init

Bithumb_minute.py

@@ -10,7 +10,7 @@ from math import nan
 import plotly.io as po
 from plotly import subplots
 import plotly.graph_objects as go
-from datetime import datetime
+from datetime import datetime, timedelta
 from sklearn.linear_model import LinearRegression
 from sklearn.preprocessing import StandardScaler, MinMaxScaler
 
@@ -211,10 +211,10 @@ class Bithumb_minute(HTS):
 
         data = data.astype(
             {
-                'open': 'int',
+                'open': 'float',
-                'high': 'int',
+                'high': 'float',
-                'low': 'int',
+                'low': 'float',
-                'close': 'int',
+                'close': 'float',
                 'avg3': 'float',
                 'avg4': 'float',
                 'avg5': 'float',
@@ -234,8 +234,8 @@ class Bithumb_minute(HTS):
                 'disparity_avg20': 'float',
                 'disparity_avg60': 'float',
                 'disparity_avg120': 'float',
-                'buy': 'int',
+                'buy': 'float',
-                'sell': 'int',
+                'sell': 'float',
                 'slow_k': 'float',
                 'slow_d': 'float',
                 'macd': 'float',
@@ -459,142 +459,6 @@ class Bithumb_minute(HTS):
                     continue
 
             if i > 10:
-                """
-                # 만약 전일 저가와 오늘 종의 차이가 1만원이 넘으면 향후 60일은 분석하지 않는다.
-                if data1['high'][i] < int(data1['low'][i - 1] * 0.7):
-                    gap_state = True
-                    gap_interval -= 1
-                    continue
-                if gap_state:
-                    if gap_interval <= 0:
-                        gap_state = False
-                        gap_interval = 60
-                    else:
-                        gap_interval -= 1
-                        continue
-
-                if data1['disparity'][i] < 2:
-                    check = True
-                    for l in range(i - 3, i):
-                        if (
-                                data1['gradients_avg60'][l - 1] > data1['gradients_avg60'][l] or
-                                data1['gradients_avg20'][l - 1] > data1['gradients_avg20'][l] or
-                                data1['gradients_low'][l - 1] > data1['gradients_low'][l] or
-                                data1['disparity_avg5'][l - 1] > data1['disparity_avg5'][l] or
-                                data1['disparity'][l - 1] < data1['disparity'][l]
-                        ):
-                            check = False
-                            break
-                    if check and 99 < sum(data1['disparity_avg5'][i - 4:i + 1]) / 5 < 100 and 99 < sum(
-                            data1['disparity_avg60'][i - 4:i + 1]) / 5 < 100:
-                        buy = data1['low'][i]
-                        data1['buy'][i] = buy
-                        bsLine['buy'][i] = buy
-                        bsLine['buy_weight'][i] = 0.1
-
-                    check = True
-                    for l in range(i - 2, i):
-                        if (
-                                data1['gradients_avg60'][l - 1] > data1['gradients_avg60'][l] or
-                                data1['gradients_low'][l - 1] > data1['gradients_low'][l]
-                        ):
-                            check = False
-                            break
-                    if (
-                            check and
-                            -0.0011 < data1['gradients_low'][i] < 0 and -0.007 < data1['gradients_avg5'][i] < 0.001 and
-                            -0.0012 < data1['gradients_avg60'][i] < 0 and
-                            98.90 < data1['disparity_avg5'][i] < 101
-                    ):
-                        buy = data1['low'][i]
-                        data1['buy'][i] = buy
-                        bsLine['buy'][i] = buy
-                        bsLine['buy_weight'][i] = 0.1
-
-                check = True
-                for l in range(i - 6, i):
-                    if (
-                            data1['gradients_avg60'][l - 1] < data1['gradients_avg60'][l] or
-                            data1['gradients_avg20'][l - 1] < data1['gradients_avg20'][l] or
-                            data1['gradients_low'][l - 1] < data1['gradients_low'][l] or
-                            -0.039 < data1['gradients_low'][l - 1] < -0.35 or
-                            -0.05 < data1['gradients_avg20'][l - 1] < -0.30 or
-                            -0.40 < data1['gradients_avg60'][l - 1] < -0.30
-                    ):
-                        check = False
-                        break
-                if check and 99 < min(data1['disparity_avg5'][i - 6:i]) < max(data1['disparity_avg5'][i - 6:i]) < 101:
-                    buy = data1['low'][i]
-                    data1['buy'][i] = buy
-                    bsLine['buy'][i] = buy
-                    bsLine['buy_weight'][i] = 0.1
-
-                check = True
-                for l in range(i - 3, i):
-                    if (
-                            data1['gradients_low'][l - 1] < data1['gradients_low'][l] or
-                            data1['gradients_avg60'][l - 1] < data1['gradients_avg60'][l] or
-                            data1['gradients_avg20'][l - 1] < data1['gradients_avg20'][l] or
-                            0.01 < data1['gradients_low'][l - 1] < 0.21 or
-                            -0.09 < data1['gradients_avg20'][l - 1] < -0.002 or
-                            0.01 < data1['gradients_avg60'][l - 1] < 0.021
-                    ):
-                        check = False
-                        break
-                if check:
-                    buy = data1['low'][i]
-                    data1['buy'][i] = buy
-                    bsLine['buy'][i] = buy
-                    bsLine['buy_weight'][i] = 0.1
-
-                if (data1['disparity'][i] < 5 and 99.0 < data1['disparity_avg60'][i] < 99.1 and
-                        -0.009 < data1['gradients_avg60'][i] < -0.008 and 0.015 < data1['gradients_avg20'][i] < 0.016 and
-                        -0.006 < data1['gradients_avg5'][i] < -0.005 and -0.009 < data1['gradients_low'][i] < -0.008):
-                    check = True
-                    for l in range(i - 5, i):
-                        if (
-                                data1['gradients_avg60'][l - 1] > data1['gradients_avg60'][l] or
-                                data1['gradients_low'][l - 1] > data1['gradients_low'][l] or
-                                data1['disparity'][l - 1] < data1['disparity'][l]
-                        ):
-                            check = False
-                            break
-                    if check:
-                        buy = data1['low'][i]
-                        data1['buy'][i] = buy
-                        bsLine['buy'][i] = buy
-                        bsLine['buy_weight'][i] = 0.1
-
-                if data1['macd'][i] < -4000:
-                    if data1['macd'][i - 1] < data1['macd'][i]:
-                        if not self.notBuy(data1, i) and data1['slow_k'][i] < 30:
-                            buy = data1['low'][i]
-                            data1['buy'][i] = buy
-                            bsLine['buy'][i] = buy
-                            bsLine['buy_weight'][i] = 0.1
-
-                # macd 이전에 없던 바닥인 경우 상승할 찰나 매수
-                if data1['macds'][i - 1] < min(data1['macds'][:i - 1]):
-                    if data1['macds'][i - 1] < data1['macds'][i]:
-                        if not self.notBuy(data1, i) and data1['slow_k'][i] < 30:
-                            buy = data1['low'][i]
-                            data1['buy'][i] = buy
-                            bsLine['buy'][i] = buy
-                            bsLine['buy_weight'][i] = 0.1
-
-                if (
-                        98 < data1['disparity_avg5'][i] < 100 and data1['disparity_avg20'][i] < 93.5 and
-                        data1['disparity_avg60'][i] < 89 and
-                        -0.014 < data1['gradients_avg60'][i] < -0.013 and -0.03 < data1['gradients_avg20'][i] < -0.02 and -0.014 < data1['gradients_low'][i] < -0.013 and
-                        data1['slow_k'][i] < 11
-                ):
-                    if not self.notBuy(data1, i):
-                        buy = data1['low'][i]
-                        data1['buy'][i] = buy
-                        bsLine['buy'][i] = buy
-                        bsLine['buy_weight'][i] = 0.1
-
-                """
 
                 if data1['slow_k'][i] < 20 and data1['slow_k'][i - 1] < data1['slow_d'][i - 1] and data1['slow_d'][i] < data1['slow_k'][i]:
                     buy = data1['low'][i]
@@ -646,17 +510,24 @@ class Bithumb_minute(HTS):
         df.set_index('datetime', inplace=True)
         return df
 
-    def cancel_order(self, log_df, min=10):
+    def cancel_order(self, log_df, log_filename, min=5):
-        df = log_df[pd.DatetimeIndex(log_df.index).minute >= min]
+        now = datetime.now() - timedelta(minutes=min)
+        df = log_df.loc[(log_df.index <= now) ]
         df.reset_index()
 
         if df is not None:
-            for i in range(len(df)):
+             for i in range(len(df)):
                 order = (df['order0'][i], df['order1'][i], df['order2'][i], df['order3'][i])
                 cancel = self.bithumb.cancel_order(order)
 
-        log_df = log_df[pd.DatetimeIndex(log_df.index).minute < min]
+        log_df = log_df.loc[(log_df.index > now)]
-        return
+        if len(log_df) == 0:
+            log_df["datetime"] = ""
+        else:
+            log_df["datetime"] = log_df.index
+        log_df.to_csv(log_filename, index=False)
+
+        return log_df
 
     def getStock(self, ticker, analyzed_day, minute=5):
         stock = {"CODE": ticker, "NAME": ticker, "PRICE": []}
@@ -707,14 +578,15 @@ class Bithumb_minute(HTS):
             log_filename = os.path.join(RESOURCE_PATH, 'order', "bithumb"+"_"+today + '.log')
             if os.path.exists(log_filename):
                 log_df = pd.read_csv(log_filename)
-                log_df.columns = ["type", "datetime", "order0", "order1", "order2", "order3", "slow_k", "price", "count"]
+                log_df.columns = ["type", "order0", "order1", "order2", "order3", "slow_k", "price", "count", "datetime"]
+                log_df["datetime"] = pd.to_datetime(log_df["datetime"], format='%Y-%m-%d %H:%M:%S')
             else:
                 log_df = pd.DataFrame(columns=["type", "datetime", "order0", "order1", "order2", "order3", "slow_k", "price", "count"])
             log_df['datetime'] = pd.to_datetime(log_df['datetime'], unit='s')
             log_df.set_index('datetime', inplace=True)
 
             # 10분이 지난 미체결은 취소한다.
-            self.cancel_order(log_df, 10)
+            log_df = self.cancel_order(log_df, log_filename, 5)
             if isRealTime:
                 if max(bsLine['buy'][len(bsLine['buy']) - 2:]) > 100:
                     tmp = self.bithumb.get_balance(ticker)
@@ -724,10 +596,16 @@ class Bithumb_minute(HTS):
                     # order: ('bid', 'BTC', 'C0101000000322993432', 'KRW')
                     if len(stock1['close']) > 0:
                         print(ticker, "/", datetime.now().strftime('%Y-%m-%d %H:%M:%S'), "/", stock1['close'][len(stock1['close']) - 1], "/ BUY / ", stock1['slow_k'][len(stock1['slow_k']) - 1], "/", bsLine['buy'][len(bsLine['buy']) - 1], "/", count)
-                    value = {"type": "BUY", "datetime": datetime.now().strftime('%Y-%m-%d %H:%M:%S'), "order0": order[0], "order1": order[1], "order2": order[2], "order3": order[3], "slow_k": stock1['slow_k'][len(stock1['slow_k']) - 1], "price": bsLine['buy'][len(bsLine['buy']) - 1], "count": count}
+                    value = {"type": "BUY", "order0": order[0], "order1": order[1], "order2": order[2], "order3": order[3], "slow_k": stock1['slow_k'][len(stock1['slow_k']) - 1], "price": bsLine['buy'][len(bsLine['buy']) - 1], "count": count}
-                    log_df = log_df.append(value, ignore_index=True)
+                    datetime_value = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
-                    log_df['datetime'] = pd.to_datetime(log_df['datetime'], unit='s')
+                    value_df = pd.DataFrame(value, index=[datetime_value])
-                    log_df.set_index('datetime', inplace=True)
+
+                    indexes = log_df.index.tolist()
+                    indexes.append(datetime_value)
+
+                    log_df = log_df.append(value_df, ignore_index = True)
+                    log_df.index = indexes
+                    log_df['datetime'] = log_df.index
                     log_df.to_csv(log_filename, index=False)
 
                     dirName = os.path.join(RESOURCE_PATH, 'analysis', 'bithumb')
@@ -741,10 +619,16 @@ class Bithumb_minute(HTS):
                     order = self.bithumb.sell_limit_order(ticker, bsLine['sell'][len(bsLine['sell'])-1], count)
                     if len(stock1['close'])>0:
                         print(ticker, "/", datetime.now().strftime('%Y-%m-%d %H:%M:%S'), "/", stock1['close'][len(stock1['close']) - 1], "/ SELL / ", stock1['slow_k'][len(stock1['slow_k']) - 1], "/", bsLine['sell'][len(bsLine['sell']) - 1], "/", count)
-                    value = {"type": "SELL", "datetime": datetime.now().strftime('%Y-%m-%d %H:%M:%S'), "order0": order[0], "order1": order[1], "order2": order[2], "order3": order[3], "slow_k": stock1['slow_k'][len(stock1['slow_k']) - 1], "price": bsLine['buy'][len(bsLine['buy']) - 1], "count": count}
+                    value = {"type": "SELL", "order0": order[0], "order1": order[1], "order2": order[2], "order3": order[3], "slow_k": stock1['slow_k'][len(stock1['slow_k']) - 1], "price": bsLine['buy'][len(bsLine['buy']) - 1], "count": count}
-                    log_df = log_df.append(value, ignore_index=True)
+                    datetime_value = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
-                    log_df['datetime'] = pd.to_datetime(log_df['datetime'], unit='s')
+                    value_df = pd.DataFrame(value, index=[datetime_value])
-                    log_df.set_index('datetime', inplace=True)
+
+                    indexes = log_df.index.tolist()
+                    indexes.append(datetime_value)
+
+                    log_df = log_df.append(value_df, ignore_index=True)
+                    log_df.index = indexes
+                    log_df['datetime'] = log_df.index
                     log_df.to_csv(log_filename, index=False)
 
                     dirName = os.path.join(RESOURCE_PATH, 'analysis', 'bithumb')
@@ -780,7 +664,7 @@ if __name__ == "__main__":
                 #size = len(data_daily)
                 #if data_daily['slow_k'] < 30:
                 bithumb.buyRealTime(ticker, analyzed_day, isRealTime)
-            time.sleep(300)
+            time.sleep(30)
     else:
         for ticker in tickers:
             bithumb.buyRealTime(ticker, analyzed_day, isRealTime)