-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcount_angles.R
232 lines (178 loc) · 7.13 KB
/
count_angles.R
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
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
# install.packages("ggplot2")
# install.packages("tidyverse")
# install.packages("gridExtra")
# install.packages("RColorBrewer")
# install.packages("reshape")
# install.packages("plyr")
# install.packages("FunChisq")
# install.packages("REdaS")
# install.packages("Morpho")
# install.packages("scales")
library(ggplot2)
library(tidyverse)
library(gridExtra)
library(RColorBrewer)
library(reshape)
library(plyr)
library(FunChisq)
library(REdaS)
library(Morpho)
library(scales)
# langs_short <- c('basque', 'eng', 'finnish', 'french', 'german', 'greek_modern', 'hebrew_modern', 'hindi', 'indonesian', 'japanese', 'korean', 'mandarin', 'persian', 'russian', 'spanish', 'tagalog', 'thai', 'turkish', 'vietnamese')
# langs_full <- c('Basque', 'English', 'Finnish', 'French', 'German', 'Greek', 'Hebrew', 'Hindi', 'Indonesian', 'Japanese', 'Korean', 'Mandarin', 'Persian', 'Russian', 'Spanish', 'Tagalog', 'Thai', 'Turkish', 'Vietnamese')
# langs_short <- c('hebrew_modern', 'hindi', 'indonesian', 'japanese', 'korean', 'mandarin', 'persian', 'tagalog', 'thai', 'turkish', 'vietnamese')
# langs_full <- c('Hebrew', 'Hindi', 'Indonesian', 'Japanese', 'Korean', 'Mandarin', 'Persian', 'Tagalog', 'Thai', 'Turkish', 'Vietnamese')
langs_short <- c('eng', 'fin', 'tur')
langs_full <- c('English', 'Finnish', 'Turkish')
# langs_short <- c('Bagirmi_bmi', 'Burushaski_bsk', 'Dani_LowerGrandValley_dni',
# 'Imonda_imn', 'Kayardild_gyd', 'Lavukaleve_lvk',
# 'Makah_myh', 'Martuthunira_vma', 'Maybrat_ayz',
# 'Ngiyambaa_wyb', 'Piraha_myp', 'Rama_rma', 'Tiwi_tiw')
# langs_full <- langs_short
Language <- c(NA)
Angle <- c(NA)
Mode <- c(NA)
Setting <- c(NA)
new_results <- data.frame(Language, Angle, Mode, Setting)
language <- c(NA)
k <- c(NA)
b <- c(NA)
rad_angle <- c(NA)
deg_angle <- c(NA)
results <- data.frame(language, k, b, rad_angle, deg_angle)
Var1 <- c(NA)
Var2 <- c(NA)
Freq <- c(NA)
lang <- c(NA)
new_angle <- data.frame(Var1, Var2, Freq, lang)
x <- c(NA)
values <- c(NA)
diagonals <- c(NA)
lang <- c(NA)
new_data_fun <- data.frame(x, values, diagonals, lang)
angles <- rep(c(0),times=100)
settings <- c('bpe-mr', 'bpe04V', 'manual', 'morf', 'spm', 'wordpiece')
for (s in 1:length(settings)) {
setting <- settings[s]
for (l in 1:length(langs_short)) {
print(langs_short[l])
df <- read.csv(paste('../baseline_files/aalto_3lang/5_random_seeds/5/', setting, '/', langs_short[l], '_baseline.csv', sep=''), sep = '\t')
df <- df[grepl(',', df$segments_lengths),]
lang_short <- langs_short[l]
lang_full <- langs_full[l]
for (iter in 1:100) {
# print(iter)
# df1 <- df[df$language == lang_short,]
df1 <- df
freqs <- data.frame(table(df1$index, df1$word_length))
freqs <- freqs[!(freqs$Freq == 0),]
freqs$lang <- lang_short
freqs$Var1 <- as.integer(as.character(freqs$Var1))
freqs$Var2 <- as.integer(as.character(freqs$Var2))
freqs <- subset(freqs, select = c(1,2,3))
freqs <- add.noise(freqs, 0.0001, "house", 2)
freqs <- as.data.frame(freqs)
names(freqs)[names(freqs)=="V1"] <- "Var1"
names(freqs)[names(freqs)=="V2"] <- "Var2"
names(freqs)[names(freqs)=="V3"] <- "Freq"
p <- ggplot(freqs,
aes(x = Var2,
y = Var1)) +
labs(title = paste(lang_full, ', ', lang_short, sep = ''),
x = 'word length in characters',
y = 'unevenness index',
size = 'frequency of\nthe unevenness index\nfor the given word length') +
geom_point(aes(size = Freq)) +
scale_size(labels = comma) +
scale_x_continuous(limits = c(0, 50)) +
scale_y_continuous(limits = c(0, 50)) +
geom_density_2d_filled(alpha=.5, bins=3, color='blue') +
guides(size = guide_legend(order=1),
levels = guide_legend(order=2))
ggbld <- ggplot_build(p)
gdata <- ggbld$data[[2]]
sub <- gdata[gdata$fill == '#21908CFF',]
sub %>%
ggplot() +
geom_point(aes(x, y)) +
geom_polygon(aes(x, y))
# find 2 data points
# find x, where y is max
max_y <- max(sub$y)
max_x <- sub[sub$y == max_y,]$x[[1]]
# find starting point for y
#mid_y <- (max_y - 2) / 2
max_right_x <- max(sub$x)
max_right_y <- sub[sub$x == max_right_x,]$y[[1]]
x1 <- max_x
y1 <- max_y
x2 <- max_right_x
y2 <- max_right_y
df_line <- data.frame(x1, y1)
df_line <- rbind(df_line, c(x2, y2))
X <- matrix(c(x1, 1,
x2, 1), 2, 2, byrow=TRUE)
y <- c(y1, y2)
coef <- solve(X, y)
# extract k, b
k <- coef[1]
b <- coef[2]
theta1 <- atan(1)
theta2 <- atan(k)
rad_angle <- pi - abs(theta1 - theta2)
# radians to degrees
deg_angle <- rad2deg(rad_angle)
# print(deg_angle)
# plot angle
y <- function(x) { x - 2 } # Create own functions
right_diagonal <- function(x) { k*x + b}
curve(y, from = 0, to = 100, col = 2) # Draw Base R plot
curve(right_diagonal, from = 0, to = 100, col = 3, add = TRUE)
data_fun <- data.frame(x = - 100:100, # Create data for ggplot2
values = c(y(- 100:100),
right_diagonal(- 100:100)),
diagonals = rep(c("left diagonal", "right diagonal"), each = 201))
data_fun$lang <- lang_short
p2 <- ggplot(freqs,
aes(x = Var2,
y = Var1)) +
ggtitle(paste(lang_full, '\nBaseline: ', setting, sep = '')) +
labs(x = 'word length in characters',
size = 'frequency of\nthe unevenness index\nfor the given word length',
colour = 'angle',
levels = 'density level') +
geom_point(aes(size = Freq)) +
scale_size(labels = comma) +
scale_x_continuous(limits = c(0, 30)) +
scale_y_continuous(limits = c(0, 30)) +
geom_density_2d_filled(alpha=.5, bins=3, color='blue') +
geom_line(data = data_fun, aes(x, values, col = diagonals)) +
scale_color_manual(values=c("bisque", "bisque")) +
theme(
axis.title.y = element_blank(),
legend.position = 'none',
plot.title=element_text(face='bold', color='white', size=18, hjust=0.08, vjust=-15, margin = margin(t=-30,b=0)))
# ggsave(paste('../plots/bpe-min-r/', lang_full, '_', iter, '.png', sep = ''),
# plot = p2,
# width = 9, height = 5)
# p2
results <- rbind(results, c(lang_full, k, b, rad_angle, deg_angle))
freqs$lang <- paste(lang_full, ', ', lang_short, sep = '')
new_angle <- rbind(new_angle, freqs)
data_fun$lang <- paste(lang_full, ', ', lang_short, sep = '')
new_data_fun <- rbind(new_data_fun, data_fun)
angles[iter] <- deg_angle
}
p2
closest<-function(xv,sv){
xv[which(abs(xv-sv)==min(abs(xv-sv)))] }
closest(angles, mean(angles))
index <- which(abs(angles-mean(angles))==min(abs(angles-mean(angles))))
print(lang_full)
print(index)
print(mean(angles))
new_results[nrow(new_results) + 1,] <- c(lang_full, round(mean(angles), digits = 5), 'Baseline', setting)
}
}
na.omit(new_results)
new_results