/
frame-cbsm-experiments-penn.tex
96 lines (90 loc) · 3.2 KB
/
frame-cbsm-experiments-penn.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
% SPDX-License-Identifier: CC-BY-4.0
% Copyright 2018 Toni Dietze
\documentclass[beamer]{standalone}
\input{preamble.tex}
\begin{document}
\begin{standaloneframe}[t]{\jobname}
\centering\small
\begin{tikzpicture}
[ data/.style = {align = center, draw, rounded corners}
, func/.style = {align = center, draw}
, node distance = 1.5em
]
\newcommand{\stacklist}
{ 1.25em/0.390625%
, 1.1580882352941178em/0.78125%
, 1.0546875em/1.5625%
, 0.9375em/3.125%
, 0.8035714285714286em/6.25%
, 0.6490384615384616em/12.5%
, 0.46875em/25.0%
, 0.2556818181818182em/50.0%
, 0.0em/100.0%
}
\node[data, text = HKS41K70] (ptb0) {Penn Treebank section 0 \\ (1\,921 trees)};
\node[data, right = 0.3333em of ptb0, text = HKS57K100] (ptb23) {Penn Treebank section 23 \\ (2\,416 trees)};
\node[data, below = 0.6666em of ptb23.south west, anchor = north west, xshift = 0.3333em + \pgflinewidth] (sents) {sequence of \\ sentences};
\node[func, right = of sents] (yield) {\strut yield};
\node[func, below = of sents] (parse) {\strut parse};
\node[data, right = of parse] (trees) {sequence \\ of trees};
\node[func, right = of trees] (evalb) {\strut evalb};
\foreach \t/\c in \stacklist {
\node[xshift = \t, yshift = \t, color = black!\c, fill = white, data] (ptas) at (ptb0 |- parse) {\strut sequence of pta\s};
\node[xshift = \t, yshift = \t, color = black!\c, fill = white, data, right = of evalb, text = HKS57K100] (Fmeas) {bracketing \\ F-measure};
}
\node[func, above = of ptas] (cbsm) {\strut cbsm};
\node[func, fit = (sents) (yield) (parse) (trees) (evalb)] {};
\begin{scope}[->, rounded corners]
\draw (ptb0) -- (cbsm) ;
\draw (cbsm) -- (ptas) ;
\draw (ptas) -- (parse);
\draw (sents) -- (parse);
\draw (parse) -- (trees);
\draw (yield) -- (sents);
\draw (trees) -- (evalb);
\draw (evalb) -- (Fmeas);
%\draw (yield |- ptb23.south) -- (yield);
%\draw (yield |- ptb23.south) -- ++(0, -0.8em) -| (evalb);
\draw (ptb23) -- (evalb |- ptb23.east) |- (yield);
\draw (ptb23) -| (evalb);
\end{scope}
%\node[below = 0.8em of ptas, draw, shape = ellipse, inner sep = 0.1em, fill = white] {\emph{for each pta}};
\end{tikzpicture}
\\[0.5em]
\only<2->{%
\begin{tikzpicture}
\pgfplotstableread[col sep = comma]{csv-07-evalb-wsj-2300-2320-len-restricted.csv}\dataEvalb
\begin{axis}
[ extra tick style = {grid = major, ticklabel pos = upper, tick style = {draw = none}}
, group/group size = 1 by 3
, group/vertical sep = 1.5cm
, group/xlabels at = edge bottom
, height = 11.5em
, scaled ticks = false
, set layers
, tick pos = lower
, width = 0.85\linewidth
, xlabel = {iteration (from 11\,729 to 11\,808)}
, xmin = 11726
, xmax = 11811
, xtick distance = 20
, extra x ticks = {11803}
, extra y ticks = {69.25, 71.06}
, only marks
, point meta min = 0
, point meta max = 1
, scatter
, scatter src = explicit
, ylabel = {Bracketing \\ F-measure [\%]}
, ylabel style = {align = left}
]
\addplot table
[ meta expr = \thisrow{Number of Skip sentence}/\thisrow{Number of sentence}
, x = iteration
, y = Bracketing FMeasure
] {\dataEvalb};
\end{axis}
\end{tikzpicture}%
}
\end{standaloneframe}
\end{document}