -\/- les erreurs de prédiction sont d'abord affichées, puis
-quelques graphes de courbes réalisées/prévues (sur le jour "en moyenne
-le plus facile" à gauche, et "en moyenne le plus difficile" à droite).
-Ensuite plusieurs types de graphes apportant des précisions sur la
-nature et la difficulté du problème viennent compléter ces premières
-courbes. Concernant les graphes de filaments, la moitié gauche du graphe
-correspond aux jours similaires au jour courant, tandis que la moitié
-droite affiche les lendemains : ce sont donc les voisinages tels
-qu'utilisés dans l'algorithme.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}1}]:} \PY{n}{library}\PY{p}{(}\PY{n}{talweg}\PY{p}{)}
-
- \PY{n}{P} \PY{o}{=} \PY{l+m+mi}{7} \PY{c+c1}{\PYZsh{}instant de prévision}
- \PY{n}{H} \PY{o}{=} \PY{l+m+mi}{17} \PY{c+c1}{\PYZsh{}horizon (en heures)}
-
- \PY{n}{ts\PYZus{}data} \PY{o}{=} \PY{n}{read}\PY{o}{.}\PY{n}{csv}\PY{p}{(}\PY{n}{system}\PY{o}{.}\PY{n}{file}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{extdata}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{pm10\PYZus{}mesures\PYZus{}H\PYZus{}loc\PYZus{}report.csv}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}
- \PY{n}{package}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{talweg}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{)}
- \PY{n}{exo\PYZus{}data} \PY{o}{=} \PY{n}{read}\PY{o}{.}\PY{n}{csv}\PY{p}{(}\PY{n}{system}\PY{o}{.}\PY{n}{file}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{extdata}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{meteo\PYZus{}extra\PYZus{}noNAs.csv}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{package}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{talweg}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{)}
- \PY{c+c1}{\PYZsh{} NOTE: \PYZsq{}GMT\PYZsq{} because DST gaps are filled and multiple values merged in above dataset.}
- \PY{c+c1}{\PYZsh{} Prediction from P+1 to P+H included.}
- \PY{n}{data} \PY{o}{=} \PY{n}{getData}\PY{p}{(}\PY{n}{ts\PYZus{}data}\PY{p}{,} \PY{n}{exo\PYZus{}data}\PY{p}{,} \PY{n}{input\PYZus{}tz} \PY{o}{=} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{GMT}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{working\PYZus{}tz}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{GMT}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{predict\PYZus{}at}\PY{o}{=}\PY{n}{P}\PY{p}{)}
-
- \PY{n}{indices\PYZus{}ch} \PY{o}{=} \PY{n}{seq}\PY{p}{(}\PY{k}{as}\PY{o}{.}\PY{n}{Date}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{2015\PYZhy{}01\PYZhy{}18}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{,}\PY{k}{as}\PY{o}{.}\PY{n}{Date}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{2015\PYZhy{}01\PYZhy{}24}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{,}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{days}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}
- \PY{n}{indices\PYZus{}ep} \PY{o}{=} \PY{n}{seq}\PY{p}{(}\PY{k}{as}\PY{o}{.}\PY{n}{Date}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{2015\PYZhy{}03\PYZhy{}15}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{,}\PY{k}{as}\PY{o}{.}\PY{n}{Date}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{2015\PYZhy{}03\PYZhy{}21}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{,}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{days}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}
- \PY{n}{indices\PYZus{}np} \PY{o}{=} \PY{n}{seq}\PY{p}{(}\PY{k}{as}\PY{o}{.}\PY{n}{Date}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{2015\PYZhy{}04\PYZhy{}26}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{,}\PY{k}{as}\PY{o}{.}\PY{n}{Date}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{2015\PYZhy{}05\PYZhy{}02}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}\PY{p}{,}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{days}\PY{l+s+s2}{\PYZdq{}}\PY{p}{)}
-\end{Verbatim}
-
- \subsection*{Pollution par chauffage}
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}2}]:} \PY{n}{p1} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ch}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{mix}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{FALSE}\PY{p}{)}
- \PY{n}{p2} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ch}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{endo}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{p3} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ch}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{none}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{p4} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ch}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Average}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{)}
- \PY{n}{p5} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ch}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Persistence}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{same\PYZus{}day}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
-\end{Verbatim}
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}3}]:} \PY{n}{e1} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e2} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e3} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e4} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p4}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e5} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p5}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{options}\PY{p}{(}\PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{width}\PY{o}{=}\PY{l+m+mi}{9}\PY{p}{,} \PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{height}\PY{o}{=}\PY{l+m+mi}{7}\PY{p}{)}
- \PY{n}{plotError}\PY{p}{(}\PY{n+nb}{list}\PY{p}{(}\PY{n}{e1}\PY{p}{,} \PY{n}{e5}\PY{p}{,} \PY{n}{e4}\PY{p}{,} \PY{n}{e2}\PY{p}{,} \PY{n}{e3}\PY{p}{)}\PY{p}{,} \PY{n}{cols}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{,}\PY{n}{colors}\PY{p}{(}\PY{p}{)}\PY{p}{[}\PY{l+m+mi}{258}\PY{p}{]}\PY{p}{,}\PY{l+m+mi}{4}\PY{p}{,}\PY{l+m+mi}{6}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} noir: Neighbors non\PYZhy{}local (p1), bleu: Neighbors local endo (p2),}
- \PY{c+c1}{\PYZsh{} mauve: Neighbors local none (p3), vert: moyenne (p4),}
- \PY{c+c1}{\PYZsh{} rouge: persistence (p5)}
-
- \PY{n}{sum\PYZus{}p123} \PY{o}{=} \PY{n}{e1}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices} \PY{o}{+} \PY{n}{e2}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices} \PY{o}{+} \PY{n}{e3}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices}
- \PY{n}{i\PYZus{}np} \PY{o}{=} \PY{n}{which}\PY{o}{.}\PY{n}{min}\PY{p}{(}\PY{n}{sum\PYZus{}p123}\PY{p}{)} \PY{c+c1}{\PYZsh{}indice de (veille de) jour \PYZdq{}facile\PYZdq{}}
- \PY{n}{i\PYZus{}p} \PY{o}{=} \PY{n}{which}\PY{o}{.}\PY{n}{max}\PY{p}{(}\PY{n}{sum\PYZus{}p123}\PY{p}{)} \PY{c+c1}{\PYZsh{}indice de (veille de) jour \PYZdq{}difficile\PYZdq{}}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_4_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- L'erreur absolue dépasse 20 sur 1 à 2 jours suivant les modèles (graphe
-en haut à droite). C'est au-delà de ce que l'on aimerait voir (disons
-+/- 5 environ). Sur cet exemple le modèle à voisins "contraint"
-(local=TRUE) utilisant des pondérations basées sur les similarités de
-forme (simtype="endo") obtient en moyenne les meilleurs résultats, avec
-un MAPE restant en général inférieur à 30\% de 8h à 19h (7+1 à 7+12 :
-graphe en bas à gauche).
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}4}]:} \PY{n}{options}\PY{p}{(}\PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{width}\PY{o}{=}\PY{l+m+mi}{9}\PY{p}{,} \PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{height}\PY{o}{=}\PY{l+m+mi}{4}\PY{p}{)}
- \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p3 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p3 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} Bleu: prévue, noir: réalisée}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_6_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_6_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_6_2.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Le jour "facile à prévoir", à gauche, se décompose en deux modes : un
-léger vers 10h (7+3), puis un beaucoup plus marqué vers 19h (7+12). Ces
-deux modes sont retrouvés par les trois variantes de l'algorithme à
-voisins, bien que l'amplitude soit mal prédite. Concernant le jour
-"difficile à prévoir" il y a deux pics en tout début et toute fin de
-journée (à 9h et 23h), qui ne sont pas du tout anticipés par le
-programme ; la grande amplitude de ces pics explique alors l'intensité
-de l'erreur observée.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}5}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}np1} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}p1} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{f\PYZus{}np2} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}p2} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_8_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_8_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Les voisins du jour courant (période de 24h allant de 8h à 7h le
-lendemain) sont affichés avec un trait d'autant plus sombre qu'ils sont
-proches. On constate dans le cas non contraint (en haut) une grande
-variabilité des lendemains, très nette sur le graphe en haut à droite.
-Ceci indique une faible corrélation entre la forme d'une courbe sur une
-période de 24h et la forme sur les 24h suivantes ; \textbf{cette
-observation est la source des difficultés rencontrées par l'algorithme
-sur ce jeu de données.}
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}6}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotFilamentsBox}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{FilBox p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotFilamentsBox}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{FilBox p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} En pointillés la courbe du jour courant + lendemain (à prédire)}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_10_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Sur cette boxplot fonctionnelle (voir la fonction fboxplot() du package
-R "rainbow") l'on constate essentiellement deux choses : le lendemain
-d'un voisin "normal" peut se révéler être une courbe atypique, fort
-éloignée de ce que l'on souhaite prédire (courbes bleue et rouge à
-gauche) ; et, dans le cas d'une courbe à prédire atypique (à droite) la
-plupart des voisins sont trop éloignés de la forme à prédire et forcent
-ainsi un aplatissement de la prédiction.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}7}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np2}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p2}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} Variabilité globale en rouge ; sur les 60 voisins (+ lendemains) en noir}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_12_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_12_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Ces graphes viennent confirmer l'impression visuelle après observation
-des filaments. En effet, la variabilité globale en rouge (écart-type
-heure par heure sur l'ensemble des couples "aujourd'hui/lendemain" du
-passé) devrait rester nettement au-dessus de la variabilité locale,
-calculée respectivement sur un voisinage d'une soixantaine de jours
-(pour p1) et d'une dizaine de jours (pour p2). Or on constate que ce
-n'est pas du tout le cas sur la période "lendemain", sauf en partie pour
-p2 le jour 4 \(-\) mais ce n'est pas suffisant.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}8}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} \PYZhy{} pollué à gauche, + pollué à droite}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_14_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_14_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Les poids se concentrent près de 0 dans le cas "non local" (p1), et se
-répartissent assez uniformément dans \([0,0.2]\) dans le cas "local"
-(p2). C'est ce que l'on souhaite observer pour éviter d'effectuer une
-simple moyenne.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}9}]:} \PY{c+c1}{\PYZsh{} Fenêtres sélectionnées dans ]0,7] / non\PYZhy{}loc 2 premières lignes, loc ensuite}
- \PY{n}{p1}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
- \PY{n}{p1}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
-
- \PY{n}{p2}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
- \PY{n}{p2}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
-\end{Verbatim}
-
- \begin{enumerate*}
-\item 0.168824188864717
-\item 0.336969608767438
-\end{enumerate*}
-
-
- \begin{enumerate*}
-\item 0.18004595760919
-\item 0.353963007643311
-\end{enumerate*}
-
-
- 1.16620655388085
-
-
- 1.18148881881259
-
-
- \subsection*{Pollution par épandage}
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}10}]:} \PY{n}{p1} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ep}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{mix}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{FALSE}\PY{p}{)}
- \PY{n}{p2} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ep}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{endo}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{p3} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ep}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{none}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{p4} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ep}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Average}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{)}
- \PY{n}{p5} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}ep}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Persistence}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{same\PYZus{}day}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
-\end{Verbatim}
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}11}]:} \PY{n}{e1} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e2} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e3} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e4} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p4}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e5} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p5}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{options}\PY{p}{(}\PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{width}\PY{o}{=}\PY{l+m+mi}{9}\PY{p}{,} \PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{height}\PY{o}{=}\PY{l+m+mi}{7}\PY{p}{)}
- \PY{n}{plotError}\PY{p}{(}\PY{n+nb}{list}\PY{p}{(}\PY{n}{e1}\PY{p}{,} \PY{n}{e5}\PY{p}{,} \PY{n}{e4}\PY{p}{,} \PY{n}{e2}\PY{p}{,} \PY{n}{e3}\PY{p}{)}\PY{p}{,} \PY{n}{cols}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{,}\PY{n}{colors}\PY{p}{(}\PY{p}{)}\PY{p}{[}\PY{l+m+mi}{258}\PY{p}{]}\PY{p}{,}\PY{l+m+mi}{4}\PY{p}{,}\PY{l+m+mi}{6}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} noir: Neighbors non\PYZhy{}local (p1), bleu: Neighbors local endo (p2),}
- \PY{c+c1}{\PYZsh{} mauve: Neighbors local none (p3), vert: moyenne (p4),}
- \PY{c+c1}{\PYZsh{} rouge: persistence (p5)}
-
- \PY{n}{sum\PYZus{}p123} \PY{o}{=} \PY{n}{e1}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices} \PY{o}{+} \PY{n}{e2}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices} \PY{o}{+} \PY{n}{e3}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices}
- \PY{n}{i\PYZus{}np} \PY{o}{=} \PY{n}{which}\PY{o}{.}\PY{n}{min}\PY{p}{(}\PY{n}{sum\PYZus{}p123}\PY{p}{)} \PY{c+c1}{\PYZsh{}indice de (veille de) jour \PYZdq{}facile\PYZdq{}}
- \PY{n}{i\PYZus{}p} \PY{o}{=} \PY{n}{which}\PY{o}{.}\PY{n}{max}\PY{p}{(}\PY{n}{sum\PYZus{}p123}\PY{p}{)} \PY{c+c1}{\PYZsh{}indice de (veille de) jour \PYZdq{}difficile\PYZdq{}}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_19_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Il est difficile dans ce cas de déterminer une méthode meilleure que les
-autres : elles donnent toutes de plutôt mauvais résultats, avec une
-erreur absolue moyennée sur la journée dépassant presque toujours 15
-(graphe en haut à droite).
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}12}]:} \PY{n}{options}\PY{p}{(}\PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{width}\PY{o}{=}\PY{l+m+mi}{9}\PY{p}{,} \PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{height}\PY{o}{=}\PY{l+m+mi}{4}\PY{p}{)}
- \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p3 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p3 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} Bleu: prévue, noir: réalisée}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_21_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_21_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_21_2.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Dans le cas d'un jour "facile" à prédire \(-\) à gauche \(-\) la forme
-est plus ou moins retrouvée, mais le niveau moyen est trop bas (courbe
-en bleu). Concernant le jour "difficile" à droite, non seulement la
-forme n'est pas anticipée mais surtout le niveau prédit est très
-inférieur au niveau de pollution observé. Comme on le voit ci-dessous
-cela découle d'un manque de voisins au comportement similaire.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}13}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}np1} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}p1} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{f\PYZus{}np2} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}p2} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_23_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_23_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Les observations sont les mêmes qu'au paragraphe précédent : trop de
-variabilité des lendemains (et même des voisins du jour courant).
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}14}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotFilamentsBox}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{FilBox p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotFilamentsBox}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{FilBox p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} En pointillés la courbe du jour courant + lendemain (à prédire)}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_25_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- On constate la présence d'un voisin au lendemain complètement atypique
-avec un pic en début de journée (courbe en vert à gauche), et d'un autre
-phénomène semblable avec la courbe rouge sur le graphe de droite. Ajouté
-au fait que le lendemain à prévoir est lui-même un jour "hors norme",
-cela montre l'impossibilité de bien prévoir une courbe en utilisant
-l'algorithme à voisins.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}15}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np2}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p2}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} Variabilité globale en rouge ; sur les 60 voisins (+ lendemains) en noir}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_27_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_27_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Comme précédemment les variabilités locales et globales sont confondues
-dans les parties droites des graphes \(-\) sauf pour la version "locale"
-sur le jour "facile" ; mais cette bonne propriété n'est pas suffisante
-si l'on ne trouve pas les bons poids à appliquer.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}16}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} \PYZhy{} pollué à gauche, + pollué à droite}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_29_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_29_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- En comparaison avec le pragraphe précédent on retrouve le même (bon)
-comportement des poids pour la version "non locale". En revanche la
-fenêtre optimisée est trop grande sur le jour "facile" pour la méthode
-"locale" (voir affichage ci-dessous) : il en résulte des poids tous
-semblables autour de 0.084, l'algorithme effectue donc une moyenne
-simple \(-\) expliquant pourquoi les courbes mauve et bleue sont très
-proches sur le graphe d'erreurs.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}17}]:} \PY{c+c1}{\PYZsh{} Fenêtres sélectionnées dans ]0,7] / non\PYZhy{}loc 2 premières lignes, loc ensuite}
- \PY{n}{p1}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
- \PY{n}{p1}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
-
- \PY{n}{p2}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
- \PY{n}{p2}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
-\end{Verbatim}
-
- \begin{enumerate*}
-\item 0.206604806619633
-\item 0.661854053860987
-\end{enumerate*}
-
-
- \begin{enumerate*}
-\item 0.367945958636072
-\item 0.244429852740092
-\end{enumerate*}
-
-
- 6.99993248587025
-
-
- 1.24825506305085
-
-
- \subsection*{Semaine non polluée}
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}18}]:} \PY{n}{p1} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}np}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{mix}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{FALSE}\PY{p}{)}
- \PY{n}{p2} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}np}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{endo}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{p3} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}np}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Neighbors}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{simtype}\PY{o}{=}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{none}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{local}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{p4} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}np}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Average}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{)}
- \PY{n}{p5} \PY{o}{=} \PY{n}{computeForecast}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{indices\PYZus{}np}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Persistence}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Zero}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,} \PY{n}{horizon}\PY{o}{=}\PY{n}{H}\PY{p}{,}
- \PY{n}{same\PYZus{}day}\PY{o}{=}\PY{n}{FALSE}\PY{p}{)}
-\end{Verbatim}
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}19}]:} \PY{n}{e1} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e2} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e3} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e4} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p4}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{e5} \PY{o}{=} \PY{n}{computeError}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p5}\PY{p}{,} \PY{n}{H}\PY{p}{)}
- \PY{n}{options}\PY{p}{(}\PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{width}\PY{o}{=}\PY{l+m+mi}{9}\PY{p}{,} \PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{height}\PY{o}{=}\PY{l+m+mi}{7}\PY{p}{)}
- \PY{n}{plotError}\PY{p}{(}\PY{n+nb}{list}\PY{p}{(}\PY{n}{e1}\PY{p}{,} \PY{n}{e5}\PY{p}{,} \PY{n}{e4}\PY{p}{,} \PY{n}{e2}\PY{p}{,} \PY{n}{e3}\PY{p}{)}\PY{p}{,} \PY{n}{cols}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{,}\PY{n}{colors}\PY{p}{(}\PY{p}{)}\PY{p}{[}\PY{l+m+mi}{258}\PY{p}{]}\PY{p}{,}\PY{l+m+mi}{4}\PY{p}{,}\PY{l+m+mi}{6}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} noir: Neighbors non\PYZhy{}local (p1), bleu: Neighbors local endo (p2),}
- \PY{c+c1}{\PYZsh{} mauve: Neighbors local none (p3), vert: moyenne (p4),}
- \PY{c+c1}{\PYZsh{} rouge: persistence (p5)}
-
- \PY{n}{sum\PYZus{}p123} \PY{o}{=} \PY{n}{e1}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices} \PY{o}{+} \PY{n}{e2}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices} \PY{o}{+} \PY{n}{e3}\PY{err}{\PYZdl{}}\PY{n+nb}{abs}\PY{err}{\PYZdl{}}\PY{n}{indices}
- \PY{n}{i\PYZus{}np} \PY{o}{=} \PY{n}{which}\PY{o}{.}\PY{n}{min}\PY{p}{(}\PY{n}{sum\PYZus{}p123}\PY{p}{)} \PY{c+c1}{\PYZsh{}indice de (veille de) jour \PYZdq{}facile\PYZdq{}}
- \PY{n}{i\PYZus{}p} \PY{o}{=} \PY{n}{which}\PY{o}{.}\PY{n}{max}\PY{p}{(}\PY{n}{sum\PYZus{}p123}\PY{p}{)} \PY{c+c1}{\PYZsh{}indice de (veille de) jour \PYZdq{}difficile\PYZdq{}}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_34_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Dans ce cas plus favorable les intensité des erreurs absolues ont
-clairement diminué : elles restent souvent en dessous de 5. En revanche
-le MAPE moyen reste au-delà de 20\%, et même souvent plus de 30\%. Comme
-dans le cas de l'épandage on constate une croissance globale de la
-courbe journalière d'erreur absolue moyenne (en haut à gauche) ; ceci
-peut être dû au fait que l'on ajuste le niveau du jour à prédire en le
-recollant sur la dernière valeur observée.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}20}]:} \PY{n}{options}\PY{p}{(}\PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{width}\PY{o}{=}\PY{l+m+mi}{9}\PY{p}{,} \PY{n+nb}{repr}\PY{o}{.}\PY{n}{plot}\PY{o}{.}\PY{n}{height}\PY{o}{=}\PY{l+m+mi}{4}\PY{p}{)}
- \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p3 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotPredReal}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p3}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{PredReal p3 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} Bleu: prévue, noir: réalisée}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_36_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_36_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_36_2.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- La forme est raisonnablement retrouvée pour les méthodes "locales",
-l'autre version lissant trop les prédictions. Le biais reste cependant
-important, surtout en fin de journée sur le jour "difficile".
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}21}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}np1} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}p1} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{f\PYZus{}np2} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{f\PYZus{}p2} \PY{o}{=} \PY{n}{computeFilaments}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{,} \PY{n}{plot}\PY{o}{=}\PY{n}{TRUE}\PY{p}{)}
- \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Filaments p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_38_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_38_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Les graphes de filaments ont encore la même allure, avec une assez
-grande variabilité observée. Cette observation est cependant trompeuse,
-comme l'indique plus bas le graphe de variabilité relative.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}22}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotFilamentsBox}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{FilBox p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotFilamentsBox}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{FilBox p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} En pointillés la courbe du jour courant + lendemain (à prédire)}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_40_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- On peut réappliquer les mêmes remarques qu'auparavant sur les boxplots
-fonctionnels : lendemains de voisins atypiques, courbe à prévoir
-elle-même légèrement "hors norme".
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}23}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p1}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}np2}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotRelVar}\PY{p}{(}\PY{n}{data}\PY{p}{,} \PY{n}{f\PYZus{}p2}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{StdDev p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} Variabilité globale en rouge ; sur les 60 voisins (+ lendemains) en noir}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_42_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_42_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Cette fois la situation idéale est observée : la variabilité globale est
-nettement au-dessus de la variabilité locale. Bien que cela ne suffise
-pas à obtenir de bonnes prédictions de forme, on constate au moins
-l'amélioration dans la prédiction du niveau.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}24}]:} \PY{n}{par}\PY{p}{(}\PY{n}{mfrow}\PY{o}{=}\PY{n}{c}\PY{p}{(}\PY{l+m+mi}{1}\PY{p}{,}\PY{l+m+mi}{2}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p1}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p1 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{p}{)}
- \PY{n}{plotSimils}\PY{p}{(}\PY{n}{p2}\PY{p}{,} \PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{;} \PY{n}{title}\PY{p}{(}\PY{n}{paste}\PY{p}{(}\PY{l+s+s2}{\PYZdq{}}\PY{l+s+s2}{Weights p2 day}\PY{l+s+s2}{\PYZdq{}}\PY{p}{,}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{p}{)}
-
- \PY{c+c1}{\PYZsh{} \PYZhy{} pollué à gauche, + pollué à droite}
-\end{Verbatim}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_44_0.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- \begin{center}
- \adjustimage{max size={0.9\linewidth}{0.9\paperheight}}{output_44_1.png}
- \end{center}
- { \hspace*{\fill} \\}
-
- Concernant les poids en revanche, deux cas a priori mauvais se cumulent
-: * les poids dans le cas "non local" ne sont pas assez concentrés
-autour de 0, menant à un lissage trop fort \(-\) comme observé sur les
-graphes des courbes réalisées/prévues ; * les poids dans le cas "local"
-sont trop semblables (à cause de la trop grande fenêtre optimisée par
-validation croisée, cf. ci-dessous), résultant encore en une moyenne
-simple \(-\) mais sur moins de jours, plus proches du jour courant.
-
- \begin{Verbatim}[commandchars=\\\{\}]
-{\color{incolor}In [{\color{incolor}25}]:} \PY{c+c1}{\PYZsh{} Fenêtres sélectionnées dans ]0,7] / non\PYZhy{}loc 2 premières lignes, loc ensuite}
- \PY{n}{p1}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
- \PY{n}{p1}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
-
- \PY{n}{p2}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}np}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
- \PY{n}{p2}\PY{err}{\PYZdl{}}\PY{n}{getParams}\PY{p}{(}\PY{n}{i\PYZus{}p}\PY{p}{)}\PY{err}{\PYZdl{}}\PY{n}{window}
-\end{Verbatim}
-
- \begin{enumerate*}
-\item 0.205055690975915
-\item 0.703482647754766
-\end{enumerate*}
-
-
- \begin{enumerate*}
-\item 1.1038650998802
-\item 0.885155748316133
-\end{enumerate*}
-
-
- 3.64336124381868
-
-
- 6.99994501761361
-
-
- \subsection*{Bilan}
-
-Nos algorithmes à voisins ne sont pas adaptés à ce jeu de données où la
-forme varie considérablement d'un jour à l'autre. Plus généralement
-cette décorrélation de forme rend ardue la tâche de prévision pour toute
-autre méthode \(-\) du moins, nous ne savons pas comment procéder pour
-parvenir à une bonne précision.
-
-Toutefois, un espoir reste permis par exemple en aggréger les courbes
-spatialement (sur plusieurs stations situées dans la même agglomération
-ou dans une même zone).
-
-
- % Add a bibliography block to the postdoc
-
-
-
- \end{document}
projects / talweg.git / blobdiff