def test_summarize():
p = pd.DataFrame({
'price_mean':[diamonds.price.mean()],
'price_std':[diamonds.price.std()]
})
assert p.equals(diamonds >> summarize(price_mean=X.price.mean(),
price_std=X.price.std()))
pcut = pd.DataFrame({
'cut':['Fair','Good','Ideal','Premium','Very Good']
})
pcut['price_mean'] = [diamonds[diamonds.cut == c].price.mean() for c in pcut.cut.values]
pcut['price_std'] = [diamonds[diamonds.cut == c].price.std() for c in pcut.cut.values]
assert pcut.equals(diamonds >> groupby('cut') >>
summarize(price_mean=X.price.mean(), price_std=X.price.std()))
def test_load_country_continent():
ccdf = utils.load_country_continent()
logging.debug(f"{len(ccdf)} countries parsed")
cont = ccdf >> group_by(X.Continent_Code, X.Continent_Name) >> summarize(num_countries=n(X.Country_Number))
logging.debug(cont)
assert cont.shape[0] == 6
def test_load_countries():
cdf = utils.load_countries()
cont = cdf >> group_by(X.continent_code) >> summarize(num_countries=n(X.country_code))
logging.debug(cont)
assert {'country_code', 'continent_code', 'country_name'}.issubset(set(cdf.columns))
# ***********************************************
fileInfo1 = glob.glob('{}/{}'.format(globalVar['inpPath'],
'LSH0167_dataL2.csv'))
dataL2 = pd.read_csv(fileInfo1[0], na_filter=False)
# breakpoint()
# ***********************************************
# 데이터 요약 (요약통계량)
# ***********************************************
# 연소득당 거래금액 따른 기초 통계량
dataL2.describe()
# 법정동에 따른 연소득당 거래금액 따른 기초 통계량
dataL3 = ((dataL2 >> dfply.group_by(dfply.X.d2) >>
dfply.summarize(meanVal=dfply.mean(dfply.X.val))))
# *******************************************************
# 데이터 요약 (표/그래프 활용)
# *******************************************************
# 연소득당 거래금액 따른 히스토그램
saveImg = '{}/{}_{}.png'.format(globalVar['figPath'], serviceName,
'연소득당 거래금액 따른 히스토그램')
sns.distplot(dataL2['val'], kde=True, rug=False)
plt.show()
plt.savefig(saveImg, dpi=600, bbox_inches='tight')
# 법정동에 따른 연소득당 거래금액 히스토그램
saveImg = '{}/{}_{}.png'.format(globalVar['figPath'], serviceName,
'법정동에 따른 연소득당 거래금액 히스토그램')
# Solution 2 (as series of list)
emails.str.findall(pattern, flags=re.IGNORECASE)
# Solution 3 (as list)
[x[0] for x in [re.findall(pattern, email) for email in emails] if len(x) > 0]
#%% Mean of a series grouped by another series
fruit = pd.Series(np.random.choice(['apple', 'banana', 'carrot'], 10))
weights = pd.Series(np.linspace(1, 10, 10))
print(weights.tolist())
print(fruit.tolist())
import dfply as dplyr
df=pd.DataFrame({'weights':weights, 'fruits':fruit})
df >> dplyr.group_by(X.fruits) >> dplyr.summarize(mean=X.weights.mean())
weights.groupby(fruit).mean()
#%% Local maxima/minima in a series
ser = pd.Series([2, 10, 3, 4, 9, 10, 2, 7, 3])
# Solution
dd = np.diff(np.sign(np.diff(ser)))
peak_locs = np.where(dd == -2)[0] + 1
#%% Fill Nas
df = pd.DataFrame({"A":[None, 1, 2, 3, None, None],
"B":[11, 5, None, None, None, 8],
"C":[None, 5, 10, 11, None, 8]})
df.bfill(axis ='rows')
def exec(self):
try:
log.info('[START] {}'.format("exec"))
# breakpoint()
# python -c "import nltk; nltk.download('punkt')"
# nltk.download('stopwords')
# 1) https://edition.cnn.com/2020/06/02/world/nodosaur-fossil-stomach-contents-scn-trnd/index.html에서 기사 내용을 스크랩하십시오.
html = urlopen(
"https://edition.cnn.com/2020/06/02/world/nodosaur-fossil-stomach-contents-scn-trnd/index.html"
)
# html = requests.get(url)
soup = BeautifulSoup(html, 'html.parser')
section = soup.select('section.zn-body-text')
liGetText = []
for i in section:
getText = i.get_text()
log.info("getText : {%s} : {%s}", len(getText), getText)
# 단어 추출
wordTokens = word_tokenize(getText)
# 불용어
stopWords = set(stopwords.words('english'))
# log.info("wordTokens : {%s} : {%s}", len(wordTokens), wordTokens)
# log.info("stopWords : {%s} : {%s}", len(stopWords), stopWords)
# 2) 기사 내용을 사전 처리하여 불용어없이 단수 명사 목록을 얻습니다.
for j in wordTokens:
if j not in stopWords:
liGetText.append(j)
log.info("liGetText : {%s} : {%s}", len(liGetText), liGetText)
data = pd.DataFrame({'type': liGetText})
# 3) 빈도분포 및 워드 클라우드 시각화
dataL1 = ((data >> filter_by(
X.type != '.', X.type != ',', X.type != "'", X.type != "''",
X.type != "``", X.type != "'s") >> group_by(X.type) >>
summarize(number=n(X.type)) >> ungroup() >> arrange(
X.number, ascending=False)))
log.info("dataL1 : {%s} : {%s}", len(dataL1), dataL1)
# 데이터 시각화를 위한 전처리
objData = {}
for i in dataL1.values:
key = i[0]
val = i[1]
objData[key] = val
log.info("objData : {%s} : {%s}", len(objData), objData)
wordcloud = WordCloud(
width=1000, height=1000,
background_color="white").generate_from_frequencies(objData)
saveImg = '{}/{}_{}'.format(globalVar['figPath'], serviceName,
'워드 클라우드.png')
log.info('[CHECK] saveFile : {}'.format(saveImg))
plt.imshow(wordcloud, interpolation="bilinear")
plt.axis("off")
plt.savefig(saveImg, dpi=600, bbox_inches='tight')
plt.show()
except Exception as e:
log.error("Exception : {}".format(e))
raise e
finally:
log.info('[END] {}'.format("exec"))