def hot_min(self, date, time, kline):
data = ModelData.read_data(table_name=kline,
field=['stock_code', 'close', 'open'],
date=date,
time=time)
if len(data) <= 0 and time == 1130:
data = ModelData.read_data(table_name=kline,
field=['stock_code', 'close', 'open'],
date=date,
time=1300)
data = data.drop(['_id'], axis=1)
data['profit'] = data.close / data.open - 1
data['profit'] = data[abs(data.profit) < 0.11].profit
data['pro'] = data.profit
# data['pro'] = data.profit.map(lambda x: 1 if x > 0 else 0)
data = data.dropna()
data = pd.merge(stock_codes, data, on=['stock_code'])
data = data.fillna(0)
# print(data.stock_code.tolist())
# data=data[0:56*56]
arr = np.array(data.pro.tolist() + [0] *
(64 * 64 - len(data))).reshape(64, 64)
# ax = sns.heatmap(arr, vmin=-0.03, vmax=0.03,cbar=False)
# plt.matshow(arr, cmap='hot')
# plt.show()
BaseModel('hot_min5_version4').insert_batch(
dict(date=date, time=time, value=arr.tolist()))
def get_k(cls, stock_code, sampleNo, kline, date, index_profit):
# print(len(date))
kline = "simulation_" + kline
profit = Stochastic.get_random(sigma=0.01, sampleNo=sampleNo)
high_increse = np.abs(
Stochastic.get_random(sigma=0.005, sampleNo=sampleNo))
low_increase = np.abs(
Stochastic.get_random(sigma=0.005, sampleNo=sampleNo))
resel = []
reselh = []
resell = []
r = 1
for i, ss in enumerate(profit):
r *= 1 + profit[i] + index_profit[i]
h = (r *
1.1) if (1 +
high_increse[i]) > 1.1 else r * (1 + high_increse[i])
l = (r * 0.9) if r * (1 - low_increase[i]) < 0.9 else r * (
1 - low_increase[i])
resel.append(r)
reselh.append(h)
resell.append(l)
dd = pd.DataFrame({
'stock_code': stock_code,
'date': date,
'low': resell,
'high': reselh,
'close': resel,
'open': resel
})
ModelData.insert_data(table_name=kline, data=dd)
def get_stock_basics(cls): #上市公司基本情况
data = ts.get_stock_basics()
data['code'] = data.index
data.reset_index(drop=True, inplace=True)
ModelData.remove_data(table_name='stock_basics')
print('remove success')
ModelData.insert_data(table_name='stock_basics', data=data)
print('write success')
def get_cashflow_data(cls, year, quarter):
data = ts.get_cashflow_data(year, quarter)
data['year'] = year
data['quarter'] = quarter
data.reset_index(drop=True, inplace=True)
ModelData.remove_data(table_name='cashflow_data',
year=year,
quarter=quarter)
print('remove success')
ModelData.insert_data(table_name='cashflow_data', data=data)
print('write success')
def get_profit_data(cls, year, quarter): #反应盈利能力的
data = ts.get_profit_data(year, quarter)
data['year'] = year
data['quarter'] = quarter
data.reset_index(drop=True, inplace=True)
ModelData.remove_data(table_name='profit_data',
year=year,
quarter=quarter)
print('remove success')
ModelData.insert_data(table_name='profit_data', data=data)
print('write success')
print(data.head(5))
def hot_day(self, date):
data = ModelData.read_data(table_name='kline_day',
field=['stock_code', 'close', 'open'],
date=date)
data = data.drop(['_id'], axis=1)
data['profit'] = data.close / data.open - 1
data['profit'] = data[abs(data.profit) < 0.11].profit
data['pro'] = data.profit.map(lambda x: True if x > 0 else False)
data = data.dropna()
# data=data[0:56*56]
arr = np.array(data.pro.tolist() + [0] *
(64 * 64 - len(data))).reshape(64, 64)
# ax = sns.heatmap(arr, vmin=-0.03, vmax=0.03,cbar=False)
plt.matshow(arr, cmap='hot')
plt.show()
def add_industry(cls, table):
from Calf import ModelData
import pymongo
url = 'mongodb://*****:*****@192.168.1.104:27017/?connectTimeoutMS=2000;authSource=admin'
industry = ModelData.read_data(table_name='platestock')
dic = {}
for i, r in industry.iterrows():
dic[r.stock_code] = r.hangye
myclient = pymongo.MongoClient(url)
mydb = myclient["ppp"]
mycol = mydb[table]
for s_c in industry.stock_code.tolist():
myquery = {"stock_code": s_c}
newvalues = {"$set": {"hangye": dic[s_c]}}
x = mycol.update_many(myquery, newvalues)
print(x.modified_count, s_c + "文档已修改")
myclient.close()
try:
data, address = s.recvfrom(65535)
data = data.decode('utf-8')
print('Server received from {}:{}'.format(address, data))
data = json.loads(data)
if data['identification'] == 'getdata':
result[data['kline']] = True
if result['index_min5'] and result['kline_min5']:
print('start')
# dosomething
# your function
result = {'kline_min5': False, 'index_min5': False}
except Exception as e:
print(e)
# Messages.receive()
# import datetime as dt
# date=dt.datetime.now()
# # messages = {'identification': 'getdata', 'status': 200, 'kline': 'index_min5', 'other': 'ok','datetime':date}
# messages2 = {'identification': 'getdata', 'status': 200, 'kline': 'kline_min30', 'other': 'ok','datetime':str(date)}
# # Messages.send(messages)
# Messages.send(messages2)
import datetime as dt
kline = 'kline_min5'
date = dt.datetime(2018, 8, 6)
data = ModelData.read_data(table_name=kline,
field=['stock_code', 'close', 'open'],
date=date,
time=1130)
print(data.stock_code.tolist())
# plt.show()
BaseModel('hot_min5_version4').insert_batch(
dict(date=date, time=time, value=arr.tolist()))
def fun(data):
for i, r in data.iterrows():
try:
print('date', r.date, 'time', r.time)
Hot().hot_min(date=r.date, time=r.time, kline='kline_min5')
except Exception as e:
print(e)
# Hot().hot_day(date=dt.datetime(2018,8,1))
if __name__ == '__main__':
data = ModelData.read_data(table_name='kline_min5',
field=['date', 'time'],
stock_code='000001',
date={
'$gte': dt.datetime(2016, 1, 1),
'$lte': dt.datetime(2018, 8, 6)
})
m = 10
# data=data[0:1]
li = [data[i:i + m] for i in range(0, len(data), m)]
pool = multiprocessing.Pool(processes=3)
pool.map(fun, li)