Python WordCloud.generate方法代码示例

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def wordcloud(datafile):

    #remove stop words, the most common words in a language
    vectorizer=CountVectorizer(stop_words='english')

    for word in vectorizer.get_stop_words():
        STOPWORDS.add(word)
    STOPWORDS.add("said")

    pony_mask = np.array(Image.open("../pinkyB.jpg"))
    wc = WordCloud(background_color="black", max_words=2000, mask=pony_mask, stopwords=STOPWORDS)

    #init dictionary with the five categories
    categoriesSet = set(datafile["Category"])
    categoriesDict = dict.fromkeys(categoriesSet,"")

    #Conditional Selection
    # business = datafile.ix[datafile["Category"]=="Business"]
    # print business["Content"].size

    #fill index with data from cv
    for index, row in datafile.iterrows():
        categoriesDict[row["Category"]] += str(row["Content"])

    for category, text in categoriesDict.iteritems():
        wc.generate(text)
        image = wc.to_image()
        image.save("../wordcloud/wordcloud_" + category + ".jpg")
    return

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def generateWordCloud(node, contribs, wordsToShow=None, normalize=True, normMin=0, normMax=1):
    contrib = contribs[node]
    
    if (normalize):
        contrib = normalizeContrib(contrib, normMin, normMax)
    
    # generate text
    text = generateText(contrib, wordsToShow)
    
    # load mask
    d = path.dirname(__file__)
    circle_mask = imread(path.join(d, "black_circle_mask_whitebg.png"))
    
    # gen word cloud
    wc = WordCloud(background_color="white", max_words=2000, mask=circle_mask)
    wc.generate(text)

    # store to file
    wc.to_file(path.join(d, "node.png"))

    # show
    useColorFunc = False #normalize
    if (useColorFunc):
        plt.imshow(wc.recolor( color_func=pos_neg_color_func ))
    else:
        plt.imshow(wc)
        
    plt.axis("off")
    plt.show()

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def txt2pic(txt_file, out_png, font_path, mask_file):
  text_address = path.abspath(txt_file)
  text = open(text_address).read()   #读取文本
  text_cut = jieba.cut(text)   #分词
  new_textlist = ' '.join(text_cut)   #组合
  pic_address = path.abspath(mask_file)
  pic = imread(pic_address)  #读取图片
  pic_color = ImageColorGenerator(pic)   #根据图片生成颜色函数
  wc = WordCloud(background_color='white',    #构造wordcloud类
    mask=pic,
    width = 750,
    height = 750,
    max_font_size = 80,
    random_state=30,
    font_path=font_path,
    max_words=500,
    min_font_size=2,
    color_func=pic_color
  )
  wc.generate(new_textlist)    #生成词云图
  plt.figure()    #画图
  plt.imshow(wc)
  plt.axis("off")
  plt.show()
  wc.to_file(out_png)   #保存图片

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def main():
	wr=WordReader()
	# wlist=wr.word_reader('data1/dt01.txt')

	wcount=''
	for root,dirs,files in os.walk('data2'):
		for file in files:
			file_path=os.path.join(root,file)
			wlist=wr.word_reader(file_path)
			wcount+=wlist

	back_coloring = np.array(Image.open("./sky.png"))
	wc = WordCloud(
                background_color="white", #背景颜色  
                max_words=1000,# 词云显示的最大词数  
                mask=back_coloring,#设置背景图片  
                max_font_size=150, #字体最大值  
                random_state=42,  
                )
                
	wc.generate(wcount) 
	# 
	# wc.generate_from_frequencies(word_list)
	# wc.fit_words(word_list)
	plt.figure() 
	plt.imshow(wc)  
	plt.axis("off")
	plt.show()  

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def run_yt():
    yt = ds.acquire_youtube()
    ytimg = imread("ytlogo.png")
    wc = WordCloud(mask=ytimg)
    image_colors = ImageColorGenerator(ytimg)
    wc.generate(word_list_yt(ds.mean(yt[0]), ds.standard_deviation(yt[0])))
    
    plt.imshow(wc.recolor(color_func = image_colors))
    plt.axis("off")
    plt.savefig('popularWordsYT.png', bbox_inches = 'tight', dpi = 200)
    
    words, vidcount = ds.word_count_yt('title', ds.mean(yt[0]), ds.standard_deviation(yt[0]))
    data1 = []
    data2 = []
    labels = []
    count = 0
    for w in sorted(words, key=words.get, reverse=True):   
        labels.append(w)
        data1.append(1000 * words[w]/vidcount)
        count +=1
        if count == 10:
            break
    words, vidcount = ds.word_count_dailymotion( 0, 0 )
    for w in labels:
        data2.append(1000 * words[w]/vidcount)
    create_dualbargraph(data1, data2, labels, 'wordUseCompYT.png')
    Theta = da.yt_thetas()
    for x in xrange(len(Theta)):
        Theta[x] = Theta[x]/10000
    print Theta
    create_bargraph(Theta,('duration', 'date created', 'y-intercept'), 'barGraphYT.png')

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def test_repeat():
    short_text = "Some short text"
    wc = WordCloud(stopwords=[]).generate(short_text)
    assert_equal(len(wc.layout_), 3)
    wc = WordCloud(max_words=50, stopwords=[], repeat=True).generate(short_text)
    # multiple of word count larger than max_words
    assert_equal(len(wc.layout_), 51)
    # relative scaling doesn't work well with repeat
    assert_equal(wc.relative_scaling, 0)
    # all frequencies are 1
    assert_equal(len(wc.words_), 3)
    assert_array_equal(list(wc.words_.values()), 1)
    frequencies = [w[0][1] for w in wc.layout_]
    assert_array_equal(frequencies, 1)
    repetition_text = "Some short text with text"
    wc = WordCloud(max_words=52, stopwords=[], repeat=True)
    wc.generate(repetition_text)
    assert_equal(len(wc.words_), 4)
    # normalized frequencies
    assert_equal(wc.words_['text'], 1)
    assert_equal(wc.words_['with'], .5)
    assert_equal(len(wc.layout_), wc.max_words)
    frequencies = [w[0][1] for w in wc.layout_]
    # check that frequencies are sorted
    assert_true(np.all(np.diff(frequencies) <= 0))

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def generate_cloud():
    d = path.dirname(__file__)
    janice = open(path.join(d, 'messages.txt')).read()
    group_mask = misc.imread(path.join(d, "mask.png"), flatten=True)
    wc = WordCloud(background_color="white", max_words = 2000, mask=group_mask)
    wc.generate(text)
    wc.to_file(path.join(d, "masked.jpg"))

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
    def wordcloudplot_focus(self, yizhongzhazha=None, backimage=None):
        """Do wordcloud plots for contacts. need to run relationship()
        first to get self._relationship.
        Parameters
            yizhongzhazha: pandas object by loading the data
            backimage: background image file's directory

        Returns: basic word cloud plots saved in files
        """
        if yizhongzhazha is None:
            print("Need load message table first.")
            return
        if self._contacts_topN is None:
            print("need to run relationship() first.")
            return
        if backimage is not None:
            custompic = imread(backimage)
        else:
            custompic = None
        if not os.path.exists('./wordcloud'):
            os.makedirs('./wordcloud')
        wordcloud = WordCloud(background_color="white", mask=custompic,
                              max_words=2000,scale=3)
        for k in range(len(self._contacts_topN)):
            text=self._relationship.iloc[:,k]
            text_to_wordcloud=[]
            for i in range(len(text)):
                text_to_wordcloud.append((text.index.values[i]+' ')*text[i])
            text=''.join(text_to_wordcloud)
            wordcloud.generate(text)
            wordcloud.to_file("./wordcloud/"+self._relationship.columns[k]+'2.png')

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def genwordcloud(texts,mask=None,font_path=None,background_color='white'):
    '''生成词云
    parameter
    ----------
    mask: RGBA模式数组，最后一个分量是alpha通道, 默认会生成一个900*1200的椭圆
    font_path: 采用的字体，建议采用安卓默认字体DroidSansFallback.ttf
    
    return
    -------
    img:可以直接img.save('test.png')
    '''
    from PIL import Image
    try:
        from wordcloud import WordCloud
    except:
        #raise Exception('wordcloud need install wordcloud package.')
        print('wordcloud need install wordcloud package.')
        return None
    if mask is None:
        tmp=np.zeros((900,1200),dtype=np.uint8)
        for i in range(tmp.shape[0]):
            for j in range(tmp.shape[1]):
                if (i-449.5)**2/(430**2)+(j-599.5)**2/(580**2)>1:
                    tmp[i,j]=255
        mask=np.zeros((900,1200,4),dtype=np.uint8)
        mask[:,:,0]=tmp
        mask[:,:,1]=tmp
        mask[:,:,2]=tmp
        mask[:,:,3]=255
    else:
        mask=np.array(Image.open(mask))
    wordcloud = WordCloud(background_color = background_color,font_path=font_path, mask = mask)
    wordcloud.generate(texts)
    img=wordcloud.to_image()
    return img

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def run_dm():
    dm = ds.acquire_dailymotion()
    dmimg = imread("dmlogo.png")
    # Read the whole text.
    wc = WordCloud(mask=dmimg)
    image_colors = ImageColorGenerator(dmimg)
    wc.generate(word_list_dailymotion(ds.mean(dm[0]), ds.standard_deviation(dm[0])))
    
    
    # Open a plot of the generated image.
    plt.imshow(wc.recolor(color_func=image_colors))
    plt.axis("off")
    plt.savefig('popularWordsDM.png', bbox_inches = 'tight', dpi = 200)
    words, vidcount = ds.word_count_dailymotion(ds.mean(dm[0]), ds.standard_deviation(dm[0]))
    data1 = []
    data2 = []
    labels = []
    count = 0
    for w in sorted(words, key=words.get, reverse=True):   
        labels.append(w)
        data1.append(1000 * words[w]/vidcount)
        count +=1
        if count == 10:
            break
    words, vidcount = ds.word_count_dailymotion( 0, 0 )
    for w in labels:
        data2.append(1000 * words[w]/vidcount)
    create_dualbargraph(data1, data2, labels, 'wordUseCompDM.png')
    create_bargraph(da.dm_thetas(),('fans','duration','date created', 'y-intercept'), 'barGraphDM.png')

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def create_word_cloud(ballots, chart_directory, image_name, mask_file,
                      stop_words, word_counts=None):
    """
    Generates a word cloud from given ballots.
    """
    if word_counts is None:
        word_counts=[25, 50, 100, 1000]
    text = ''
    for ballot in ballots:
        text = ''.join((text, ballot.feedback,))
    all_stop_words = STOPWORDS
    all_stop_words |= set(stop_words)
    for word_count in word_counts:
        if mask_file:
            color_mask = imread(mask_file)
            image_colors = ImageColorGenerator(color_mask)
            wc = WordCloud(background_color="white", max_words=word_count,
                           mask=color_mask,
                           stopwords=all_stop_words,
                           color_func=image_colors,
                           max_font_size=80, random_state=42)
        else:
            wc = WordCloud(background_color="white", max_words=word_count,
                           stopwords=all_stop_words,
                           max_font_size=80, random_state=42)
        wc.generate(text)
        axis_image = plt.imshow(wc)
        plt.axis("off")
        image_name_with_count = '{0}-{1}.png'.format(image_name, str(word_count))
        logger.info('...creating word cloud {0}'.format(image_name_with_count))
        save_location = os.path.join(chart_directory, image_name_with_count)
        plt.savefig(save_location)
        plt.close()

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def make_word_cloud(text, save_path, background_color='black'):
    # text expected to a string or a list of [(word, count), ...]
    from wordcloud import WordCloud
    import os

    def col_fun(word, *args, **kw):
        return '#333'

    if type(text) == str:
        big_string = text
    else:
        big_string = ''
        for word in text:
            big_string = big_string + ''.join((word[0]+' ') * word[1])

    # print 'trying to make cloud: %s' % save_path
    # print os.getcwd()
    wc = WordCloud(background_color=background_color,
                   color_func=col_fun,
                   max_words=10000,
                   height=200,
                   width=700,
                   font_path='app/static/fonts/NanumScript.ttc').generate(big_string)
    wc.generate(big_string)
    wc.to_file('app/%s' % save_path)

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def word_cloud(posts):
    text = u' '.join(post['content'] for post in posts)
    
    '''
    if os.path.isdir('/home/public/stanford-corenlp-full-2015-04-20/'):
        proc = CoreNLP("pos", corenlp_jars=["/home/public/stanford-corenlp-full-2015-04-20/*"])
        sentenses = proc.parse_doc(text)['sentences']
        
        text = ''

        for sentence in sentenses:
            text += u' '.join(sentence['lemmas']) + u' '
    '''

    wordcloud = WordCloud(background_color="white", width=1200, height=900, margin=0)
    wordcloud.generate(text)
    fig = plt.gcf()
    # fig.set_size_inches(15, 8.5)
    # Open a plot of the generated image.
    plt.imshow(wordcloud)
    plt.axis("off")

    imgdata = StringIO.StringIO()
    fig.savefig(imgdata, format='png', bbox_inches='tight')
    imgdata.seek(0)  # rewind the data
    plt.close()
    uri = urllib.quote(base64.b64encode(imgdata.buf))
    return uri

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def main(save_files = False, db_filename = '../output/database.sqlite'):
    conn = sqlite3.connect(db_filename)
    c = conn.cursor()

    # Retrieve papers
    c.execute('''SELECT *
                 FROM Papers''')

    paper_content = c.fetchall()
    conn.close()

    titles = ''

    for pc in paper_content:
        titles += pc[1]

    # A Marvin Minsky mask
    mask = np.array(Image.open("../files/minsky_mask.png"))

    wc = WordCloud(background_color="white", max_words=2000, mask=mask, stopwords=STOPWORDS.copy())
    # Generate word cloud
    wc.generate(titles)
    
    if (save_files):
        # Store to file
        wc.to_file("../files/title_cloud.png")
    
    # Show word cloud
    plt.imshow(wc)
    plt.axis("off")
    # Show mask
#    plt.figure()
#    plt.imshow(mask, cmap=plt.cm.gray)
#    plt.axis("off")
    plt.show()

# 需要导入模块: from wordcloud import WordCloud [as 别名]
# 或者: from wordcloud.WordCloud import generate [as 别名]
def get_word_cloud(content, file_name,
                   dict = BASE_DICT,
                   folder_path=BASE_PATH+os.sep+"WordCloud",
                   font_path=BASE_PATH+os.sep+"WordCloud"+os.sep+"幼圆.ttf",
                   width=400,
                   height=200,
                   margin=5,
                   ranks_only=False,
                   prefer_horizontal=0.9,
                   mask=None,
                   scale=1,
                   color_func=None,
                   max_words=200,
                   stopwords=None,
                   random_state=None,
                   background_color='white',
                   max_font_size=None):
    wc = WordCloud(font_path=font_path,width=width,height=height,margin=margin,ranks_only=ranks_only,
                   prefer_horizontal=prefer_horizontal,mask=mask,scale=scale,max_words=max_words,
                   stopwords=stopwords,random_state=random_state,background_color=background_color,max_font_size=max_font_size#,color_func=color_func
                )
    # if dict:
    #     jieba.load_userdict(dict)
    # after = ' '.join(jieba.cut(content, cut_all=False))

    file_with_path = "{BASE_PATH}{sep}{file}".format( BASE_PATH=folder_path ,sep=os.sep,file=file_name)
    wc.generate(content)
    print(file_with_path)
    wc.to_file(file_with_path)
    return file_name