Reputation: 143
I'm trying to write some code that will allow me quickly determine whether a variable is higher, lower, or equal to another variable within a dataframe at each index point.
As shown here:
import numpy as np
import datetime
import pandas as pd
import operator
previous_cross = 'both'
ops = {
'+' : operator.add,
'-' : operator.sub,
'*' : operator.mul,
'/' : operator.truediv, # use operator.div for Python 2
'%' : operator.mod,
'^' : operator.xor,
'<' : operator.lt,
'<=' : operator.le,
'==' : operator.eq,
'!=' : operator.ne,
'>=' : operator.ge,
'>' : operator.gt
}
#//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
df = pd.read_csv(r'C:\EURUSD_mt5_ticks.csv', parse_dates=['Date'])
df.columns = ['Date', 'Ask_Price', 'Ask_Volume', 'Bid_Price', 'Bid_Volume', 'Spread']
df['Date'] = pd.to_datetime(df['Date'].astype('datetime64[ns]'), exact=True, cache=True, format='%Y-%m-%d %H:%M:%S.%f')
MA1 = df['Ask_Price'].rolling(4).mean()
MA2 = df['Ask_Price'].rolling(21).mean()
df['MA1'] = MA1
df['MA2'] = MA2
#//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
def tbar_number(n):
return tbar_count - n
def vcross(variable1_column, direction, variable2_column, vdataframe, width):
global previous_cross
#number of bars previous from the current bar,
width_value = tbar_number(width)
for i in range(width_value+1):
width_value_count = i
#data from the index of each bar within the 'width' bar range
v1_width_data = df.loc[width_value_count, variable1_column]
v2_width_data = df.loc[width_value_count, variable2_column]
#if variable 1 crosses above or is equal to(depending on chosen symbol) variable 2, return true
if (direction == '>' or '>=') & (ops[direction](v1_width_data, v2_width_data)) & (previous_cross == 'short' or 'both'):
if (direction == '>=') and (v1_width_data == v2_width_data):
return True
if direction == '>':
previous_cross = 'long'
return True
#if variable 1 crosses below or is equal to(depending on chosen symbol) variable 2, return true
elif (direction == '<' or '<=') & (ops[direction](variable1, variable2)) & (previous_cross == 'long' or 'both'):
if (direction == "<=") and (v1_width_data == v2_width_data):
return True
if direction == '<':
previous_cross = 'short'
return True
#if variable 1 is equal to(depending on chosen symbol) variable 2, return true
elif (direction == '==') & (ops[direction](variable1, variable2)):
return True
else:
return False
#/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
for tbar_count in range(len(df)):
if vcross('MA1', '>', 'MA2', df, 1):
print("hi")
Date Ask_Price Ask_Volume Bid_Price Bid_Volume Spread MA1 MA2
0 2021-08-11 00:00:00.686 1.17199 0.75 1.17195 0.75 0.00004 NaN NaN
1 2021-08-11 00:00:00.989 1.17201 2.43 1.17195 3.37 0.00006 NaN NaN
2 2021-08-11 00:00:05.004 1.17198 0.75 1.17196 4.87 0.00002 NaN NaN
3 2021-08-11 00:00:05.203 1.17200 0.75 1.17197 0.10 0.00003 1.171995 NaN
4 2021-08-11 00:00:10.654 1.17201 0.75 1.17198 0.10 0.00003 1.172000 NaN
But I get this error
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-7-f2dbb6db13b8> in <module>
80 for tbar_count in range(len(df)):
81
---> 82 if vcross('MA1', '>', 'MA2', df, 1):
83 print("hi")
84
<ipython-input-7-f2dbb6db13b8> in vcross(variable1_column, direction, variable2_column, vdataframe, width)
55
56 #if variable 1 crosses above or is equal to(depending on chosen symbol) variable 2, return true
---> 57 if (direction == '>' or '>=') & (ops[direction](v1_width_data, v2_width_data)) & (previous_cross == 'short' or 'both'):
58 if (direction == '>=') and (v1_width_data == v2_width_data):
59 return True
TypeError: ufunc 'bitwise_and' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''
I haven't been able to find the solution through google. Most of the time the issue seems to be related to not encapsulating the conditions when using & but I have encapsulated everything already.
Can anyone lend me a hand?
Upvotes: 0
Views: 817
Reputation: 231475
Careful with the use of or
in your tests. For example:
In [351]: direction='++'; direction=='>' or '>='
Out[351]: '>='
In [352]: direction='>'; direction=='>' or '>='
Out[352]: True
This does a direction=='>'
; if that is True, it returns True; if False, it returns the value after the or
. It is not the same as
In [354]: direction='++';(direction=='>') or (direction=='>=')
Out[354]: False
In [355]: direction='>';(direction=='>') or (direction=='>=')
Out[355]: True
In [356]: direction='>=';(direction=='>') or (direction=='>=')
Out[356]: True
and equivalent test is
direction in ['>','>=']
Upvotes: 1