XWilliamY

#0
from googletrans import Translator
#1
translator = Translator()
#2
translated = translator.translate("Hello there, I like pie.")
#3
detected = translator.detect("Hello there, I like pie.")

XWilliamY

def keep_emoji_strip_nonalphanum(text):
    # split the text by emoji; multi-emoji emojis will be kept together
    split_on_emoji = emoji.get_emoji_regexp().split(text)
    
    # strip all whitespaces
    rejoined_no_whitespace = " ".join(split_on_emoji).split()
    
    # check each substring and apply strip symbols if it doesn't contain emojis
    cleaned = [strip_symbols(sub) if sub[0] not in emoji.UNICODE_EMOJI \
                                  else sub \

XWilliamY

# set readability and styling
sns.set(font_scale=1.4)
sns.set_style("whitegrid")
f = plt.figure(figsize=(12, 10), dpi=1000)

# logic to obtain and plot Replies Count
grouper = blackpink.groupby([pd.Grouper(freq='D')])
grouper['Replies Count'].sum().plot()

# more styling and labeling

XWilliamY

# get relevant indexes
text_lengths = copy_detected_langs['text_length']\
               .value_counts(normalize=True)\
               .sort_values(ascending=False)\
               .to_frame()[:20].index

# set styling
sns.set(font_scale=1.4)
sns.set_style("whitegrid")
fig, axes = plt.subplots(1, figsize=(15, 10), dpi=1000)

XWilliamY

target_y = 'predicted_conf'
sns.boxplot(x='predicted_lang', 
            y=target_y,
            data=copy_detected_langs[(copy_detected_langs['predicted_conf'] >= 0) & \
                                     (copy_detected_langs['predicted_lang'].isin(names))],
            order = names
           )

XWilliamY

XWilliamY

XWilliamY

XWilliamY

XWilliamY