python_trader/calculate_renko.py at master · JackWBoynton/python_trader · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
'''
Python3
Main running script -- for pyrenko live brick calculation and indicator calculation
Jack Boynton 2019
'''
import pandas as pd
import pyrenko
import helper as helper
import sys
import datetime
import argparse
import matplotlib.pyplot as plt
import time
import calendar
from tqdm import tqdm
import matplotlib
from blocks import get_data

parser = argparse.ArgumentParser()
parser.add_argument("fast", nargs=3, type=int)
parser.add_argument("-t", "--trade", type=int)
parser.add_argument('-r', '--tr', type=str)
parser.add_argument('-b', '--brick_size', type=int)
parser.add_argument('-d', '--days', type=int)
parser.add_argument('-o', '--order_type', type=str)
parser.add_argument("-p",  "--plot", type=bool)
args = parser.parse_args()
if args.order_type and args.order_type == 'Market' or args.order_type == 'Limit':
    pass
else:
    print('must set order_type to market or limit')
    sys.exit(0)
args.trade = bool(args.trade)
print(args.trade)
print('fast ma length: {}'.format(args.fast[0]), 'slow ma length: {}'.format(
    args.fast[1]), 'signal length: {}'.format(args.fast[2]), 'ord_type: ' + str(args.order_type))
time = datetime.date.today() - datetime.timedelta(days=1) # cant get todays data until tomorrow
sta = []
for i in range(args.days):  # gets all date csv files in home directory
    sta.append('../' + datetime.datetime.strftime(time -
                                                  datetime.timedelta(days=i), "%Y%m%d") + '.csv')

print('starting to load csv backtest data... days: ' + str(args.days))

#data = pd.DataFrame(helper.load_dfs_mult('XBTUSD', files=sta, location='../'))  # uses multiprocessing to parse huge csv datafiles
data = get_data(sta) # get_data is a generator that returns a pandas dataframe for 5 day chunks of data

print('finished loading csv backtest data... starting renko brick calculation')
renko_obj = pyrenko.renko(plot=False, j_backtest=True, fast=int(args.fast[0]), slow=int(
    args.fast[1]), signal_l=int(args.fast[2]), to_trade=args.trade, strategy=args.tr, ordtype=args.order_type)
renko_obj.set_brick_size(brick_size=args.brick_size, auto=False)  # sets brick_size hyperparam in dollars
while True:
    try:
        renko_obj.build_history(prices=next(data), timestamps=[''])  # builds renko backtest
    except Exception as e:
        print(e)
        break

trades = renko_obj.plot_renko()  # starts live renko brick calculation
# 2019-12-17D23:09:17.575367000 # sample timestamp from bitmex
if args.plot:
    macd_ = renko_obj.macd()
    fast, slow = renko_obj.ma()
    times = renko_obj.act_timestamps
    wma, times_wma = renko_obj.wma(9)
    p = "%Y-%m-%d%H:%M:%S.%f000"
    timestampss = []
    """
    plt.figure(figsize=(20,20))
    for n, point in tqdm(enumerate(data[1]),total=len(data[1])): ## somehow improve parsing time???
        point = point.replace("D","")
        st = (datetime.datetime.strptime(point,p))
        timestampss.append(calendar.timegm(st.timetuple()))
    macd_timestamps = []
    for m, pt in tqdm(enumerate(times[-len(macd_):]),total=len(times[-len(macd_):])):
        pt = pt.replace("D","")
        ac = (datetime.datetime.strptime(pt,p))
        macd_timestamps.append(calendar.timegm(ac.timetuple()))
    timestamps_wma = []
    for a, b in tqdm(enumerate(times_wma),total=len(times_wma)):
        b = b.replace("D","")
        ad = (datetime.datetime.strptime(b,p))
        timestamps_wma.append(calendar.timegm(ad.timetuple()))
    #plt.plot(macd_timestamps, macd_)
    plt.plot(timestampss, data[2])
    plt.plot(macd_timestamps, slow)
    plt.plot(macd_timestamps, fast)
    plt.plot(timestamps_wma, wma, c="#00ff00")
    for i in trades:
        if i[0] == 1:
            point = i[1].replace("D","")
            timestamp = (datetime.datetime.strptime(point,p))
            plt.scatter(calendar.timegm(timestamp.timetuple()),[i[2]], c="#00ff00")
        else:
            point = i[1].replace("D","")
            timestamp = (datetime.datetime.strptime(point,p))
            plt.scatter(calendar.timegm(timestamp.timetuple()),[i[2]], c="#ff0000")

    plt.savefig("out.jpg")
    plt.close()
    """
    rsi = []
    for i in renko_obj.rsi():
        rsi.append(i[0])
    #print(rsi)
    plt.plot(list(range(len(rsi))),rsi)
    plt.show()