class CReivew:
def __init__(self, dbinfo=ct.DB_INFO, redis_host=None):
self.dbinfo = dbinfo
self.logger = getLogger(__name__)
self.tu_client = get_tushare_client()
self.doc = CDoc()
self.redis = create_redis_obj(
) if redis_host is None else create_redis_obj(redis_host)
self.mysql_client = CMySQL(dbinfo, iredis=self.redis)
self.margin_client = Margin(dbinfo=dbinfo, redis_host=redis_host)
self.rstock_client = RIndexStock(dbinfo=dbinfo, redis_host=redis_host)
self.sh_market_client = StockExchange(ct.SH_MARKET_SYMBOL)
self.sz_market_client = StockExchange(ct.SZ_MARKET_SYMBOL)
self.emotion_client = Emotion()
def get_industry_data(self, cdate):
ri = RIndexIndustryInfo()
df = ri.get_k_data(cdate)
if df.empty: return df
df = df.reset_index(drop=True)
df = df.sort_values(by='amount', ascending=False)
df['money_change'] = (df['amount'] - df['preamount']) / 1e8
industry_info = IndustryInfo.get()
df = pd.merge(df, industry_info, how='left', on=['code'])
return df
def get_index_data(self, cdate):
df = pd.DataFrame()
for code, name in ct.TDX_INDEX_DICT.items():
data = CIndex(code).get_k_data(cdate)
data['name'] = name
data['code'] = code
df = df.append(data)
df = df.reset_index(drop=True)
return df
def get_market_data(self, market, start_date, end_date):
if market == ct.SH_MARKET_SYMBOL:
df = self.sh_market_client.get_k_data_in_range(
start_date, end_date)
df = df.loc[df.name == '上海市场']
else:
df = self.sz_market_client.get_k_data_in_range(
start_date, end_date)
df = df.loc[df.name == '深圳市场']
df = df.round(2)
df = df.drop_duplicates()
df = df.reset_index(drop=True)
df = df.sort_values(by='date', ascending=True)
df.negotiable_value = (df.negotiable_value / 2).astype(int)
return df
def get_rzrq_info(self, market, start_date, end_date):
df = self.margin_client.get_k_data_in_range(start_date, end_date)
if market == ct.SH_MARKET_SYMBOL:
df = df.loc[df.code == 'SSE']
df['code'] = '上海市场'
else:
df = df.loc[df.code == 'SZSE']
df['code'] = '深圳市场'
df = df.round(2)
df['rzye'] = df['rzye'] / 1e+8
df['rzmre'] = df['rzmre'] / 1e+8
df['rzche'] = df['rzche'] / 1e+8
df['rqye'] = df['rqye'] / 1e+8
df['rzrqye'] = df['rzrqye'] / 1e+8
df = df.drop_duplicates()
df = df.reset_index(drop=True)
df = df.sort_values(by='date', ascending=True)
return df
def get_index_df(self, code, start_date, end_date):
cindex_client = CIndex(code)
df = cindex_client.get_k_data_in_range(start_date, end_date)
df['time'] = df.index.tolist()
df = df[[
'time', 'open', 'high', 'low', 'close', 'volume', 'amount', 'date'
]]
return df
def update(self, cdate=datetime.now().strftime('%Y-%m-%d')):
start_date = get_day_nday_ago(cdate, 100, dformat="%Y-%m-%d")
end_date = cdate
try:
#market info
sh_df = self.get_market_data(ct.SH_MARKET_SYMBOL, start_date,
end_date)
sz_df = self.get_market_data(ct.SZ_MARKET_SYMBOL, start_date,
end_date)
date_list = list(
set(sh_df.date.tolist()).intersection(set(
sz_df.date.tolist())))
sh_df = sh_df[sh_df.date.isin(date_list)]
sh_df = sh_df.reset_index(drop=True)
sz_df = sz_df[sz_df.date.isin(date_list)]
sz_df = sz_df.reset_index(drop=True)
#rzrq info
sh_rzrq_df = self.get_rzrq_info(ct.SH_MARKET_SYMBOL, start_date,
end_date)
sz_rzrq_df = self.get_rzrq_info(ct.SZ_MARKET_SYMBOL, start_date,
end_date)
date_list = list(
set(sh_rzrq_df.date.tolist()).intersection(
set(sz_rzrq_df.date.tolist())))
sh_rzrq_df = sh_rzrq_df[sh_rzrq_df.date.isin(date_list)]
sh_rzrq_df = sh_rzrq_df.reset_index(drop=True)
sz_rzrq_df = sz_rzrq_df[sz_rzrq_df.date.isin(date_list)]
sz_rzrq_df = sz_rzrq_df.reset_index(drop=True)
#average price info
av_df = self.get_index_df('880003', start_date, end_date)
#limit up and down info
limit_info = CLimit(self.dbinfo).get_data(cdate)
stock_info = self.rstock_client.get_data(cdate)
stock_info = stock_info[stock_info.volume >
0] #get volume > 0 stock list
stock_info = stock_info.reset_index(drop=True)
#index info
index_info = self.get_index_data(end_date)
#industry analysis
industry_info = self.get_industry_data(cdate)
#all stock info
all_stock_info = self.rstock_client.get_k_data_in_range(
start_date, end_date)
#gen review file and make dir for new data
self.doc.generate(cdate, sh_df, sz_df, sh_rzrq_df, sz_rzrq_df,
av_df, limit_info, stock_info, industry_info,
index_info, all_stock_info)
##gen review animation
#self.gen_animation()
except Exception as e:
self.logger.error(e)
traceback.print_exc()
def gen_animation(self, sfile=None):
style.use('fivethirtyeight')
Writer = animation.writers['ffmpeg']
writer = Writer(fps=1, metadata=dict(artist='biek'), bitrate=1800)
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
_today = datetime.now().strftime('%Y-%m-%d')
cdata = self.mysql_client.get('select * from %s where date = "%s"' %
(ct.ANIMATION_INFO, _today))
if cdata is None: return None
cdata = cdata.reset_index(drop=True)
ctime_list = cdata.time.unique()
name_list = cdata.name.unique()
ctime_list = [
datetime.strptime(ctime, '%H:%M:%S') for ctime in ctime_list
]
frame_num = len(ctime_list)
if 0 == frame_num: return None
def animate(i):
ax.clear()
ax.xaxis.set_major_formatter(mdates.DateFormatter('%H:%M:%S'))
ax.xaxis.set_major_locator(mdates.DayLocator())
ax.set_title('盯盘', fontproperties=get_chinese_font())
ax.set_xlabel('时间', fontproperties=get_chinese_font())
ax.set_ylabel('增长', fontproperties=get_chinese_font())
ax.set_ylim((-6, 6))
fig.autofmt_xdate()
for name in name_list:
pchange_list = list()
price_list = cdata[cdata.name == name]['price'].tolist()
pchange_list.append(0)
for _index in range(1, len(price_list)):
pchange_list.append(
10 * (price_list[_index] - price_list[_index - 1]) /
price_list[0])
ax.plot(ctime_list[0:i],
pchange_list[0:i],
label=name,
linewidth=1.5)
if pchange_list[i - 1] > 1 or pchange_list[i - 1] < -1:
ax.text(ctime_list[i - 1],
pchange_list[i - 1],
name,
font_properties=get_chinese_font())
ani = animation.FuncAnimation(fig,
animate,
frame_num,
interval=60000,
repeat=False)
sfile = '/data/animation/%s_animation.mp4' % _today if sfile is None else sfile
ani.save(sfile, writer)
plt.close(fig)
class MarauderMap():
def __init__(self):
self.ris = RIndexStock()
self.logger = getLogger(__name__)
def get_data(self, mdate):
return self.ris.get_data(mdate)
def plot(self, cdate, fdir, fname):
df = self.ris.get_data(cdate)
if df.empty: return
fig, ax = plt.subplots()
#get min profit day
min_pday = df.pday.values.min()
max_pday = df.pday.values.max()
#get max profit day
min_profit = df.profit.values.min()
max_profit = df.profit.values.max()
#set axis for map
xmax = max(abs(min_pday), max_pday)
ymax = max(abs(min_profit), max_profit)
ax.set_xlim(-xmax, xmax)
ax.set_ylim(-ymax, ymax)
ax.spines['top'].set_color('none')
ax.spines['right'].set_color('none')
ax.xaxis.set_ticks_position('bottom')
ax.spines['bottom'].set_position(('data', 0))
ax.yaxis.set_ticks_position('left')
ax.spines['left'].set_position(('data', 0))
for code in df.code.tolist():
pday = df.loc[df.code == code, 'pday']
profit = df.loc[df.code == code, 'profit']
ax.scatter(pday, profit, s=5, alpha=1, linewidths=0.1)
plt.savefig('%s/%s.png' % (fdir, fname), dpi=1000)
def gen_animation(self, end_date, days):
import matplotlib.animation as animation
matplotlib.use('Agg')
start_date = get_day_nday_ago(end_date, num=days, dformat="%Y-%m-%d")
df = self.ris.get_k_data_in_range(start_date, end_date)
fig, ax = plt.subplots()
#get min profit day
min_pday = df.pday.values.min()
max_pday = df.pday.values.max()
#get max profit day
min_profit = df.profit.values.min()
max_profit = df.profit.values.max()
#set axis for map
xmax = max(abs(min_pday), max_pday)
ymax = max(abs(min_profit), max_profit)
groups = df.groupby(df.date)
dates = list(set(df.date.tolist()))
dates.sort()
Writer = animation.writers['ffmpeg']
writer = Writer(fps=2, metadata=dict(artist='biek'), bitrate=-1)
def init():
ax.clear()
ax.set_xlim(-xmax, xmax)
ax.set_ylim(-ymax, ymax)
ax.spines['top'].set_color('none')
ax.spines['right'].set_color('none')
ax.xaxis.set_ticks_position('bottom')
ax.spines['bottom'].set_position(('data', 0))
ax.yaxis.set_ticks_position('left')
ax.spines['left'].set_position(('data', 0))
def animate(i):
cdate = dates[i]
df = groups.get_group(cdate)
init()
print(cdate, len(df))
bull_stock_num = len(df[df.profit >= 0])
for code in df.code.tolist():
pday = df.loc[df.code == code, 'pday']
profit = df.loc[df.code == code, 'profit']
ax.scatter(pday, profit, color='black', s=1)
ax.set_title("日期：%s 股票总数：%s 牛熊股比:%s" %
(cdate, len(df), 100 * bull_stock_num / len(df)),
fontproperties=get_chinese_font())
ani = animation.FuncAnimation(fig,
animate,
frames=len(dates),
init_func=init,
interval=1000,
repeat=False)
sfile = '/code/panimation.mp4'
ani.save(sfile, writer)
ax.set_title('Marauder Map for date')
ax.grid(True)
plt.close(fig)
elif code in black_set:
return -1
else:
return 0
#if __name__ == '__main__':
# mval_client = MValuation()
# mval_client.update_index(end_date = '2019-08-13')
if __name__ == '__main__':
try:
mdate = 20190815
mval_client = MValuation()
rindex_client = RIndexStock()
df = rindex_client.get_data(transfer_int_to_date_string(mdate))
df['mv'] = df['totals'] * df['close']
df['mv'] = df['mv'] / 100000000
df = df[df.mv > 100]
df = df[(df.profit > 1) & (df.profit < 6.5)]
#黑名单
black_set = set(ct.BLACK_DICT.keys())
white_set = set(ct.WHITE_DICT.keys())
if len(black_set.intersection(white_set)) > 0:
raise Exception("black and white has intersection.")
df['history'] = df.apply(
lambda row: get_hist_val(black_set, white_set, row['code']),
axis=1)
df = df[df['history'] > -1]
#basic_info
base_df = mval_client.stock_info_client.get()
import traceback
import const as ct
import pandas as pd
from rstock import RIndexStock
from cstock_info import CStockInfo
from industry_info import IndustryInfo
if __name__ == '__main__':
try:
mdate = '2019-08-02'
cobj = CStockInfo()
robj = RIndexStock()
iobj = IndustryInfo()
black_list = list(ct.BLACK_DICT.keys())
bdf = cobj.get()
stock_info = robj.get_data(mdate)
idf = iobj.get_csi_industry_data(mdate)
df = pd.merge(bdf, idf, how='left', on=['code'])
df = pd.merge(stock_info, df, how='inner', on=['code'])
df = df[~df.code.isin(black_list)]
df = df[(df.profit > 1) & (df.profit < 3) & (df.pday > 30) &
(df.timeToMarket < 20150101)]
df = df.reset_index(drop=True)
#df = df[['code', 'name', 'industry', 'profit', 'pday', 'pind_name', 'sind_name', 'tind_name', 'find_name']]
df = df[['code', 'name', 'profit', 'pday', 'find_name']]
for name, contains in df.groupby('find_name'):
if len(contains) > 2:
print("--------------------")
print(name)
print(contains)
print("====================")
class MarauderMap():
def __init__(self, code_list):
self.codes = code_list
self.ris = RIndexStock()
self.logger = getLogger(__name__)
def plot(self, cdate, fdir, fname):
df = self.ris.get_data(cdate)
if df.empty: return
fig, ax = plt.subplots()
#get min profit day
min_pday = df.pday.values.min()
max_pday = df.pday.values.max()
#get max profit day
min_profit = df.profit.values.min()
max_profit = df.profit.values.max()
#set axis for map
xmax = max(abs(min_pday), max_pday)
ymax = max(abs(min_profit), max_profit)
ax.set_xlim(-xmax, xmax)
ax.set_ylim(-ymax, ymax)
ax.spines['top'].set_color('none')
ax.spines['right'].set_color('none')
ax.xaxis.set_ticks_position('bottom')
ax.spines['bottom'].set_position(('data', 0))
ax.yaxis.set_ticks_position('left')
ax.spines['left'].set_position(('data', 0))
for code in self.codes:
pday = df.loc[df.code == code, 'pday']
profit = df.loc[df.code == code, 'profit']
ax.scatter(pday, profit, s=5, alpha=1, linewidths=0.1)
plt.savefig('%s/%s.png' % (fdir, fname), dpi=1000)
def gen_animation(self, cdate):
df = self.get_data(cdate)
fig, ax = plt.subplots()
Writer = animation.writers['ffmpeg']
writer = Writer(fps=30, metadata=dict(artist='biek'), bitrate=1800)
#get min profit day
min_pday = df.pday.values.min()
max_pday = df.pday.values.max()
#get max profit day
min_profit = df.profit.values.min()
max_profit = df.profit.values.max()
#set axis for map
xmax = max(abs(min_pday), max_pday)
ymax = max(abs(min_profit), max_profit)
groups = list(df.groupby(df.time))
def init():
ax.clear()
ax.set_xlim(-xmax, xmax)
ax.set_ylim(-ymax, ymax)
ax.spines['top'].set_color('none')
ax.spines['right'].set_color('none')
ax.xaxis.set_ticks_position('bottom')
ax.spines['bottom'].set_position(('data', 0))
ax.yaxis.set_ticks_position('left')
ax.spines['left'].set_position(('data', 0))
def animate(n):
val = groups[n][1].values.tolist()[0]
for code in self.codes:
pday = val[1]
profit = val[2]
ax.scatter(pday, profit, s=5, alpha=1, linewidths=0.1)
ani = animation.FuncAnimation(fig,
animate,
frames=300,
init_func=init,
interval=1,
repeat=False)
sfile = '/code/animation.mp4'
ani.save(sfile, writer, fps=60, dpi=100)
ax.set_title('Marauder Map for date')
ax.grid(True)
plt.close(fig)
class BullStockRatio:
def __init__(self, index_code, dbinfo=ct.DB_INFO, redis_host=None):
self.dbinfo = dbinfo
self.index_code = index_code
self.index_obj = CIndex(index_code,
dbinfo=self.dbinfo,
redis_host=redis_host)
self.db_name = self.index_obj.get_dbname(index_code)
self.logger = getLogger(__name__)
self.ris = RIndexStock(dbinfo, redis_host)
self.bull_stock_ratio_table = self.get_table_name()
self.redis = create_redis_obj(
) if redis_host is None else create_redis_obj(redis_host)
self.mysql_client = CMySQL(self.dbinfo,
dbname=self.db_name,
iredis=self.redis)
if not self.create(): raise Exception("create emotion table failed")
def delete(self):
self.mysql_client.delete(self.bull_stock_ratio_table)
def get_table_name(self):
return "%s_%s" % (self.db_name, ct.BULLSTOCKRATIO_TABLE)
def create(self):
if self.bull_stock_ratio_table not in self.mysql_client.get_all_tables(
):
sql = 'create table if not exists %s(date varchar(10) not null, ratio float, PRIMARY KEY (date))' % self.bull_stock_ratio_table
if not self.mysql_client.create(sql, self.bull_stock_ratio_table):
return False
return True
def is_date_exists(self, table_name, cdate):
if self.redis.exists(table_name):
return cdate in set(
str(tdate, encoding=ct.UTF8)
for tdate in self.redis.smembers(table_name))
return False
def get_k_data_between(self, start_date, end_date):
sql = "select * from %s where date between \"%s\" and \"%s\"" % (
self.get_table_name(), start_date, end_date)
return self.mysql_client.get(sql)
def get_components(self, cdate):
df = self.index_obj.get_components_data(cdate)
if df is None: return list()
if df.empty: return list()
if self.index_code == '000001': df = df[df.code.str.startswith('6')]
return df.code.tolist()
def get_data(self, cdate):
return self.ris.get_data(cdate)
def update(self, end_date=None, num=30):
if end_date is None: end_date = datetime.now().strftime('%Y-%m-%d')
#start_date = "1997-12-30"
start_date = get_day_nday_ago(end_date, num=num, dformat="%Y-%m-%d")
succeed = True
code_list = self.get_components(end_date)
if 0 == len(code_list):
self.logger.error("%s code_list for %s is empty" %
(end_date, self.index_code))
return False
for mdate in get_dates_array(start_date, end_date):
if CCalendar.is_trading_day(mdate, redis=self.redis):
if not self.set_ratio(code_list, mdate):
self.logger.error("set %s score for %s set failed" %
(self.index_code, mdate))
succeed = False
return succeed
def get_profit_stocks(self, df):
data = df[df.profit >= 0]
return data.code.tolist()
def set_ratio(self,
now_code_list,
cdate=datetime.now().strftime('%Y-%m-%d')):
if self.is_date_exists(self.bull_stock_ratio_table, cdate):
self.logger.debug("existed date:%s, date:%s" %
(self.bull_stock_ratio_table, cdate))
return True
code_list = self.get_components(cdate)
if len(code_list) == 0: code_list = now_code_list
df = self.get_data(cdate)
if df is None: return False
if df.empty: return False
df = df[df.code.isin(code_list)]
if df.empty: return True
profit_code_list = self.get_profit_stocks(df)
bull_stock_num = len(profit_code_list)
bull_ration = 100 * bull_stock_num / len(df)
data = {'date': [cdate], 'ratio': [bull_ration]}
df = pd.DataFrame.from_dict(data)
if self.mysql_client.set(df, self.bull_stock_ratio_table):
return self.redis.sadd(self.bull_stock_ratio_table, cdate)
return False
class Emotion:
def __init__(self, dbinfo = ct.DB_INFO, redis_host = None):
self.dbinfo = dbinfo
self.emotion_table = ct.EMOTION_TABLE
self.redis = create_redis_obj() if redis_host is None else create_redis_obj(redis_host)
self.mysql_client = CMySQL(self.dbinfo, iredis = self.redis)
self.rstock_client = RIndexStock(dbinfo, redis_host)
self.logger = getLogger(__name__)
if not self.create(): raise Exception("create emotion table failed")
def create(self):
if self.emotion_table not in self.mysql_client.get_all_tables():
sql = 'create table if not exists %s(date varchar(10) not null, score float, PRIMARY KEY (date))' % self.emotion_table
if not self.mysql_client.create(sql, self.emotion_table): return False
return True
def get_score(self, cdate = None):
if cdate is None:
sql = "select * from %s" % self.emotion_table
else:
sql = "select * from %s where date=\"%s\"" %(self.emotion_table, cdate)
return self.mysql_client.get(sql)
def get_stock_data(self, cdate):
df_byte = self.redis.get(ct.TODAY_ALL_STOCK)
if df_byte is None: return None
df = _pickle.loads(df_byte)
return df.loc[df.date == date]
def update(self, end_date = None, num = 3):
if end_date is None: end_date = datetime.now().strftime('%Y-%m-%d')
start_date = get_day_nday_ago(end_date, num = num, dformat = "%Y-%m-%d")
succeed = True
for mdate in get_dates_array(start_date, end_date):
if CCalendar.is_trading_day(mdate, redis = self.redis):
if not self.set_score(mdate):
succeed = False
self.logger.info("set score for %s set failed" % mdate)
return succeed
def set_score(self, cdate = datetime.now().strftime('%Y-%m-%d')):
stock_info = self.rstock_client.get_data(cdate)
limit_info = CLimit(self.dbinfo).get_data(cdate)
if stock_info.empty or limit_info.empty:
self.logger.error("info is empty failed")
return False
limit_up_list = limit_info[(limit_info.pchange > 0) & (limit_info.prange != 0)].reset_index(drop = True).code.tolist()
limit_down_list = limit_info[limit_info.pchange < 0].reset_index(drop = True).code.tolist()
limit_up_list.extend(limit_down_list)
total = 0
for _index, pchange in stock_info.pchange.iteritems():
code = stock_info.loc[_index, 'code']
if code in limit_up_list:
total += 2 * pchange
else:
total += pchange
aver = total / len(stock_info)
data = {'date':[cdate], 'score':[aver]}
df = pd.DataFrame.from_dict(data)
return self.mysql_client.set(df, self.emotion_table)