def bubble_data():
pytrends = TrendReq(hl='en-US', tz=360)
initial_keywords = ['obesity', 'diet', 'exercise', 'disease', 'cancer']
keywords = {}
pytrends.build_payload(kw_list=initial_keywords)
print('Fecthing Related Terms')
related = pytrends.related_queries()
related_data = [related[x]['top'][:5].to_numpy() for x in related]
related_terms = [x[:, 0] for x in related_data]
print('Related Terms obtained')
for x in range(len(initial_keywords)):
keywords[initial_keywords[x]] = related_terms[x].tolist()
print('Getting related Terms')
for x in keywords.copy():
subkeywords = keywords[x]
pytrends.build_payload(kw_list=subkeywords)
related = pytrends.related_queries()
related_data = [related[x]['top'][:3].to_numpy() for x in related]
related_terms = [x[:, 0] for x in related_data]
for i in range(len(subkeywords)):
keywords[x].extend(related_terms[i].tolist())
print('Related terms obtained')
data = {}
for keyword in keywords:
subkeywords = keywords[keyword]
data[keyword] = {}
chunkedList = list()
for index in range(5, len(subkeywords) + 1, 5):
chunkedList.append(subkeywords[index - 5:index])
for currList in chunkedList:
if len(currList) > 0:
currList = list(dict.fromkeys(currList))
print(currList)
pytrends.build_payload(kw_list=currList)
interest = pytrends.interest_over_time()
for subkey in currList:
interest_data = {
interest.index[x].strftime("%m/%d/%Y"):
int(interest[subkey][x])
for x in range(len(interest))
}
data[keyword][subkey] = interest_data
data[keyword][subkey]['start_date'] = interest.index[
0].strftime("%m/%d/%Y")
data[keyword][subkey]['end_date'] = interest.index[
len(interest) - 1].strftime("%m/%d/%Y")
def search_generate(seed_keyword, top=True):
pytrends = TrendReq()
pytrends.build_payload([seed_keyword])
if top:
return pytrends.related_queries()[seed_keyword]['top']['query'].tolist(
)
else:
return pytrends.related_queries(
)[seed_keyword]['rising']['query'].tolist()
class GoogleTrendAPI():
def __init__(self):
self.pytrend = TrendReq()
self.result = ''
self.none_type_dic = ''
def get_interest_over_time(self, cel):
time.sleep(1)
self.pytrend.build_payload(kw_list=[cel.celebCode],
timeframe=cel.totalPeriod(),
geo=cel.convertAlpha())
time.sleep(2)
try:
time_result = self.pytrend.interest_over_time()[cel.celebCode]
except:
time_result = pd.DataFrame({cel.celebCode: [cel.celebName]})
pass
return time_result
def get_related_queries(self, cel):
self.pytrend.build_payload(kw_list=[cel.celebCode],
timeframe=cel.totalPeriod(),
geo=cel.convertAlpha())
time.sleep(1)
if self.pytrend.related_queries()[cel.celebCode]['rising'] is None:
self.none_type_dic = pd.DataFrame({'query': [cel.celebName]})
self.result = self.none_type_dic
else:
query_len = len(
self.pytrend.related_queries()[cel.celebCode]['rising'])
if query_len >= 3:
self.result = self.pytrend.related_queries()[
cel.celebCode]['rising'].head(3)
else:
self.result = self.pytrend.related_queries()[
cel.celebCode]['rising'].head(query_len)
return self.result
def add_slash_query(self, related_query):
query_array = []
for i in tqdm(range(len(related_query))):
sub_query = []
for idx, row in related_query[i].iterrows():
sub_query.append(row['query'])
query_array.append('/'.join(sub_query))
return query_array
def post_message(keyword):
pytrends = TrendReq(hl='ja-JP', tz=540)
pytrends.build_payload([keyword], timeframe='now 4-H', geo='JP')
trends = list(pytrends.related_queries().values())[0]
top_queries = trends.get('top')
rising_queries = trends.get('rising')
text = u''
if rising_queries is not None and not rising_queries.empty:
text += u'[%s] Rising queries\n' % keyword
for index, top_query in rising_queries.iterrows():
text += u'%2s. %3s%% \u2191 %s\n' % (index + 1, top_query.value,
top_query.query)
text += u'\n\n'
if top_queries is not None and not top_queries.empty:
text += u'[%s] Top queries\n' % keyword
for index, top_query in top_queries.iterrows():
text += u'%2s. %3s%% %s\n' % (index + 1, top_query.value,
top_query.query)
slack_data = {
'text': text,
'channel': CHANNEL,
'username': USERNAME
}
requests.post(WEBHOOK_URL,
data=json.dumps(slack_data),
headers={'Content-Type': 'application/json'})
def parse(self, response):
try:
pytrends = TrendReq(hl='en-US', tz=360)
pytrends.build_payload(self.search_term.split('|||'),
cat=self.cat,
timeframe='{} {}'.format(
self.start_date, self.end_date),
geo=self.geo,
gprop=self.gprop)
df = pytrends.related_queries()
keys = ['rising', 'top']
for key in keys:
json_data = json.loads(
df[self.search_term][key].to_json(orient="table"))
for item in json_data['data']:
for search_term in self.search_term.split('|||'):
yield RelatedQueriesItem(
search_term=search_term,
search_phrase=item['query'],
top=item['value'] if key == 'top' else '',
breakout=item['value'] if key == 'rising' else '')
except Exception as e:
print(e)
pass
def get_gtrend(code):
all_edinetcodeinfo = EdinetCodeInfo.query.filter(EdinetCodeInfo.securitiescode==code).all()
for edinetcodeinfo in all_edinetcodeinfo:
gtrend = edinetcodeinfo.submitter.replace('株式会社','').replace('株','')
pytrends = TrendReq(hl='ja-JP', tz=360)
keyword=''.join(gtrend.split())
kw_list = [keyword]
pytrends.build_payload(kw_list, timeframe='today 1-m', geo='JP')
df = pytrends.interest_over_time()
dfi = df.drop("isPartial",axis=1)
dt=[]
for d in dfi.index.values:
dt.append(str(d)[0:10])
dfi['date'] = dt
dfi.columns = ['name','date']
trends = pytrends.related_queries()
if(trends[keyword]['top'] is None):
top = ['No data']
else:
top = trends[keyword]['top'].values.tolist()
if(trends[keyword]['rising'] is None):
rising = ['No data']
else:
rising = trends[keyword]['rising'].values.tolist()
return render_template("gtrend.html",code=code,all_edinetcodeinfo=all_edinetcodeinfo, gtrend_g=dfi,gtrend_t=top,gtrend_r=rising)
class GtrendsDownloaderMiddleware:
"""
"""
def __init__(self, proxies=None):
self._proxies = proxies
if self._proxies:
proxie = choice(self._proxies)
pytrends = TrendReq(hl='en-US', tz=360, proxies = {"https": proxie})
else:
self._trends = TrendReq(hl='en-US', tz=360)
@classmethod
def from_crawler(cls, crawler):
proxies = crawler.settings.get("PROXIES")
return cls(proxies)
def process_request(self, request, spider):
self._trends.build_payload(kw_list=request.keywords)
related_queries = self._trends.related_queries()
return GtrendsResponse(related_queries=related_queries)
def process_response(self, request, response, spider):
return response
async def google_trends(request):
if request.method == 'GET':
topic = request.args.get('topic').split(',')
lang = request.args.get('lang')
region = request.args.get('reg')
else:
return json({'response': False, 'message': 'Wrong HTTP method', 'results': []})
if topic is None or topic is '':
return json({'response': False, 'message': 'Not enough arguments', 'results': []})
from pytrends.request import TrendReq
pytrends = TrendReq(hl='{}-{}'.format(lang, region), tz=360)
pytrends.build_payload(topic)
res = pytrends.interest_over_time()
res.drop('isPartial', axis=1, inplace=True)
rel_topics = pytrends.related_topics()[topic[0]]
rel_topics.drop('mid', axis=1, inplace=True)
rel_queries = pytrends.related_queries()[topic[0]]
countries = pytrends.interest_by_region(resolution='COUNTRY')
countries = countries[(countries.T != 0).any()]
from json import loads
countries = loads(countries.to_json())[topic[0]]
return json({'message':'done', 'response': True, 'result': {'interest': res.to_json(), 'related_topics': rel_topics.to_json(),
'top_queries': rel_queries['top'].to_json(), 'rising_queries': rel_queries['rising'].to_json(), 'countries': countries}})
def queries(l_args, s_ticker):
parser = argparse.ArgumentParser(
prog='queries',
description=
"""Print top related queries with this stock's query. [Source: Google]"""
)
parser.add_argument('-n',
"--num",
action="store",
dest="n_num",
type=check_positive,
default=10,
help='number of top related queries to print.')
(ns_parser, l_unknown_args) = parser.parse_known_args(l_args)
if l_unknown_args:
print(
f"The following args couldn't be interpreted: {l_unknown_args}\n")
return
pytrend = TrendReq()
pytrend.build_payload(kw_list=[s_ticker])
df_related_queries = pytrend.related_queries()
df_related_queries = df_related_queries[s_ticker]['top'].head(
ns_parser.n_num)
df_related_queries['value'] = df_related_queries['value'].apply(
lambda x: str(x) + '%')
print(f"Top {s_ticker}'s related queries")
print(df_related_queries.to_string(index=False))
print("")
def get_related_queries(keyword, location, category):
path = ""
pytrend = TrendReq()
kw_list = [keyword]
# print "## get_related_queries ##"
print kw_list
print location
# category = 184
print category
pytrend.build_payload(kw_list, geo=location, cat=category)
# print "edo1"
try:
related_queries_dict = pytrend.related_queries()
# print "edo2"
rising_df = related_queries_dict[keyword]['rising']
# pprint.pprint(rising_df)
# print "edo3"
# print 'printing dictionary top'
top_df = related_queries_dict[keyword]['top']
related_queries = {
'rising': rising_df.to_dict(orient='record'),
'top': top_df.to_dict(orient='record')
}
except:
related_queries = {'rising': "", 'top': ""}
return related_queries
def get_related_queries(keyword, category):
path = ""
pytrend = TrendReq()
kw_list = [keyword]
pytrend.build_payload(kw_list, cat=category)
# pytrend.build_payload(kw_list=['obama', 'trump'])
related_queries_dict = pytrend.related_queries()
# print 'printing dictionary '
# pprint.pprint(related_queries_dict)
# print 'printing dictionary rising'
rising_df = related_queries_dict[keyword]['rising']
# pprint.pprint(rising_df)
# print 'printing dictionary top'
top_df = related_queries_dict[keyword]['top']
# pprint.pprint(top_df)
# print 'interest_over_time'
related_queries = {
'rising': rising_df.to_dict(orient='record'),
'top': top_df.to_dict(orient='record')
}
# pprint.pprint(interest_over_time)
# related_queries_json = json.dumps(related_queries_dict)
return related_queries
def get_data(Start, End):
Trending_Terms = TrendReq(hl='en-US', tz=360)
Keywords = ['share price', 'stock price', 'price']
Trending_Terms.build_payload(kw_list=Keywords,
cat=0,
timeframe=f'{Start} {End}',
geo='US',
gprop='')
Interest_Over_Time = Trending_Terms.interest_over_time()
Related_Queries = Trending_Terms.related_queries()
top_queries = []
rising_queries = []
for key, value in Related_Queries.items():
for k1, v1 in value.items():
if (k1 == "top"):
top_queries.append(v1)
elif (k1 == "rising"):
rising_queries.append(v1)
top_searched = pd.DataFrame(top_queries[1])
rising_searched = pd.DataFrame(rising_queries[1])
return top_searched, rising_searched
def rise(l_args, s_ticker):
parser = argparse.ArgumentParser(
add_help=False,
prog="rise",
description="""Print top rising related queries with this stock's query.
[Source: Google]""",
)
parser.add_argument(
"-n",
"--num",
action="store",
dest="n_num",
type=check_positive,
default=10,
help="number of top rising related queries to print.",
)
try:
ns_parser = parse_known_args_and_warn(parser, l_args)
if not ns_parser:
return
pytrend = TrendReq()
pytrend.build_payload(kw_list=[s_ticker])
df_related_queries = pytrend.related_queries()
df_related_queries = df_related_queries[s_ticker]["rising"].head(
ns_parser.n_num
)
print(f"Top rising {s_ticker}'s related queries")
print(df_related_queries.to_string(index=False))
print("")
except Exception as e:
print(e)
print("")
def queries(l_args, s_ticker):
parser = argparse.ArgumentParser(
prog="queries",
description=
"""Print top related queries with this stock's query. [Source: Google]""",
)
parser.add_argument(
"-n",
"--num",
action="store",
dest="n_num",
type=check_positive,
default=10,
help="number of top related queries to print.",
)
try:
ns_parser = parse_known_args_and_warn(parser, l_args)
pytrend = TrendReq()
pytrend.build_payload(kw_list=[s_ticker])
df_related_queries = pytrend.related_queries()
df_related_queries = df_related_queries[s_ticker]["top"].head(
ns_parser.n_num)
df_related_queries["value"] = df_related_queries["value"].apply(
lambda x: str(x) + "%")
print(f"Top {s_ticker}'s related queries")
print(df_related_queries.to_string(index=False))
print("")
except Exception as e:
print(e)
print("")
def get_response(
pytrends_session: TrendReq, keyword_list: List[str], cat: int = 0, geo: str = ""
) -> dict:
"""Returns a dictionary with a dataframe for each keyword
Calls pytrend's related_queries()
Args:
pytrends_session (object): TrendReq() session of pytrend
keyword_list (list): Used as input for query and passed to TrendReq().build_payload()
cat (int): see https://github.com/pat310/google-trends-api/wiki/Google-Trends-Categories
geo (str): Geolocation like US, UK
Returns:
Dictionary: Dict with dataframes with related query results
"""
assert isinstance(
keyword_list, list
), f"keyword_list should be string. Instead of type {type(keyword_list)}"
pytrends_session.build_payload(keyword_list, cat=cat, geo=geo)
response = pytrends_session.related_queries()
if response is not None:
logger.info(f"Query succeeded for {*keyword_list ,}")
else:
logger.warning(
f"Query failed for {*keyword_list ,}, obtained df_related_queries is None."
)
return response
def get_gtrend(keyword, geo='', timeframe='now 7-d'):
''' Gets gGoogle trend data using the Pytrends module
This functions returns a df with the top related queries related to keyword.
It also loops over these queries and gets the relative search volume over the last week
for each search term. This is stored in the rising_queries_interest dict. The key over this
dict are the search terms and the values are a df with the relative search volume
'''
pytrends = TrendReq()
pytrends.build_payload([keyword], geo=geo, timeframe=timeframe)
while True:
try:
results = pytrends.related_queries()
rising_queries = results[keyword]['rising']
except:
time.sleep(60)
continue
break
rising_queries_interest = {}
if not rising_queries is None:
for i, row in results[keyword]['rising'].iterrows():
while True:
try:
pytrends.build_payload([row.query], geo=geo, timeframe=timeframe)
rising_queries_interest[row.query] = pytrends.interest_over_time()
except:
time.sleep(60)
continue
break
return(rising_queries, rising_queries_interest)
class Trends:
def __init__(self, keywords):
self.pytrends = TrendReq(hl='uk', tz=360)
self.pytrends.build_payload(keywords)
def interest_over_time(self):
return self.pytrends.interest_over_time()
def interest_by_region(self, resolution='COUNTRY', inc_low_vol=False):
return self.pytrends.interest_by_region(resolution=resolution,
inc_low_vol=inc_low_vol)
def related_topics(self):
return self.pytrends.related_topics()
def related_querries(self):
return self.pytrends.related_queries()
def trending_searches(self, pn='ukraine'):
return self.pytrends.trending_searches(pn=pn)
def top_charts(self, date, hl='uk', tz=360, geo='GLOBAL'):
return self.pytrends.top_charts(date, hl=hl, tz=tz, geo=geo)
def suggestions(self, keyword):
return self.pytrends.suggestions(keyword)
def get_related_queries(sessiontrend=None,
term='',
geo='GB',
timeframe='today 5-y'):
kw_list = [term]
if sessiontrend is None:
pytrends = TrendReq(hl='en-US', tz=0)
pytrends.build_payload(kw_list,
cat=0,
timeframe=timeframe,
geo=geo,
gprop='')
else:
pytrends = sessiontrend
related_queries = pytrends.related_queries()
#print(related_queries)
try:
if 0 < len(related_queries[term]['rising'].index) <= 3:
related_queries = related_queries[term]['rising'].values.tolist()
elif len(related_queries[term]['rising'].index) == 0:
related_queries = []
else:
related_queries = related_queries[term]['rising'].values.tolist()
except:
related_queries = []
# return related_queries
return json.dumps(related_queries)
def retrieve_google_trends(self, search, date_range):
# Set up the trend fetching object
pytrends = TrendReq(hl='en-US', tz=360)
kw_list = [search]
try:
# Create the search object
pytrends.build_payload(kw_list,
cat=0,
timeframe=date_range[0],
geo='',
gprop='news')
# Retrieve the interest over time
trends = pytrends.interest_over_time()
related_queries = pytrends.related_queries()
except Exception as e:
print('\nGoogle Search Trend retrieval failed.')
print(e)
return
return trends, related_queries
def trending(k):
import pandas as pd
from pytrends.request import TrendReq
pytrend = TrendReq()
pytrend.build_payload(kw_list=[k])
related_queries = pytrend.related_queries()
return '\n'.join((list((((related_queries)[k])["top"])["query"].to_frame()["query"])))
def get_info_world(timeframe):
res_df = []
all_words = [['ada','altcoin','altcoins','bitcoin','bitconnect'],['bitminter', 'blockchain', 'blockchain%20mining','btc', 'cardano'],['crypto', 'cryptocurrencies','cryptocurrency', 'cryptographic', 'dash'], ['doge', 'dogecoin', 'eos','ethereum', 'hashing'],['hashocean', 'hodl', 'ico', 'iota', 'litecoin'],['localbitcoins', 'ltc', 'ltc', 'mercado', 'mining%20cryptocurrency'], ['montero', 'onecoin', 'satoshi', 'stellar', 'stratis'], ['tron', 'xlm', 'xmr', 'xrp', 'zcash']]
for k in all_words:
list_query_top = []
list_query_top_value = []
list_query_rising = []
list_query_rising_value = []
pytrend = TrendReq()
pytrend.build_payload(kw_list=k ,timeframe=timeframe)
related_queries_dict = pytrend.related_queries()
for f in k:
s_top = related_queries_dict[f]['top']
s_rising = related_queries_dict[f]['rising']
if s_rising is not None:
df_rising = s_rising.where(s_rising > 500)
df_rising = df_rising.dropna()
list_rising = df_rising['query'].tolist()
list_rising2 = df_rising['value'].tolist()
result_rising = zip(list_rising, list_rising2)
if s_top is not None:
df_top = s_top.where(s_top > 50)
df_top = df_top.dropna()
list_query_top = df_top['query'].tolist()
list_query_top2 = df_top['value'].tolist()
result_top = zip(list_query_top, list_query_top2)
print(list(result_top))
for j in result_top:
url = j[0].replace(" ", "+")
url2 = 'https://trends.'+'google.com/trends/explore?q=' + f + '+' + url + '&date={}'.format(timeframe)
pytrend.build_payload(kw_list=[j[0]],timeframe='now 1-d')
interest_by_region_df = pytrend.interest_by_region()
s = interest_by_region_df[j[0]]
df = s.where(s > 50)
df = df.dropna()
df = df.sort_values(ascending=False)
interest_by_region = pd.Series(df.index.values.tolist())
weight = pd.Series(list(df.values))
res_df.append(pd.DataFrame({'Location': pd.Series('Wordwide'), 'Keyword': pd.Series(f), 'Related Queries': pd.Series(j[0]), 'Percentage': pd.Series(str(j[1] * 10)), 'Interest by Region':interest_by_region, 'Weight': weight, 'Google Search Link': pd.Series('<a href="{}">{}</a>'.format(url2, url2))}))
for j in result_rising:
url = j[0].replace(" ", "+")
url2 = 'https://trends.'+'google.com/trends/explore?q=' + f + '+' + url + '&date={}'.format(timeframe)
pytrend.build_payload(kw_list=[j[0]],timeframe=timeframe)
interest_by_region_df = pytrend.interest_by_region()
s = interest_by_region_df[j[0]]
df = s.where(s > 50)
df = df.dropna()
df = df.sort_values(ascending=False)
interest_by_region = pd.Series(df.index.values.tolist())
weight = pd.Series(list(df.values))
res_df.append(pd.DataFrame({'Location': pd.Series('Wordwide'), 'Keyword': pd.Series(f), 'Related Queries': pd.Series(j[0]), 'Percentage': pd.Series('Breakout'), 'Interest by Region':interest_by_region, 'Weight': weight, 'Google Search Link': pd.Series('<a href="{}">{}</a>'.format(url2, url2))}))
df = pd.concat(res_df)
return df
def related_keyword(string):
try:
pytrend_session = TrendReq()
pytrend_session.build_payload(kw_list=[string])
related_keyword = pytrend_session.related_queries()
except KeyError as e: # 키워드가 입력되지 않은 경우
related_keyword[string]['top'] = None
related_keyword[string]['rising'] = None
return related_keyword[string]
return related_keyword[string]
class Trends:
'''
Wrapper class for pytrends library.
'''
def __init__(self, keywords):
'''
Initialize trends object.
'''
self.pytrends = TrendReq(hl='uk', tz=360)
self.pytrends.build_payload(keywords)
def interest_over_time(self):
'''
Get interest overt time.
'''
return self.pytrends.interest_over_time()
def interest_by_region(self, resolution='COUNTRY', inc_low_vol=False):
'''
Get interest by region.
'''
return self.pytrends.interest_by_region(resolution=resolution, inc_low_vol=inc_low_vol)
def related_topics(self):
'''
Get related topics.
'''
return self.pytrends.related_topics()
def related_querries(self):
'''
Get related search querries.
'''
return self.pytrends.related_queries()
def trending_searches(self, pn='ukraine'):
'''
Get trending searhes by country.
'''
return self.pytrends.trending_searches(pn=pn)
def top_charts(self, date, hl='uk', tz=360, geo='GLOBAL'):
'''
Get top charts by date.
'''
return self.pytrends.top_charts(date, hl=hl, tz=tz, geo=geo)
def suggestions(self, keyword):
'''
Get suggestions for keyword.
'''
return self.pytrends.suggestions(keyword)
class Gtrend:
def __init__(self, brand):
kw_list = [brand]
self.brand = brand
self.pytrends = TrendReq()
self.pytrends.build_payload(kw_list, cat=0, timeframe='today 12-m', geo='', gprop='froogle')
def trend(self):
return self.pytrends.interest_over_time().drop(columns=['isPartial'])
def queries(self):
return self.pytrends.related_queries()[self.brand]
def getSearchQueries(searchWord):
pytrends = TrendReq(hl='en-US', tz=360)
kw_list = [searchWord]
pytrends.build_payload(kw_list,
cat=0,
timeframe='today 5-y',
geo='',
gprop='')
df = pytrends.related_queries()
et = pytrends.suggestions(searchWord)
print(et)
print(df)
def get_related_queries():
string = request.args.get('keyword')
try:
pytrend_session = TrendReq(hl='en-US', tz=120)
pytrend_session.build_payload(kw_list=[string], cat=30, timeframe='today 12-m', geo='', gprop='')
related_keyword = pytrend_session.related_queries()
if related_keyword[string]['top'] is None:
raise KeyError
except KeyError as e: # 키워드가 입력되지 않은 경우
return ''
return render_template('search/delayed/related_keywords.html', related_keyword=related_keyword[string]['top']['query'])
def get_trending_searches():
list = []
pytrends = TrendReq(hl='en-US', tz=360)
pytrends.build_payload(kw_list=['coronavirus'])
related_queries = pytrends.related_queries()
for i in related_queries.values():
df = i['top']
for index, row in df.iterrows():
# print(row['query'])
list.append(row['query'])
return list
def related_table_data():
final_mapping = {}
most_recent_nonfail = {}
start = datetime.datetime.strptime("19-04-2015", "%d-%m-%Y")
end = datetime.datetime.strptime("12-04-2020", "%d-%m-%Y")
dates = list(rrule(
WEEKLY,
dtstart=start,
until=end,
))
pytrends = TrendReq(hl='en-US', tz=360)
initial_keywords = ['obesity', 'diet', 'exercise', 'disease', 'cancer']
with open('final_related_data.json') as json_file:
init_related = json.load(json_file)
for item in initial_keywords:
final_mapping[item] = {}
most_recent_nonfail[item] = list(init_related[item].keys())
count = 0
for date in dates:
try:
print('Starting: ' + date.strftime("%Y-%m-%d"))
date_key = date.strftime("%m/%d/%Y")
next_day = date + timedelta(days=7)
dates_str = date.strftime("%Y-%m-%d") + " " + next_day.strftime(
"%Y-%m-%d")
pytrends.build_payload(kw_list=initial_keywords,
timeframe=dates_str)
related = pytrends.related_queries()
for key in related:
if type(related[key]['top']) == type(None):
print("FAILED " + str(key))
final_mapping[key][date_key] = most_recent_nonfail[key]
else:
val = related[key]['top'][:5].to_numpy()
val = val[:, 0].tolist()
final_mapping[key][date_key] = val
most_recent_nonfail[key] = val
print('Complete: ' + date.strftime("%Y-%m-%d"))
except:
date_key = date.strftime("%m/%d/%Y")
print("*" * 40 + 'FAILED AT: ' + date_key + "*" * 40)
for key in initial_keywords:
final_mapping[key][date_key] = most_recent_nonfail[key]
print('Complete: ' + date.strftime("%Y-%m-%d"))
print('Done')
def getRelatedQueries(keywords, interval, country_id):
""" Returns top queries for given inputs
Parameters:
keywords: list of keywords
interval: interval with format %Y-%m-%d
country_id: country ID
"""
pytrend = TrendReq()
pytrend.build_payload(kw_list=keywords, timeframe=interval, geo=country_id)
queries = pytrend.related_queries()
queries2 = pd.DataFrame()
for key in queries.keys():
queries2 = pd.concat([queries2, queries[key]['top']])
return list(queries2['query'].unique())
def get_queries(ticker: str) -> pd.DataFrame:
"""Get related queries from google api [Source: google]
Parameters
----------
ticker : str
Stock ticker to compare
Returns
-------
pd.DataFrame
Dataframe of related queries
"""
pytrend = TrendReq()
pytrend.build_payload(kw_list=[ticker])
return pytrend.related_queries()
# Login to Google. Only need to run this once, the rest of requests will use the same session. pytrend = TrendReq() # Create payload and capture API tokens. Only needed for interest_over_time(), interest_by_region() & related_queries() pytrend.build_payload(kw_list=['pizza', 'bagel']) # Interest Over Time interest_over_time_df = pytrend.interest_over_time() print(interest_over_time_df.head()) # Interest by Region interest_by_region_df = pytrend.interest_by_region() print(interest_by_region_df.head()) # Related Queries, returns a dictionary of dataframes related_queries_dict = pytrend.related_queries() print(related_queries_dict) # Get Google Hot Trends data trending_searches_df = pytrend.trending_searches() print(trending_searches_df.head()) # Get Google Top Charts top_charts_df = pytrend.top_charts(cid='actors', date=201611) print(top_charts_df.head()) # Get Google Keyword Suggestions suggestions_dict = pytrend.suggestions(keyword='pizza') print(suggestions_dict)
def test_related_queries(self):
pytrend = TrendReq()
pytrend.build_payload(kw_list=['pizza', 'bagel'])
self.assertIsNotNone(pytrend.related_queries())
