Datasets:
File size: 9,618 Bytes
f9c9337 |
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 |
import json
import os
import matplotlib.pyplot as plt
import numpy as np
import random
import requests
import matplotlib
import matplotlib.font_manager as font_manager
from tqdm import tqdm
# Load a font from TTF file,
# relative to this Python module
# https://stackoverflow.com/a/69016300/315168
#https://github.com/satbyy/go-noto-universal/releases/tag/v7.0
font_path = os.path.join(os.path.dirname(__file__), 'GoNotoKurrent-Regular.ttf')
assert os.path.exists(font_path)
font_manager.fontManager.addfont(font_path)
prop = font_manager.FontProperties(fname=font_path)
# Set it as default matplotlib font
matplotlib.rc('font', family='sans-serif')
matplotlib.rcParams.update({
'font.size': 26,
'font.sans-serif': prop.get_name(),
})
def load_wordlist(lang):
#https://github.com/frekwencja/most-common-words-multilingual
os.makedirs("lists", exist_ok=True)
if os.path.exists(f"lists/{lang}.json"):
lines = json.load(open(f"lists/{lang}.json", "r"))
else:
data = requests.get(f"https://raw.githubusercontent.com/frekwencja/most-common-words-multilingual/main/data/wordfrequency.info/{lang}.txt").text # it's a file like object and works just like a file
lines = data.split("\n")
lines = lines[int(len(lines) / 2):]
json.dump(lines, open(f"lists/{lang}.json", "w"))
return lines
def generate_bar_plot_configurations(n):
configurations = []
for _ in range(n):
num_bars = random.randint(4, 8)
config = {
'num_bars': num_bars,
'orientation': random.choice(['vertical', 'horizontal']),
'values': [random.randint(30, 100) for _ in range(num_bars)],
'colors': random.sample(
['blue', 'green', 'red', 'yellow', 'black', 'orange', 'purple', 'brown'],
num_bars),
'figsize': (random.uniform(5, 10), random.uniform(3, 8))
}
configurations.append(config)
return configurations
def generate_pie_plot_configurations(n):
configurations = []
for _ in range(n):
num_slices = random.randint(4, 8)
config = {
'num_slices': num_slices,
'values': [random.randint(10, 200) for _ in range(num_slices)],
'labels': [f'Label {i + 1}' for i in range(num_slices)],
'colors': random.sample(
['blue', 'green', 'red', 'yellow', 'cyan', 'orange', 'purple', 'brown'],
num_slices),
'explode': [0.1 if random.random() > 0.8 else 0 for _ in range(num_slices)], # Randomly explode some slices
'figsize': (random.uniform(5, 10), random.uniform(5, 10))
}
configurations.append(config)
return configurations
def create_and_save_pie_plots(configurations, language, output_dir='pie_plots'):
word_list = load_wordlist(language)
annotations = []
if not os.path.exists(output_dir):
os.makedirs(output_dir)
for i, config in tqdm(enumerate(configurations)):
config["labels"] = random.sample(word_list, config["num_slices"])
plt.figure(figsize=config['figsize'])
plt.pie(config['values'], labels=config['labels'], colors=config['colors'], explode=config['explode'],
autopct='%1.1f%%', startangle=140)
plt.tight_layout()
plt.savefig(os.path.join(output_dir, f'pie_plot_{i}.png'))
plt.close()
half = int(config['num_slices']//2)
questions_name = [
f"What is the label of the biggest slice?",
f"What is the label of the smallest slice?",
f"What is the label of the {config['colors'][0]} slice?",
f"What is the label of the {config['colors'][-1]} slice?",
f"What is the label of the {config['colors'][half]} slice?",
]
answer_name = [
config["labels"][np.argmax(config['values'])],
config["labels"][np.argmin(config['values'])],
config["labels"][0],
config["labels"][-1],
config["labels"][half],
]
question_ground = [
f"Is the slice with label '{config['labels'][np.argmax(config['values'])]}' the biggest?",
f"Is the slice with label '{config['labels'][(np.argmax(config['values'])+half)%config['num_slices']]}' the biggest?",
f"Is the slice with label '{config['labels'][np.argmin(config['values'])]}' the smallest?",
f"Is the slice with label '{config['labels'][(np.argmin(config['values'])+half)%config['num_slices']]}' the smallest?",
f"Is the slice with label '{config['labels'][0]}' colored in {config['colors'][0]}?",
f"Is the slice with label '{config['labels'][0]}' colored in {config['colors'][0+half]}?",
f"Is the slice with label '{config['labels'][half]}' colored in {config['colors'][half]}?",
f"Is the slice with label '{config['labels'][half]}' colored in {config['colors'][-1]}?",
]
answer_ground = [
"yes", "no",
"yes", "no",
"yes", "no",
"yes", "no",
]
annotation = {
"image": f'pie_plot_{i}.png',
"question_ground": question_ground,
"answer_ground": answer_ground,
"questions_name": questions_name,
"answer_name": answer_name,
}
annotations.append(annotation)
json.dump(annotations, open(os.path.join(output_dir, f'pie_annotations_{language}.json'), 'w', encoding='utf-8'), indent=4)
def create_and_save_bar_plots(configurations, language, output_dir='bar_plots'):
word_list = load_wordlist(language)
annotations = []
if not os.path.exists(output_dir):
os.makedirs(output_dir)
for i, config in tqdm(enumerate(configurations)):
config["labels"] = random.sample(word_list, config["num_bars"])
plt.figure(figsize=config['figsize'])
if config['orientation'] == 'vertical':
plt.bar(config['labels'], config['values'], color=config['colors'])
plt.gcf().autofmt_xdate()
else:
plt.barh(config['labels'], config['values'], color=config['colors'])
plt.tight_layout()
plt.savefig(os.path.join(output_dir, f'bar_plot_{i}.png'))
plt.close()
half = int(config['num_bars']//2)
questions_name = [
f"What is the label of the biggest bar?",
f"What is the label of the smallest bar?",
f"What is the label of the {config['colors'][0]} bar?",
f"What is the label of the {config['colors'][-1]} bar?",
f"What is the label of the {config['colors'][half]} bar?",
]
answer_name = [
config["labels"][np.argmax(config['values'])],
config["labels"][np.argmin(config['values'])],
config["labels"][0],
config["labels"][-1],
config["labels"][half],
]
question_ground = [
f"Is the bar with label '{config['labels'][np.argmax(config['values'])]}' the biggest?",
f"Is the bar with label '{config['labels'][(np.argmax(config['values'])+half)%config['num_bars']]}' the biggest?",
f"Is the bar with label '{config['labels'][np.argmin(config['values'])]}' the smallest?",
f"Is the bar with label '{config['labels'][(np.argmin(config['values'])+half)%config['num_bars']]}' the smallest?",
f"Is the bar with label '{config['labels'][0]}' colored in {config['colors'][0]}?",
f"Is the bar with label '{config['labels'][0]}' colored in {config['colors'][0+half]}?",
f"Is the bar with label '{config['labels'][half]}' colored in {config['colors'][half]}?",
f"Is the bar with label '{config['labels'][half]}' colored in {config['colors'][-1]}?",
]
answer_ground = [
"yes", "no",
"yes", "no",
"yes", "no",
"yes", "no",
]
annotation = {
"image": f'bar_plot_{i}.png',
"question_ground": question_ground,
"answer_ground": answer_ground,
"questions_name": questions_name,
"answer_name": answer_name,
}
annotations.append(annotation)
json.dump(annotations, open(os.path.join(output_dir, f'bar_annotations_{language}.json'), 'w', encoding='utf-8'), indent=4)
if __name__ == '__main__':
languages = ["en", "zu", "id", "it", "de", "th", "ar", "ko", "zh-CN", "ru", "hi"]
if not os.path.exists("bar_configs.json"):
bar_configs = generate_bar_plot_configurations(50)
json.dump(bar_configs, open("bar_configs.json", "w"))
else:
bar_configs = json.load(open("bar_configs.json"))
if not os.path.exists("pie_configs.json"):
pie_configs = generate_pie_plot_configurations(50)
json.dump(pie_configs, open("pie_configs.json", "w"))
else:
pie_configs = json.load(open("pie_configs.json"))
for language in languages:
print(language)
os.makedirs(f"/media/gregor/DATA/datasets/smpqa/{language}", exist_ok=True)
create_and_save_bar_plots(bar_configs, language, output_dir=f"/media/gregor/DATA/datasets/smpqa/{language}")
create_and_save_pie_plots(pie_configs, language, output_dir=f"/media/gregor/DATA/datasets/smpqa/{language}")
|