def make_word_cloud(color):
data_arr = get_data_arr()
df = pd.DataFrame(data_arr, columns=['id', 'word', 'length'])
text_raw = " ".join(df['word'].drop_duplicates())
wordcloud = WordCloud(background_color=color,
stopwords=STOPWORDS).generate(text_raw)
wordcloud.to_file('words_cloud.png')
def get_stat_with_frequency():
data_arr = get_data_arr()
df = pd.DataFrame(data_arr, columns=['id', 'word', 'length'])
frequency_df = df.word.value_counts().to_frame()
a = frequency_df['word'].to_dict()
xval = []
yval = []
for key, value in a.items():
xval.append(value)
yval.append(key)
x = frequency_df['word']
e = x.mean()
frequency_df['word'] = round(x - e, 2)
frequency_df['colors'] = ['red' if x < 0 else 'green' for x in frequency_df['word']]
plt.figure(figsize=(200, 200), dpi=80)
plt.hlines(y=frequency_df.index, xmin=0, xmax=frequency_df.word)
for x, y, tex in zip(frequency_df.word, frequency_df.index, frequency_df.word):
t = plt.text(x, y, tex, horizontalalignment='right' if x < 0 else 'left',
verticalalignment='center', fontdict={'color': 'red' if x < 0 else 'green', 'size': 12})
plt.yticks(frequency_df.index, frequency_df.index, fontsize=12)
plt.title('Diverging Text Bars of Car Mileage', fontdict={'size': 20})
plt.grid(linestyle='--', alpha=0.5)
# plt.xlim(-2.5, 2.5)
plt.savefig('words_freq_bar.png')
def get_top_using_words():
data_arr = get_data_arr()
df = pd.DataFrame(data_arr, columns=['id', 'word', 'length'])
# print(len(df))
frequency_df = df.word.value_counts().to_frame()
frequency_df_without_bad_words = frequency_df[frequency_df['word'].apply(
lambda x: df.word.value_counts().mean() - 3 < x < df.word.value_counts().mean() + 3)]
frequency_df_with_bad_words = frequency_df[frequency_df['word'].apply(
lambda x: df.word.value_counts().mean() - 3 >= x or x >= df.word.value_counts().mean() + 3)]
return frequency_df_without_bad_words['word'].to_dict(), frequency_df_with_bad_words['word'].to_dict()
def draw_stat_length():
data_arr = get_data_arr()
figure, axes = plt.subplots(3, figsize=(50, 50))
ax1 = axes[0]
ax2 = axes[1]
ax3 = axes[2]
ax1.set_xlabel('Length', fontsize=20)
ax1.set_ylabel('Number words', fontsize=20)
# ax1.xaxis.set_major_locator(ticker.MultipleLocator(1))
# ax1.yaxis.set_major_locator(ticker.MultipleLocator(10))
ax2.set_xlabel('Number words', fontsize=20)
ax2.set_ylabel('Length', fontsize=20)
# ax2.xaxis.set_major_locator(ticker.MultipleLocator(10))
df = pd.DataFrame(data_arr, columns=['id', 'word', 'length'])
df.length.value_counts().sort_index(ascending=False).plot(ax=ax1,
grid=True,
fontsize=20)
extent = ax1.get_window_extent().transformed(
figure.dpi_scale_trans.inverted())
ax1.figure.savefig('saved_length_statistic1.png',
bbox_inches=extent.expanded(1.1, 1.2))
df.length.value_counts().sort_index(ascending=False).plot(ax=ax2,
kind='barh',
grid=True,
fontsize=20)
extent = ax2.get_window_extent().transformed(
figure.dpi_scale_trans.inverted())
ax2.figure.savefig('saved_length_statistic2.png',
bbox_inches=extent.expanded(1.1, 1.2))
df.length.value_counts().sort_index(ascending=False).plot(
ax=ax3,
kind='pie',
autopct='%1.1f%%',
shadow=True,
ylabel='Length',
grid=True,
fontsize=20)
extent = ax3.get_window_extent().transformed(
figure.dpi_scale_trans.inverted())
ax3.figure.savefig('saved_length_statistic3.png',
bbox_inches=extent.expanded(1.1, 1.2))
plt.close()
def get_next_prev_words(word):
data_arr = get_data_arr()
df = pd.DataFrame(data_arr, columns=['id', 'word', 'length'])
list_words = df['word'].to_list()
next_words = []
prev_words = []
if len(list_words) > 1 and list_words[0] == word:
next_words.append(list_words[1])
if len(list_words) > 1 and list_words[len(list_words) - 1] == word:
prev_words.append(list_words[len(list_words) - 2])
for i in range(1, len(list_words) - 1):
if list_words[i] == word:
next_words.append(list_words[i + 1])
prev_words.append(list_words[i - 1])
return prev_words, next_words
def find_abc_in_df():
data_arr = get_data_arr()
df = pd.DataFrame(data_arr, columns=['id', 'word', 'length'])
lang = 'abcdefghijklmnopqrstuvwxyz'
LANG = 'ZYXWVUTSRQPONMLKJIHGFEDCBA'
LANG = LANG[::-1]
kol_up = {}
kol_down = {}
for symb in lang:
kol_down[symb] = 0
for symb in LANG:
kol_up[symb] = 0
for expr in df['word']:
for symb in expr:
if symb in lang or symb in LANG:
if symb.isupper():
kol_up[symb] += 1
else:
kol_down[symb] += 1
return kol_up, kol_down