Python STOPWORDS.union方法代码示例

# 需要导入模块: from wordcloud import STOPWORDS [as 别名]
# 或者: from wordcloud.STOPWORDS import union [as 别名]
    def config_stopwords(self, more_stopwords=None):
        """
          (obj) -> None

          Configuring stopwords by adding more if required
        """

        if more_stopwords is not None:
            self.STOPWORDS = STOPWORDS.union(more_stopwords)

# 需要导入模块: from wordcloud import STOPWORDS [as 别名]
# 或者: from wordcloud.STOPWORDS import union [as 别名]
def title_wordcloud(dataFrame):
    from wordcloud import WordCloud, STOPWORDS
    from PIL import Image
    #WordCloud Visualization
    text = " ".join(list(dataFrame['track_name']))
    STOPWORDS = STOPWORDS.union(["feat","Remix","Edit","Radio","Version","Mix","Remastered"])
    spotify_mask = np.array(Image.open(path.join( "spotify-logo.jpg")))
    wordcloud = WordCloud(width=2880, height=1800,background_color="white",
                          stopwords=STOPWORDS,mask = spotify_mask).generate(text)
    # Open a plot of the generated image.
    plt.figure( figsize=(10,6))
    plt.imshow(wordcloud)
    plt.axis("off")
    plt.tight_layout(pad=0)
    plt.savefig("project3_wordcloud.png")
    plt.show()

# 需要导入模块: from wordcloud import STOPWORDS [as 别名]
# 或者: from wordcloud.STOPWORDS import union [as 别名]
from wordcloud import WordCloud
from wordcloud import STOPWORDS
import matplotlib.pyplot as plt
import re
filename="/Users/chunmeiGao/Documents/Dataincubator/emailsubject.txt"

# Read the whole text.
text = open(filename).read()
print text
text=re.sub('Re:', '', text)
text=re.sub('RE:', '', text)
text=re.sub('FW:', '', text)
text=re.sub('Fwd:', '', text)
text=re.sub('Enron', '', text)
more_stopwords = {'X', 'Re', 'Fwd','ENRON','NA','FW'}
STOPWORDS = STOPWORDS.union(more_stopwords)
# Generate a word cloud image
wordcloud = WordCloud(stopwords=STOPWORDS).generate(text)

# Display the generated image:
# the matplotlib way:

plt.imshow(wordcloud)
plt.axis("off")

# take relative word frequencies into account, lower max_font_size
wordcloud = WordCloud(max_font_size=40, relative_scaling=.5).generate(text)
plt.figure()
plt.imshow(wordcloud)
plt.axis("off")
plt.show()

# 需要导入模块: from wordcloud import STOPWORDS [as 别名]
# 或者: from wordcloud.STOPWORDS import union [as 别名]
 def stop_words_configs(self):
   """
     Configuring stopwords by adding more if required
   """
   more_stopwords = {'innojam', 'video', 'cebit2014'}
   self.STOPWORDS = STOPWORDS.union(more_stopwords)

# 需要导入模块: from wordcloud import STOPWORDS [as 别名]
# 或者: from wordcloud.STOPWORDS import union [as 别名]
   pessoa = sentenca[0] 
   frase = "".join(sentenca[1:])
   
   falas.append(dict(pessoa=pessoa, frase=frase))


declarante = [
   fala['frase'].decode('utf-8') 
   for fala in falas 
   if fala['pessoa'] == 'Declarante'
]

declarante = "".join(declarante).lower()
declarante = declarante.replace(u"não sei", u"nãosei")

swords = ['que', 'eu', u'não', 'da', 'de', 'por', 'ele', u'você', u'está',
          'tem', 'um', 'uma', 'se', 'foi', u'lá', 'pra', 'para', 'vai', 
          u'já', 'na', 'era', 'em', u'aí', 'minha', u'nós', 'os', 'as', 
          'ou', 'essa', 'isso', 'como', 'aqui', 'pois', u'só', 'quando', 
          u'então', 'muito', 'porque', 'acho', 'nem', 'mais', 'meu',
          'ser', 'estou', 'vou', 'coisa', 'tenho', 'tinha', 'ter', u'quem'
          'fui', 'mas', u'são', 'muita', 'mim', 'tudo', 'toda', 'todo', 
          'deve', 'falar', 'eles', 'das']

STOPWORDS = STOPWORDS.union(swords)

wordcloud = WordCloud(width=800, height=400,
                      stopwords=STOPWORDS).generate(declarante)

wordcloud.to_file("wordcloud.png")

# 需要导入模块: from wordcloud import STOPWORDS [as 别名]
# 或者: from wordcloud.STOPWORDS import union [as 别名]
# Twitter API docs:
# https://dev.twitter.com/docs/api/1/get/search
#-----------------------------------------------------------------------
query = twitter.search.tweets(q = "modi", count=5000) #, until='2016-01-07')

#-----------------------------------------------------------------------
# How long did this query take?
#-----------------------------------------------------------------------
print ("Search complete (%.3f seconds)" % (query["search_metadata"]["completed_in"]))

#-----------------------------------------------------------------------
# Loop through each of the results, and print its content.
#-----------------------------------------------------------------------
#for result in query["statuses"]:
#	print ("(%s) @%s %s" % (result["created_at"], result["user"]["screen_name"], result["text"]))

# make a corpus from the list of tweets
status_list = [ result['text'] for result in query['statuses']]
corpus = ' '.join(status_list)

# read in the image and colors and plot the word cloud
img = Image.open("modi.jpg")
#img = img.resize((980,1080), Image.ANTIALIAS)
modi_coloring = np.array(img)
image_colors = ImageColorGenerator(modi_coloring)
#hcmask = scipy.ndimage.zoom(hcmask, 2, order=3)
STOPWORDS = STOPWORDS.union({"http","https","t","co","rt","since","towards","now","ok","okay","tag", "amp"})
#wc = WordCloud(background_color="white", max_words=2000, mask=hcmask, stopwords=STOPWORDS)
wc = WordCloud(font_path='cabin-sketch.bold.ttf', background_color="white", max_words=2000, mask=modi_coloring, color_func=image_colors, stopwords=STOPWORDS)
wc.generate(corpus)
wc.to_file("wc_color.png")

# 需要导入模块: from wordcloud import STOPWORDS [as 别名]
# 或者: from wordcloud.STOPWORDS import union [as 别名]
def info(request):	
	if request.method == 'GET' and 'screen_name' in request.GET:
		scn = request.GET['screen_name']
		if scn == "":
			return render(request, 'tweets/login.html', {'message':'Enter a valid Twitter handle'})
		else:
			# print("start", datetime.now())
			STAT_PATH = os.path.join(settings.BASE_DIR, 'tweets/static/tweets/')

			handle = tweepy.OAuthHandler(settings.CONSUMER_KEY, settings.CONSUMER_SECRET)
			handle.set_access_token(settings.ACCESS_TOKEN, settings.ACCESS_TOKEN_SECRET)
			api = tweepy.API(handle)
			try:
				user = api.get_user(screen_name=scn)
			except tweepy.TweepError:
				return render(request, 'tweets/login.html', {'message':'Enter a valid Twitter handle'})


			"""
			Tweets vs Weekday graph 
			"""
			try:
				timeline = api.user_timeline(screen_name=scn, count=3200, include_rts=True)
				mid = timeline[-1].id - 1
				while True:
					tl = api.user_timeline(screen_name=scn, count=3200, include_rts=True, max_id=mid)
					if not len(tl):
						break
					timeline += tl
					mid = tl[-1].id - 1
			except tweepy.TweepError or TypeError:
				return render(request, 'tweets/login.html', {'message':'Enter a valid Twitter handle'})

			y = [0]*7
			for tw in timeline:
				d = tw.created_at.strftime("%w")
				y[int(d)] += 1

			x = [0,1,2,3,4,5,6]
			xpoints = ['Sun', 'Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat']
			plt.xticks(x, xpoints)
			plt.plot(x, y, 'b-')
			plt.xlabel('Days of week')
			plt.ylabel('No. of tweets')
			
			path_graph = STAT_PATH + 'graph.png'
			if os.path.isfile(path_graph):
				os.remove(path_graph)

			plt.savefig(path_graph, dpi=300, bbox_inches='tight')
			plt.clf()

			"""
			Tag-cloud
			"""
			# more stopwords
			file = open(STAT_PATH + 'stopwords.txt', 'r')
			more_stops = file.readlines()
			for i in range(len(more_stops)):
				more_stops[i] = more_stops[i].rstrip('\n')
			global STOPWORDS
			STOPWORDS = STOPWORDS.union(more_stops)

			words = []
			matrix = []
			for tw in timeline:
				matrix.append(tw.text.split())
				words = words + tw.text.split()
			
			long_tweet_stripd = ""
			for w in words:
				if w != 'RT' and not(w.startswith('http')) and not(w.startswith('@')) and not(w.startswith('#')) and not(w.lower() in STOPWORDS):
					long_tweet_stripd = " ".join([long_tweet_stripd, w.lower()])
					
			un_words = long_tweet_stripd.split()

			mask = imread(STAT_PATH + 'twitter_mask.png')

			wcloud = WordCloud(max_words=50, background_color='white', stopwords=STOPWORDS, mask=mask).generate(long_tweet_stripd)
			
			# print(long_tweet_stripd)

			path_wordcloud = STAT_PATH + 'wordcloud.png'
			if os.path.isfile(path_wordcloud):
				os.remove(path_wordcloud)

			plt.imshow(wcloud)
			plt.gca().invert_yaxis()
			plt.axis('off')
			plt.savefig(path_wordcloud, dpi=600, bbox_inches='tight')
			plt.clf()
			plt.close()

			"""
			Word co-occurences matrix

			"""
			all_words_use = []
			for w in un_words:
				try:
#.........这里部分代码省略.........