dsyoon/DeepStock
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

init

Browse Source
This commit is contained in:
dsyoon
2023-01-15 12:40:10 +09:00
parent 9f9c7e290e
commit 5821f7bfa5
7 changed files with 923 additions and 307 deletions
Download Patch File Download Diff File Expand all files Collapse all files

584
hts/BuySellChecker.py
View File

@@ -1,4 +1,5 @@
import pandas as pd
from stock.analysis.Common import Common
from stock.analysis.Stochastic import Stochastic
from stock.analysis.RSI import RSI
@@ -6,7 +7,6 @@ from stock.analysis.MACD import MACD
from stock.analysis.IchimokuCloud import IchimokuCloud
class BuySellChecker:
common = None
stochastic = None
rsi = None
@@ -23,8 +23,8 @@ class BuySellChecker:
self.ichimokuCloud = IchimokuCloud()
self.BUY_COUNT = 0
return
return
def isYangbong(self, data, i):
if data['close'][i] > data['open'][i]:
@@ -67,23 +67,28 @@ class BuySellChecker:
check_under_20_for_10 = False
for c in range(i - within, i - during):
value = [1 if max(data['avg3'][d], data['avg6'][d], data['avg9'][d], data['avg12'][d]) < data['avg20'][d] else 0 for d in range(c, c + during)]
value = [
1 if max(data['avg3'][d], data['avg6'][d], data['avg9'][d], data['avg12'][d]) < data['avg20'][d] else 0
for d in range(c, c + during)]
if len(value) == during and sum(value) == during:
check_under_20_for_10 = True
break
return check_under_20_for_10
def max_min_avg(self, data, i):
return max(data['avg3'][i], data['avg6'][i], data['avg9'][i], data['avg12'][i], data['avg20'][i]) - min(data['avg3'][i], data['avg6'][i], data['avg9'][i], data['avg12'][i], data['avg20'][i])
return max(data['avg3'][i], data['avg6'][i], data['avg9'][i], data['avg12'][i], data['avg20'][i]) - min(
data['avg3'][i], data['avg6'][i], data['avg9'][i], data['avg12'][i], data['avg20'][i])
def check_inverse_arrangement_before(self, data, i, within=30, during=5):
if i - within < 381:
return False
inverse_arrangement = False
for c in range(i-within, i-during):
value = [1 if data['avg3'][d] < data['avg6'][d] < data['avg9'][d] < data['avg12'][d] < data['avg20'][d] else 0 for d in range(c, c + during)]
if len(value) == during and sum(value) == during-1:
for c in range(i - within, i - during):
value = [
1 if data['avg3'][d] < data['avg6'][d] < data['avg9'][d] < data['avg12'][d] < data['avg20'][d] else 0
for d in range(c, c + during)]
if len(value) == during and sum(value) == during - 1:
inverse_arrangement = True
break
return inverse_arrangement
@@ -91,15 +96,15 @@ class BuySellChecker:
def checkUpDirection(self, data, i):
# 0: 무추세, -1: 하락 추세, 1: 상승 추세
close = data['close'][i]
#up_count = sum([1 if data['high'][c] < close else 0 for c in range(i-20, i)])
#down_count = sum([1 if close < data['low'][c] else 0 for c in range(i - 20, i)])
# up_count = sum([1 if data['high'][c] < close else 0 for c in range(i-20, i)])
# down_count = sum([1 if close < data['low'][c] else 0 for c in range(i - 20, i)])
lagging_change = sum([1 if data['laggingSpan'][c] < data['changeLine'][c] else 0 for c in range(i-20, i)])
change_lagging = sum([1 if data['laggingSpan'][c] > data['changeLine'][c] else 0 for c in range(i-20, i)])
lagging_change = sum([1 if data['laggingSpan'][c] < data['changeLine'][c] else 0 for c in range(i - 20, i)])
change_lagging = sum([1 if data['laggingSpan'][c] > data['changeLine'][c] else 0 for c in range(i - 20, i)])
if lagging_change>10:
if lagging_change > 10:
return 1
if change_lagging==20:
if change_lagging == 20:
return -1
return 0
@@ -159,8 +164,10 @@ class BuySellChecker:
if i > START_TIME_INDEX:
# 매도 분석
if data['changeLine'][i - 1] >= data['laggingSpan'][i - 1] and data['laggingSpan'][i] < data['changeLine'][i]:
changeLine_count = sum([1 if data['changeLine'][c] <= data['laggingSpan'][c] else 0 for c in range(i - 20, i)])
if data['changeLine'][i - 1] >= data['laggingSpan'][i - 1] and data['laggingSpan'][i] < data['changeLine'][
i]:
changeLine_count = sum(
[1 if data['changeLine'][c] <= data['laggingSpan'][c] else 0 for c in range(i - 20, i)])
if changeLine_count >= 17:
sell = min(data["open"][i], data["close"][i])
weight = 1
@@ -225,8 +232,10 @@ class BuySellChecker:
if i > START_TIME_INDEX:
# 매도 분석
if data['changeLine'][i - 1] >= data['laggingSpan'][i - 1] and data['laggingSpan'][i] < data['changeLine'][i]:
changeLine_count = sum([1 if data['changeLine'][c] <= data['laggingSpan'][c] else 0 for c in range(i - 20, i)])
if data['changeLine'][i - 1] >= data['laggingSpan'][i - 1] and data['laggingSpan'][i] < data['changeLine'][
i]:
changeLine_count = sum(
[1 if data['changeLine'][c] <= data['laggingSpan'][c] else 0 for c in range(i - 20, i)])
if changeLine_count >= 17:
sell = min(data["open"][i], data["close"][i])
weight = 1
@@ -273,8 +282,8 @@ class BuySellChecker:
df = pd.DataFrame(close)
max20 = df.rolling(window=20).mean()
stddev20 = df.rolling(window=20).std()
upper_df = max20 + (stddev20 * 2) # 상단 볼린저 밴드
lower_df = max20 - (stddev20 * 2) # 하단 볼린저 밴드
upper_df = max20 + (stddev20 * 2) # 상단 볼린저 밴드
lower_df = max20 - (stddev20 * 2) # 하단 볼린저 밴드
upper, lower = [], []
for i in range(len(upper_df)):
@@ -321,13 +330,15 @@ class BuySellChecker:
temp = {
"date": point_temp,
"open": open, "high": high, "low": low, "close": close, "volume": vol,
"avg3": avg3,"avg6": avg6,"avg9": avg9,"avg12": avg12, "avg20": avg20,
"disparity_avg5": disparity_avg5, "disparity_avg10": disparity_avg10, "disparity_avg20": disparity_avg20, "disparity_avg60": disparity_avg60, "disparity_avg120": disparity_avg120,
"avg3": avg3, "avg6": avg6, "avg9": avg9, "avg12": avg12, "avg20": avg20,
"disparity_avg5": disparity_avg5, "disparity_avg10": disparity_avg10, "disparity_avg20": disparity_avg20,
"disparity_avg60": disparity_avg60, "disparity_avg120": disparity_avg120,
"upper": upper, "lower": lower,
"macd": macd, "macds": macds, "macdo": macdo,
"fast_k": fast_k, "slow_k": slow_k, "slow_d": slow_d,
"rsi": rsi, "rsis": rsis,
"changeLine": changeLine, "baseLine": baseLine, "laggingSpan": laggingSpan, "leadingSpan1": leadingSpan1, "leadingSpan2": leadingSpan2,
"changeLine": changeLine, "baseLine": baseLine, "laggingSpan": laggingSpan, "leadingSpan1": leadingSpan1,
"leadingSpan2": leadingSpan2,
"label": label
}
@@ -368,9 +379,15 @@ class BuySellChecker:
avg54_list = close_df.rolling(window=54).mean().fillna(close[0]).values.tolist()
avg54 = [item[0] for item in avg54_list]
abs_avg_1 = [max(avg3[i], avg6[i], avg9[i], avg12[i], avg27[i], avg54[i]) - min(avg3[i], avg6[i], avg9[i], avg12[i], avg27[i], avg54[i]) for i in range(0, len(close))]
abs_avg_2 = [max(avg3[i], avg6[i], avg9[i], avg12[i], avg27[i]) - min(avg3[i], avg6[i], avg9[i], avg12[i], avg27[i]) for i in range(0, len(close))]
abs_avg_3 = [max(avg3[i], avg6[i], avg9[i], avg12[i]) - min(avg3[i], avg6[i], avg9[i], avg12[i]) for i in range(0, len(close))]
abs_avg_1 = [
max(avg3[i], avg6[i], avg9[i], avg12[i], avg27[i], avg54[i]) - min(avg3[i], avg6[i], avg9[i], avg12[i],
avg27[i], avg54[i]) for i in
range(0, len(close))]
abs_avg_2 = [
max(avg3[i], avg6[i], avg9[i], avg12[i], avg27[i]) - min(avg3[i], avg6[i], avg9[i], avg12[i], avg27[i]) for
i in range(0, len(close))]
abs_avg_3 = [max(avg3[i], avg6[i], avg9[i], avg12[i]) - min(avg3[i], avg6[i], avg9[i], avg12[i]) for i in
range(0, len(close))]
abs_avg_4 = [max(avg3[i], avg6[i], avg9[i]) - min(avg3[i], avg6[i], avg9[i]) for i in range(0, len(close))]
abs_avg_5 = [max(avg3[i], avg6[i]) - min(avg3[i], avg6[i]) for i in range(0, len(close))]
@@ -391,7 +408,7 @@ class BuySellChecker:
diff_avg3, diff_avg6, diff_avg9, diff_avg12, diff_avg27, diff_avg54 = [], [], [], [], [], []
diff_avg3.append(0)
for i in range(1, len(avg3)):
diff_avg3.append(avg3[i]-avg3[i-1])
diff_avg3.append(avg3[i] - avg3[i - 1])
diff_avg6.append(0)
for i in range(1, len(avg6)):
diff_avg6.append(avg6[i] - avg6[i - 1])
@@ -408,7 +425,6 @@ class BuySellChecker:
for i in range(1, len(avg54)):
diff_avg54.append(avg54[i] - avg54[i - 1])
diff_avg3_avg6 = [avg3[i] - avg6[i] for i in range(0, len(close))]
diff_avg3_avg9 = [avg3[i] - avg9[i] for i in range(0, len(close))]
diff_avg3_avg12 = [avg3[i] - avg12[i] for i in range(0, len(close))]
@@ -429,8 +445,8 @@ class BuySellChecker:
df = pd.DataFrame(close)
max20 = df.rolling(window=20).mean()
stddev20 = df.rolling(window=20).std()
upper_df = max20 + (stddev20 * 2) # 상단 볼린저 밴드
lower_df = max20 - (stddev20 * 2) # 하단 볼린저 밴드
upper_df = max20 + (stddev20 * 2) # 상단 볼린저 밴드
lower_df = max20 - (stddev20 * 2) # 하단 볼린저 밴드
upper, lower = [], []
for i in range(len(upper_df)):
@@ -445,7 +461,8 @@ class BuySellChecker:
STOCK = []
for i in range(len(open)):
STOCK.append({'volume': vol[i], 'close': close[i], 'open': open[i], 'high': high[i], 'low': low[i],
'avg3': avg3[i], 'avg6': avg6[i],'avg9': avg9[i],'avg12': avg12[i],'avg27': avg27[i],'avg54': avg54[i]})
'avg3': avg3[i], 'avg6': avg6[i], 'avg9': avg9[i], 'avg12': avg12[i], 'avg27': avg27[i],
'avg54': avg54[i]})
# stochastic
stochastic_df = self.stochastic.apply(STOCK, n=30, m=5, t=5)
@@ -484,7 +501,6 @@ class BuySellChecker:
diff_low_lower = [low[i] - lower[i] for i in range(0, len(low))]
diff_low_upper = [low[i] - upper[i] for i in range(0, len(low))]
##### 일목균형표
diff_lead1_lead2 = [leadingSpan1[i] - leadingSpan2[i] for i in range(0, len(leadingSpan1))]
diff_change_base = [changeLine[i] - baseLine[i] for i in range(0, len(baseLine))]
@@ -512,7 +528,6 @@ class BuySellChecker:
diff_slowk_slowd = [slow_k[i] - slow_d[i] for i in range(0, len(slow_k))]
diff_rsi_rsis = [rsi[i] - rsis[i] for i in range(0, len(rsi))]
diff_macd, diff_macdo, diff_macds = [], [], []
diff_macd.append(0)
for i in range(1, len(macd)):
@@ -570,14 +585,16 @@ class BuySellChecker:
"rsi": rsi, "rsis": rsis,
"changeLine": changeLine, "baseLine": baseLine, "leadingSpan1": leadingSpan1, "leadingSpan2": leadingSpan2,
"height": height, "top_tail_height": top_tail_height, "bottom_tail_height": bottom_tail_height,
"abs_avg_1": abs_avg_1, "abs_avg_2": abs_avg_2, "abs_avg_3": abs_avg_3, "abs_avg_4": abs_avg_4, "abs_avg_5": abs_avg_5,
"abs_avg_1": abs_avg_1, "abs_avg_2": abs_avg_2, "abs_avg_3": abs_avg_3, "abs_avg_4": abs_avg_4,
"abs_avg_5": abs_avg_5,
"diff_open": diff_open, "diff_close": diff_close, "diff_low": diff_low, "diff_high": diff_high,
"diff_avg3":diff_avg3, "diff_avg6":diff_avg6, "diff_avg9":diff_avg9, "diff_avg12":diff_avg12, "diff_avg27":diff_avg27, "diff_avg54":diff_avg54,
"diff_macd":diff_macd, "diff_macdo":diff_macdo, "diff_macds":diff_macds,
"diff_fast_k":diff_fast_k, "diff_slow_k":diff_slow_k, "diff_slow_d":diff_slow_d,
"diff_rsi":diff_rsi, "diff_rsis":diff_rsis,
"diff_changeLine":diff_changeLine, "diff_baseLine":diff_baseLine,
"diff_upper":diff_upper, "diff_lower":diff_lower,
"diff_avg3": diff_avg3, "diff_avg6": diff_avg6, "diff_avg9": diff_avg9, "diff_avg12": diff_avg12,
"diff_avg27": diff_avg27, "diff_avg54": diff_avg54,
"diff_macd": diff_macd, "diff_macdo": diff_macdo, "diff_macds": diff_macds,
"diff_fast_k": diff_fast_k, "diff_slow_k": diff_slow_k, "diff_slow_d": diff_slow_d,
"diff_rsi": diff_rsi, "diff_rsis": diff_rsis,
"diff_changeLine": diff_changeLine, "diff_baseLine": diff_baseLine,
"diff_upper": diff_upper, "diff_lower": diff_lower,
"diff_avg3_avg6": diff_avg3_avg6,
"diff_avg3_avg9": diff_avg3_avg9,
"diff_avg3_avg12": diff_avg3_avg12,
@@ -790,7 +807,6 @@ class BuySellChecker:
return bsLine, data
def checkTransactionML(self, data, stock_code, predY, isRealTime=True):
# 4일치 중에서 앞에 2일은 제거한다.
date = data['date'].dt.date.unique().tolist()
@@ -841,88 +857,68 @@ class BuySellChecker:
# middle line에 맞다은 적 없이, low line에 붙었거나 아래에 있었던 캔들의 높은 가격을 얻어옴
def getPrice_UnderLowWithoutMiddle(self, last_index, data):
if data['high'][last_index] < data['envelope_middle'][last_index]:
for i in range(last_index - 1, 0, -1):
if data['high'][i] > data['envelope_middle'][i]:
return -1, -1
if data['low'][i] < data['envelope_lower'][i]:
return i, max(data['open'][i], data['close'][i])
return -1, -1
def getBuyPriceAndWeight_Envelope_trend(self, data, i):
buy, weight, type = -1, -1, -1
def getBuyPriceAndWeight_Envelope(self, i, data):
buy, weight = -1, -1
if data['close'][i-1] < data['envelope_lower'][i-1] and data['envelope_lower'][i] < data['close'][i]:
buy = data["close"][i]
weight = 1
type = 1
"""
# middle line에 맞다은 적 없이, low line에 붙었거나 아래에 있었던 캔들의 높은 가격을 얻어옴
index, price = self.getPrice_UnderLowWithoutMiddle(i, data)
if price > -1:
# 해당 가격보다 높은 가격이면 매수한다
if price < data['close'][i]:
buy = data['close'][i]
weight = 10
"""
if i > 100:
if -0.004 < data['gradient1'][i] < 0.001:
#if data['high'][i] < data['envelope_middle'][i]:
if data['slow_k'][i] < 20:
buy = data['close'][i]
weight = 10
if data['close'][i] < data['envelope_lower'][i]:
if data['rsis'][i] < data['rsi'][i]:
buy = data["close"][i]
weight = 1
type = 2
"""
if i > 100:
if min(data['gradient1'][i-5:i]) < -0.009 and -0.009 < data['gradient1'][i]:
if data['high'][i] < data['envelope_middle'][i]:
buy = data['close'][i]
weight = 10
"""
return buy, weight
if data['close'][i-1] < data['envelope_middle'][i-1] and data['envelope_middle'][i] < data['close'][i]:
if data['slow_k'][i] < 25:
buy = data["close"][i]
weight = 1
type = 3
check1 = False
check2 = False
if data['slow_k'][i] < 40:
if data['close'][i-1] < data['envelope_middle'][i-1] and data['envelope_middle'][i] < data['close'][i]:
idx = -1
for t1 in range(i-1, i-10, -1):
if data['close'][t1] < data['envelope_middle'][t1]:
check1 = True
idx = t1
break
if check1:
for t1 in range(idx-1, i - 10, -1):
if data['envelope_middle'][t1] < data['close'][t1]:
check2 = True
break
if check2:
buy = data["close"][i]
weight = 1
type = 4
if data.index[i].strftime("%Y.%m.%d") == "2021.12.21":
print(1)
check = True
if data['slow_k'][i] < 40:
if data['close'][i - 1] < data['envelope_middle'][i - 1] and data['envelope_middle'][i] < data['close'][i]:
for t1 in range(i-1, i-10, -1):
if data['envelope_middle'][t1] < data['close'][t1]:
check = False
if check:
buy = data["close"][i]
weight = 1
type = 4
return buy, weight, type
def getSellPriceAndWeight_Envelope_trend(self, data, i):
def getSellPriceAndWeight_Envelope(self, data, i):
sell, weight, type = -1, -1, -1
if data.index[i].strftime("%Y.%m.%d") == "2022.12.01":
print (1)
print(1)
# upper lined에서 처리
if data['close'][i-1] < data['envelope_upper'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i-1] <= data['slow_k'][i - 1] and data['slow_k'][i] <= data['slow_d'][i]:
if data['close'][i - 1] < data['envelope_upper'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i - 1] <= data['slow_k'][i - 1] and data['slow_k'][i] <= data['slow_d'][i]:
sell = data["close"][i]
weight = 1
type = 1
if data['envelope_upper'][i-1] < data['close'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i-1] <= data['slow_k'][i - 1] and data['slow_k'][i] <= data['slow_d'][i]:
if data['envelope_upper'][i - 1] < data['close'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i - 1] <= data['slow_k'][i - 1] and data['slow_k'][i] <= data['slow_d'][i]:
sell = data["close"][i]
weight = 1
type = 2
if data['envelope_upper'][i-1] < data['close'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i-1] +2 <= data['slow_k'][i - 1] and data['slow_d'][i]+1 == data['slow_k'][i]:
if data['envelope_upper'][i - 1] < data['close'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i - 1] + 2 <= data['slow_k'][i - 1] and data['slow_d'][i] + 1 == data['slow_k'][i]:
sell = data["close"][i]
weight = 1
type = 3
if data['envelope_upper'][i]<data['high'][i] and data['open'][i] < data['close'][i]:
if data['envelope_upper'][i] < data['high'][i] and data['open'][i] < data['close'][i]:
if data['close'][i] - data['open'][i] < data['high'][i] - data['close'][i]:
sell = data["close"][i]
weight = 1
@@ -930,109 +926,303 @@ class BuySellChecker:
return sell, weight, type
def getBuyPriceAndWeight_Envelope(self, data, i):
buy, weight, type = -1, -1, -1
# 팔아야 할 시점을 체크하기 위함
# 이전에 산 가격보다 지금 5원이상 떨어졌다면 매도 한다.
def checkBelow5WonFromPreviousBuyPrice(self, last_index, data, price):
for i in range(last_index - 1, 0, -1):
if data['sell'][i] != -1:
return False
if data['buy'][i] != -1:
if data['buy'][i] - price > 5:
return True
return
if data['close'][i-1] < data['envelope_lower'][i-1] and data['envelope_lower'][i] < data['close'][i]:
buy = data["close"][i]
weight = 1
type = 1
def checkWithEnvelope_252670(self, data, isRealTime=False):
if data['close'][i] < data['envelope_lower'][i]:
if data['rsis'][i] < data['rsi'][i]:
buy = data["close"][i]
weight = 1
type = 2
return buy, weight, type
def getSellPriceAndWeight_Envelope(self, data, i):
sell, weight, type = -1, -1, -1
if data.index[i].strftime("%Y.%m.%d") == "2022.12.01":
print (1)
# upper lined에서 처리
if data['close'][i-1] < data['envelope_upper'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i-1] <= data['slow_k'][i - 1] and data['slow_k'][i] <= data['slow_d'][i]:
sell = data["close"][i]
weight = 1
type = 1
if data['envelope_upper'][i-1] < data['close'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i-1] <= data['slow_k'][i - 1] and data['slow_k'][i] <= data['slow_d'][i]:
sell = data["close"][i]
weight = 1
type = 2
if data['envelope_upper'][i-1] < data['close'][i - 1] and data['envelope_upper'][i] < data['close'][i]:
if data['slow_d'][i-1] +2 <= data['slow_k'][i - 1] and data['slow_d'][i]+1 == data['slow_k'][i]:
sell = data["close"][i]
weight = 1
type = 3
# middle lined에서 처리
middle_up = False
if data['envelope_middle'][i-1] < data['close'][i-1] or data['envelope_middle'][i] < data['open'][i]:
if data['envelope_middle'][i-2] < data['close'][i-2] or data['envelope_middle'][i-1] < data['close'][i-1] or data['envelope_middle'][i] < data['close'][i]:
for t in range(i-1, i-20, -1):
if data['envelope_upper'][t] <= data['close'][t]:
middle_up = False
break
if data['close'][t] < data['envelope_middle'][t]:
middle_up = True
break
if middle_up:
if data['slow_d'][i-1] < data['slow_k'][i-1] and data['slow_k'][i] < data['slow_d'][i]:
sell = data["close"][i]
weight = 1
type = 4
return sell, weight, type
def checkEnvelopeTiming(self, data, stock_code, isRealTime=True):
# 어제 오늘 데이터로 분석
bsLine = {}
size = len(data["close"])
if isRealTime:
# isRealTime=True, 실시간 적용
last_index = size - 1
bsLine['buy'] = [-1 for i in range(size)]
bsLine['buy_weight'] = [-1 for i in range(size)]
bsLine['sell'] = [-1 for i in range(size)]
bsLine['sell_weight'] = [-1 for i in range(size)]
if stock_code == "252670":
buy, buy_weight, buy_type = self.getBuyPriceAndWeight_Envelope(data, last_index)
sell, sell_weight, sell_type = self.getSellPriceAndWeight_Envelope(data, last_index)
else:
buy, buy_weight, buy_type = self.getBuyPriceAndWeight_Envelope(data, last_index)
sell, sell_weight, sell_type = self.getSellPriceAndWeight_Envelope(data, last_index)
for i in range(size):
if isRealTime:
if i < size - 1:
continue
bsLine['buy'] = [buy]
bsLine['buy_weight'] = [buy_weight]
bsLine['sell'] = [sell]
bsLine['sell_weight'] = [sell_weight]
"""
# 이전에 산 가격보다 지금 5원이상 떨어졌다면 매도 한다.
price = data['close'][i] if data['close'][i] >= data['open'][i] else data['open'][i]
if self.checkBelow5WonFromPreviousBuyPrice(i, data, price):
data['sell'][i] = price
bsLine['sell'][i] = price
bsLine['sell_weight'][i] = 50
return bsLine, data
"""
else:
# Type=False, 시뮬레이션 적용
bsLine['buy'] = [-1 for i in range(size)]
bsLine['buy_weight'] = [-1 for i in range(size)]
bsLine['sell'] = [-1 for i in range(size)]
bsLine['sell_weight'] = [-1 for i in range(size)]
"""
# middle line에 맞다은 적 없이, low line에 붙었거나 아래에 있었던 캔들의 높은 가격을 얻어옴
buy, buy_weight = self.getBuyPriceAndWeight_Envelope(i, data)
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = buy_weight
return bsLine, data
"""
for i in range(size):
if stock_code == "252670":
buy, buy_weight, buy_type = self.getBuyPriceAndWeight_Envelope_trend(data, i)
sell, sell_weight, sell_type = self.getSellPriceAndWeight_Envelope_trend(data, i)
else:
buy, buy_weight, buy_type = self.getBuyPriceAndWeight_Envelope(data, i)
sell, sell_weight, sell_type = self.getSellPriceAndWeight_Envelope(data, i)
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = buy_weight
bsLine['sell'][i] = sell
bsLine['sell_weight'][i] = sell_weight
if i > 0:
if bsLine['sell'][i-1] > -1:
bsLine['sell'][i] = -1
bsLine['sell_weight'][i] = -1
if 0 < data['gradients_avg60'][i] < 0.001:
if data['high'][i] < data['envelope_middle'][i]:
if -0.015 < data['gradients_avg5'][i] and -0.007 < data['gradients_avg20'][i]:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 10
if i > 10:
if (
data['gradients_avg60'][i - 10] > 0 and data['gradients_avg60'][i - 9] > 0 and data['gradients_avg60'][i - 8] > 0 and
data['gradients_avg60'][i - 7] > 0 and data['gradients_avg60'][i - 6] > 0 and data['gradients_avg60'][i - 5] > 0 and
data['gradients_avg60'][i - 4] > 0 and data['gradients_avg60'][i - 3] > 0 and data['gradients_avg60'][i - 2] > 0 and
data['gradients_avg60'][i - 1] > 0 and data['gradients_avg60'][i] < 0
):
if data['disparity'][i] < 3:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 10
if i > 10:
if data['disparity_avg60'][i] < 65:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
if data['slow_k'][i] < 3:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 10
return bsLine, data
def checkWithEnvelope_122630(self, data, isRealTime=False):
bsLine = {}
size = len(data["close"])
bsLine['buy'] = [-1 for i in range(size)]
bsLine['buy_weight'] = [-1 for i in range(size)]
bsLine['sell'] = [-1 for i in range(size)]
bsLine['sell_weight'] = [-1 for i in range(size)]
for i in range(size):
if isRealTime:
if i < size - 1:
continue
if i > 10:
if data['disparity_avg60'][i] < 60:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
if data['macd'][i] < -1000:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
if data['slow_k'][i] < 7:
if data['slow_d'][i] < data['slow_k'][i]:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 5
return bsLine, data
def notBuy(self, data, i):
if i > 5:
check = True
for l in range(i-4, i+1):
if (
data['gradients_avg60'][l - 1] > data['gradients_avg60'][l] or
data['gradients_avg20'][l - 1] > data['gradients_avg20'][l] or
data['gradients_low'][l - 1] > data['gradients_low'][l]
):
check = False
break
if not check:
return False
return True
def checkWithEnvelope(self, data, isRealTime=False):
bsLine = {}
size = len(data["close"])
bsLine['buy'] = [-1 for i in range(size)]
bsLine['buy_weight'] = [-1 for i in range(size)]
bsLine['sell'] = [-1 for i in range(size)]
bsLine['sell_weight'] = [-1 for i in range(size)]
for i in range(size):
if isRealTime:
if i < size - 1:
continue
if i > 10:
if data['disparity'][i] < 2:
check = True
for l in range(i-3, i):
if (
data['gradients_avg60'][l-1] > data['gradients_avg60'][l] or
data['gradients_avg20'][l-1] > data['gradients_avg20'][l] or
data['gradients_low'][l-1] > data['gradients_low'][l] or
data['disparity_avg5'][l-1] > data['disparity_avg5'][l] or
data['disparity'][l-1] < data['disparity'][l]
):
check = False
break
if check and 99 < sum(data['disparity_avg5'][i-4:i+1])/5 < 100 and 99 < sum(data['disparity_avg60'][i-4:i+1])/5 < 100:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
check = True
for l in range(i - 2, i):
if (
data['gradients_avg60'][l - 1] > data['gradients_avg60'][l] or
data['gradients_low'][l - 1] > data['gradients_low'][l]
):
check = False
break
if (
check and
-0.0011 < data['gradients_low'][i] < 0 and -0.007 < data['gradients_avg5'][i] < 0.001 and
-0.0012 < data['gradients_avg60'][i] < 0 and
98.90 < data['disparity_avg5'][i] < 101
):
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
check = True
for l in range(i - 6, i):
if (
data['gradients_avg60'][l - 1] < data['gradients_avg60'][l] or
data['gradients_avg20'][l - 1] < data['gradients_avg20'][l] or
data['gradients_low'][l - 1] < data['gradients_low'][l] or
-0.039 < data['gradients_low'][l - 1] < -0.35 or
-0.05 < data['gradients_avg20'][l - 1] < -0.30 or
-0.40 < data['gradients_avg60'][l - 1] < -0.30
):
check = False
break
if check and 99 < min (data['disparity_avg5'][i - 6:i]) < max (data['disparity_avg5'][i - 6:i]) < 101:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
check = True
for l in range(i - 3, i):
if (
data['gradients_low'][l - 1] < data['gradients_low'][l] or
data['gradients_avg60'][l - 1] < data['gradients_avg60'][l] or
data['gradients_avg20'][l - 1] < data['gradients_avg20'][l] or
0.01 < data['gradients_low'][l - 1] < 0.21 or
-0.09 < data['gradients_avg20'][l - 1] < -0.002 or
0.01 < data['gradients_avg60'][l - 1] < 0.021
):
check = False
break
if check:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
if (data['disparity'][i] < 5 and 99.0 < data['disparity_avg60'][i] < 99.1 and
-0.009 < data['gradients_avg60'][i] < -0.008 and 0.015 < data['gradients_avg20'][i] < 0.016 and
-0.006 < data['gradients_avg5'][i] < -0.005 and -0.009 < data['gradients_low'][i] < -0.008):
check = True
for l in range(i-5, i):
if (
data['gradients_avg60'][l-1] > data['gradients_avg60'][l] or
data['gradients_low'][l-1] > data['gradients_low'][l] or
data['disparity'][l - 1] < data['disparity'][l]
):
check = False
break
if check:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
if data['macd'][i] < -4000:
if data['macd'][i-1] < data['macd'][i]:
if not self.notBuy(data, i) and data['slow_k'][i] < 30:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
# macd 이전에 없던 바닥인 경우 상승할 찰나 매수
if data['macds'][i-1] < min(data['macds'][:i-1]):
if data['macds'][i-1] < data['macds'][i]:
if not self.notBuy(data, i) and data['slow_k'][i] < 30:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
if (
98 < data['disparity_avg5'][i] < 100 and data['disparity_avg20'][i] < 93.5 and data['disparity_avg60'][i] < 89 and
-0.014 < data['gradients_avg60'][i] < -0.013 and -0.03 < data['gradients_avg20'][i] < -0.02 and -0.014 < data['gradients_low'][i] < -0.013 and
data['slow_k'][i] < 11
):
if not self.notBuy(data, i):
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
"""
if data['disparity_avg60'][i] < 60:
buy = data['low'][i]
data['buy'][i] = buy
bsLine['buy'][i] = buy
bsLine['buy_weight'][i] = 20
"""
return bsLine, data
def checkTransactionWithEnvelope(self, data, stock_code, isRealTime=False):
if isRealTime:
if stock_code == "252670":
bsLine, data = self.checkWithEnvelope_252670(data, isRealTime)
elif stock_code == "122630":
bsLine, data = self.checkWithEnvelope_122630(data, isRealTime)
else:
bsLine, data = self.checkWithEnvelope(data, isRealTime)
else:
# 사야 할 시점과 팔아야 할 시점을 체크한다.
if stock_code == "252670":
bsLine, data = self.checkWithEnvelope_252670(data, isRealTime)
elif stock_code == "122630":
bsLine, data = self.checkWithEnvelope_122630(data, isRealTime)
else:
bsLine, data = self.checkWithEnvelope(data, isRealTime)
return bsLine, data

317
hts/DailyStatus.py
View File

@@ -5,17 +5,21 @@ if platform.system().lower().find("window") >= 0 and platform.architecture()[0]
import win32com.client
import sqlite3
from datetime import datetime, timedelta
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import StandardScaler, MinMaxScaler
from stock.analysis.AnalyzerSqlite import AnalyzerSqlite
class DailyStatus:
tableName = None
dbFileName = None
RESOURCE_PATH = None
analyzerSqlite = None
def __init__(self, RESOURCE_PATH):
self.RESOURCE_PATH = RESOURCE_PATH
self.tableName = 'stock_analysis'
self.tableName = 'stock'
self.dbFileName = "stock.db"
self.analyzerSqlite = AnalyzerSqlite(os.path.join(self.RESOURCE_PATH, self.dbFileName))
return
def getDBData(self, stock_code, day, result):
@@ -67,13 +71,118 @@ class DailyStatus:
return True
return False
def getLastData(self, stock_code, today, n=100):
def getLastData(self, stock_code, limit=350):
stockTableName = 'stock'
conn = sqlite3.connect(os.path.join(self.RESOURCE_PATH, self.dbFileName))
cursor = conn.cursor()
stock = {"CODE": stock_code, "NAME": "", "PRICE": []}
sql = 'SELECT ymd, close, diff, open, high, low, volume FROM ' + stockTableName + ' where CODE=? order by ymd desc '
sql += ' limit ' + str(limit)
cursor.execute(sql, (stock['CODE'],))
items = cursor.fetchall()
items_reverse = reversed(items)
for item in items_reverse:
stock['PRICE'].append(
{
"ymd": item[0],
"close": item[1],
"diff": item[2],
"open": item[3],
"high": item[4],
"low": item[5],
"volume": item[6],
"avg3": -1,
"avg4": -1,
"avg5": -1,
"avg6": -1,
"avg10": -1,
"avg12": -1,
"avg20": -1,
"avg36": -1,
"avg40": -1,
"avg48": -1,
"avg60": -1,
"avg120": -1,
"avg200": -1,
"avg240": -1,
"avg300": -1,
"disparity_avg5": -1,
"disparity_avg10": -1,
"disparity_avg20": -1,
"disparity_avg60": -1,
"disparity_avg120": -1,
"bolingerband_upper": -1,
"bolingerband_lower": -1,
"bolingerband_middle": -1,
"envelope_upper": -1,
"envelope_lower": -1,
"envelope_middle": -1,
"ichimokucloud_changeLine": -1,
"ichimokucloud_baseLine": -1,
"ichimokucloud_leadingSpan1": -1,
"ichimokucloud_leadingSpan2": -1,
"stochastic_fast_k": -1,
"stochastic_slow_k": -1,
"stochastic_slow_d": -1,
"rsi": -1,
"rsis": -1,
"macd": -1,
"macds": -1,
"macdo": -1,
}
)
conn.commit()
cursor.close()
conn.close()
return stock
def analyze (self, stock, days=120):
stock["PRICE"] = sorted(stock["PRICE"], key=lambda x: x['ymd'])
self.analyzerSqlite.get_moving_average(stock["PRICE"])
# 이동 평균을 이용한 이격도 계산
self.analyzerSqlite.get_disparity(stock["PRICE"])
self.analyzerSqlite.ichimokuCloud.analyze(stock)
self.analyzerSqlite.stochastic.analyze(stock)
self.analyzerSqlite.bolingerBand.analyze(stock)
self.analyzerSqlite.envelope.analyze(stock)
self.analyzerSqlite.rsi.analyze(stock)
self.analyzerSqlite.macd.analyze(stock)
result = {
"ymd": [],
"open": [],
"close": [],
"high": [],
"low": [],
"avg3": [],
"avg4": [],
"avg5": [],
"avg6": [],
"avg10": [],
"avg12": [],
"avg20": [],
"avg36": [],
"avg40": [],
"avg48": [],
"avg60": [],
"avg120": [],
"avg200": [],
"avg240": [],
"avg300": [],
"disparity_avg5": [],
"disparity_avg20": [],
"disparity_avg60": [],
"disparity_avg120": [],
"disparity": [],
"disparity_type": [],
"envelope_upper": [],
"envelope_lower": [],
"envelope_middle": [],
@@ -82,58 +191,166 @@ class DailyStatus:
"macd": [],
"macds": [],
"slow_k": [],
"slow_d": []
"slow_d": [],
"buy": [],
"sell": [],
}
days = []
for i in range(1, n):
last_day = (datetime.strptime(today, '%Y%m%d') - timedelta(i)).strftime('%Y.%m.%d')
isValid = self.isValidYMD(stock_code, last_day)
if isValid:
days.append(last_day)
days = sorted(days)
for day in days:
self.getDBData(stock_code, day, result)
for item in stock['PRICE']:
result["ymd"].append(item['ymd'])
result["open"].append(item['open'])
result["close"].append(item['close'])
result["high"].append(item['high'])
result["low"].append(item['low'])
result["avg3"].append(item['avg3'])
result["avg4"].append(item['avg4'])
result["avg5"].append(item['avg5'])
result["avg6"].append(item['avg6'])
result["avg10"].append(item['avg10'])
result["avg12"].append(item['avg12'])
result["avg20"].append(item['avg20'])
result["avg36"].append(item['avg36'])
result["avg40"].append(item['avg40'])
result["avg48"].append(item['avg48'])
result["avg60"].append(item['avg60'])
result["avg120"].append(item['avg120'])
result["avg200"].append(item['avg200'])
result["avg240"].append(item['avg240'])
result["avg300"].append(item['avg300'])
result["disparity_avg5"].append(item['disparity_avg5'])
result["disparity_avg20"].append(item['disparity_avg20'])
result["disparity_avg60"].append(item['disparity_avg60'])
result["disparity_avg120"].append(item['disparity_avg120'])
result['disparity'].append(max(item['disparity_avg5'], item['disparity_avg20'], item['disparity_avg60']) - min(item['disparity_avg5'], item['disparity_avg20'], item['disparity_avg60']))
if item['disparity_avg60'] < item['disparity_avg20'] < item['disparity_avg5']:
result['disparity_type'].append(1)
elif item['disparity_avg5'] < item['disparity_avg20'] < item['disparity_avg60']:
result['disparity_type'].append(-1)
else:
result['disparity_type'].append(0)
result["envelope_upper"].append(item['envelope_upper'])
result["envelope_lower"].append(item['envelope_lower'])
result["envelope_middle"].append(item['envelope_middle'])
result["rsi"].append(item['rsi'])
result["rsis"].append(item['rsis'])
result["macd"].append(item['macd'])
result["macds"].append(item['macds'])
result["slow_k"].append(item['stochastic_slow_k'])
result["slow_d"].append(item['stochastic_slow_d'])
result["buy"].append(-1)
result["sell"].append(-1)
data = pd.DataFrame(result)
df_final_time = pd.DatetimeIndex(result['ymd'])
data.index = df_final_time
data = data.astype(
{
'open': 'int',
'high': 'int',
'low': 'int',
'close': 'int',
'avg3': 'float',
'avg4': 'float',
'avg5': 'float',
'avg6': 'float',
'avg10': 'float',
'avg12': 'float',
'avg20': 'float',
'avg36': 'float',
'avg40': 'float',
'avg48': 'float',
'avg60': 'float',
'avg120': 'float',
'avg200': 'float',
'avg240': 'float',
'avg300': 'float',
'disparity_avg5': 'float',
'disparity_avg20': 'float',
'disparity_avg60': 'float',
'disparity_avg120': 'float',
'buy': 'int',
'sell': 'int',
'slow_k': 'float',
'slow_d': 'float',
'macd': 'float',
'macds': 'float',
'envelope_upper': 'float',
'envelope_lower': 'float',
'envelope_middle': 'float',
'rsi': 'float',
'rsis': 'float'
}
)
scaler = StandardScaler()
low_df = pd.DataFrame(data['low'])
low_df.index = [c for c in range(len(low_df))]
low_std = scaler.fit_transform(data['low'].values.reshape(-1, 1))
low_std = pd.DataFrame(low_std, columns=['low_std'])
min_df = pd.DataFrame({'open': data['open'].to_list(), 'close': data['close'].to_list()})
min_df['min_std'] = min_df.min(axis=1)
min_df.index = [c for c in range(len(min_df))]
min_std = scaler.fit_transform(min_df['min_std'].values.reshape(-1, 1))
min_std = pd.DataFrame(min_std, columns=['min_std'])
line_fitter = LinearRegression()
size = len(data["close"])
gradients_low = []
gradients_avg5 = []
gradients_avg20 = []
gradients_avg60 = []
for i in range(size):
coef_low = -999
coef_avg5 = -999
coef_avg20 = -999
coef_avg60 = -999
if i > 0:
l = days if i >= days else i
x = pd.DataFrame([c for c in range(i - l, i + 1)])
y = pd.DataFrame(low_std.values.tolist()[i - l:i + 1])
line_fitter.fit(x.values.reshape(-1, 1), y)
coef_low = line_fitter.coef_[0][0]
l = 5 if i >= 5 else i
x = pd.DataFrame([c for c in range(i - l, i + 1)])
y = pd.DataFrame(min_std.values.tolist()[i - l:i + 1])
line_fitter.fit(x.values.reshape(-1, 1), y)
coef_avg5 = line_fitter.coef_[0][0]
l = 20 if i >= 20 else i
x = pd.DataFrame([c for c in range(i - l, i + 1)])
y = pd.DataFrame(min_std.values.tolist()[i - l:i + 1])
line_fitter.fit(x.values.reshape(-1, 1), y)
coef_avg20 = line_fitter.coef_[0][0]
l = 60 if i >= 60 else i
x = pd.DataFrame([c for c in range(i - l, i + 1)])
y = pd.DataFrame(min_std.values.tolist()[i - l:i + 1])
line_fitter.fit(x.values.reshape(-1, 1), y)
coef_avg60 = line_fitter.coef_[0][0]
gradients_low.append(coef_low)
gradients_avg5.append(coef_avg5)
gradients_avg20.append(coef_avg20)
gradients_avg60.append(coef_avg60)
gradients_low_df = pd.DataFrame(gradients_low, columns=['gradients_low'])
gradients_avg5_df = pd.DataFrame(gradients_avg5, columns=['gradients_avg5'])
gradients_avg20_df = pd.DataFrame(gradients_avg20, columns=['gradients_avg20'])
gradients_avg60_df = pd.DataFrame(gradients_avg60, columns=['gradients_avg60'])
gradients_low_df.index = df_final_time
gradients_avg5_df.index = df_final_time
gradients_avg20_df.index = df_final_time
gradients_avg60_df.index = df_final_time
data = data.merge(gradients_low_df, left_index=True, right_index=True)
data = data.merge(gradients_avg5_df, left_index=True, right_index=True)
data = data.merge(gradients_avg20_df, left_index=True, right_index=True)
data = data.merge(gradients_avg60_df, left_index=True, right_index=True)
return data
def getLastData_realtime(self, stock_code, today, n=100):
result = {
"ymd": [],
"open": [],
"close": [],
"high": [],
"low": [],
"envelope_upper": [],
"envelope_lower": [],
"envelope_middle": [],
"rsi": [],
"rsis": [],
"macd": [],
"macds": [],
"slow_k": [],
"slow_d": []
}
days = []
for i in range(1, n):
last_day = (datetime.strptime(today, '%Y%m%d') - timedelta(i)).strftime('%Y.%m.%d')
isValid = self.isValidYMD(stock_code, last_day)
if isValid:
days.append(last_day)
days = sorted(days)
for day in days:
self.getDBData(stock_code, day, result)
data = pd.DataFrame(result)
df_final_time = pd.DatetimeIndex(result['ymd'])
data.index = df_final_time
return data