Create bot_test.py

rsibot/bot_test.py

@@ -0,0 +1,121 @@
+# Importing the libraries...
+
+import ccxt
+import pandas as pd
+pd.set_option('display.max_rows', None)
+from datetime import datetime
+from ta.trend import MACD
+from ta.momentum import RSIIndicator
+import warnings
+warnings.filterwarnings('ignore')
+import  schedule as schedule
+import time
+
+# Connection to the exchange for data retreival and order execution
+exchange = ccxt.binance()
+
+# For live implementation add your API keys:
+# exchange = ccxt.binance({"apiKey": '<YOUR_API_KEY>',"secret": '<YOUR_SECRET_KEY>' })
+
+# Function for data retreival and DataFrame processing
+def technical_signals(df):
+    '''
+    This function will implement the technical indicators to the market DataFrame.
+    The second part, will define the logics with a boolean final.
+    '''
+    # ---> Example of manual calculation of indicator
+
+    # Manual MACD
+    # ShortEMA = df["close"].ewm(span=12, adjust=False).mean() 
+    # LongEMA = df["close"].ewm(span=26, adjust=False).mean() 
+    # MACD = ShortEMA - LongEMA
+    # Signal = MACD.ewm(span=9, adjust=False).mean()
+    # df['MACD'] = MACD
+    # df['Signal'] = Signal
+
+    # Get Indicators using libary MACD & RSI for the example.
+    # MACD
+    indicator_macd = MACD(df['close'])
+    df['MACD'] = indicator_macd.macd()
+    df['Signal']= indicator_macd.macd_signal()
+    df['MACD Histogram']= indicator_macd.macd_diff()
+    df['MACD_Signal'] = False
+
+    # RSI
+    indicator_rsi = RSIIndicator(df['close'], window=14)
+    df['RSI_Signal'] = False
+    df['RSI'] = indicator_rsi.rsi()
+    # As many indicators, calculations, etc as required...
+
+    # Technical indicator strategy logics. We will use simple MACD to detect signals:
+    # Crossover MACD over Signal Below 0 == Buy signal
+    # Crossunder MACD under Signal Above 0 == Sell Signal
+    for current in range(1, len(df.index)):
+        previous = current - 1
+        if (df['MACD'][current] > df['Signal'][current]) and (df['MACD'][previous] <  df['Signal'][previous]) and (df['MACD'][current]<0):
+            df['MACD_Signal'][current] = True
+        elif (df['MACD'][current] < df['Signal'][current]) and (df['MACD'][previous] >  df['Signal'][previous]):
+            df['MACD_Signal'][current] = False
+        else:
+            df['MACD_Signal'][current] = df['MACD_Signal'][previous]
+    return df
+
+# Defines if we're in the market or not, to avoid submit orders once we've submited one on the same direction.
+in_position = False
+def reading_market(df):
+    '''
+    Function to analyze the market looking for signals according to our logics.
+    '''
+    global in_position
+
+    print("Looking for signals...")
+    print(df.tail(4))
+    last_row = len(df.index) - 1
+    previous_row = last_row - 1
+
+    if not df['MACD_Signal'][previous_row] and df['MACD_Signal'][last_row]:
+        print("Uptrend activated according MACD, BUY SIGNAL triggered")
+        if not in_position:
+            order_buy = 'Here goes BUY order' #exchange.create_market_buy_order('BTC/USDT', 1)
+            print(order_buy)
+            in_position = True
+        else:
+            print("Already in position, skip BUY signal")
+
+    if df['MACD_Signal'][previous_row] and not df['MACD_Signal'][last_row]:
+        if in_position:
+            print("Downtrend activated, SELL SIGNAL triggered")
+            order_sell = 'Here goes SELL order' # exchange.create_market_sell_order('BTC/USDT', 1)
+            print(order_sell)
+            in_position = False
+        else:
+            print("Not in position, skip SELL Signal")
+
+def execute_connection(symbol='BTC/USDT', timeframe='1m'):
+    '''
+    Function for data retreival, processing, and cleaning into dataframe to be used 
+    later for technical indicators implementation
+    '''
+    raw_data = exchange.fetch_ohlcv(symbol, timeframe, limit=100)
+    # since the return is on a list format. To see ---> print(raw_data)
+
+    # Converting data into a dataframe "[:-1]" since we will check the last closed bar
+    df = pd.DataFrame(raw_data[:-1], columns=['date', 'open', 'high', 'low', 'close', 'volume'])
+
+    # Converting date into a readable date format
+    df['date'] = pd.to_datetime(df['date'], unit='ms')
+    print(f"Executing connection and data processing at... {datetime.now().isoformat()}")
+    #print(df)
+    complete_df = technical_signals(df)
+    reading_market(complete_df)
+
+# Example of data
+#execute_connection(symbol='BTC/USDT', timeframe='15m')
+
+# Running and execution
+schedule.every(10).seconds.do(execute_connection)
+
+while True:
+    schedule.run_pending()
+    time.sleep(1)
+