Trying to fix the "Truth value of a Series is Ambiguous" Error on Pandas / Python

Question

I am getting the following error:

Traceback (most recent call last):
  File "tests.py", line 101, in <module>
    test()
  File "tests.py", line 95, in test
    print(connorsRSI(df))
  File "/Users/WorkFLOW/app.py", line 14, in connorsRSI
    if data.loc[index, 'Diff'] > 0:
  File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/pandas/core/generic.py", line 1478, in __nonzero__
    .format(self.__class__.__name__))
ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

On this line if data.loc[index, 'Diff'] > 0:

And don't know how to fix it.

Anybody knows any other way to rewrite it?

I know there are other questions like this one but none has helped me solve this error.

-- UPDATE: df.head() ==>

                 High        Low       Open      Close   Volume  Adj Close
Date                                                                      
2019-11-04  98.529999  95.599998  96.349998  98.150002  3841000  98.150002
2019-11-05  98.989998  96.629997  98.129997  98.260002  2719900  98.260002
2019-11-06  99.180000  97.970001  98.680000  98.669998  2447900  98.669998
2019-11-07  99.290001  97.019997  99.290001  97.230003  2156400  97.230003
2019-11-08  98.500000  96.320000  96.910004  98.269997  2306600  98.269997

Full code here:

import pandas_datareader.data as web

def connorsRSI(df):
    data = df.copy()
    data = data[['Close']]  
    data['Close'] = round(data['Close'], 2) 
    data['Diff'] = data['Close'].diff()
    data['UpDown'] = 0

    for i, index in enumerate(data.index[1:]):
        if data.loc[index, 'Diff'] > 0:
            if data.loc[data.index[i], 'Diff'] > 0:
                data.loc[index, 'UpDown'] = data.loc[data.index[i], 'UpDown'] + 1
            else:
                data.loc[index, 'UpDown'] = 1
        else:
            if data.loc[data.index[i], 'Diff'] > 0:
                data.loc[index, 'UpDown'] = -1
            else:
                data.loc[index, 'UpDown'] = data.loc[data.index[i], 'UpDown'] - 1

    xpct = data['Diff'][-1]
    data2 = data.copy()
    data2=data2[-100:]
    counter = data2[data2['Diff'] < xpct].count()['Diff']
    percentrank = round((counter/(len(data2)))*100,2)
    rsi = round(talib.RSI(data['Close'], timeperiod=3)[-1],2)
    streak = round(talib.RSI(data['UpDown'], timeperiod=2)[-1],2)
    crsi = round((rsi+streak+percentrank)/3,2)

    return crsi

def test():
    df = web.DataReader('EA', 'yahoo', start, end)
    print(connorsRSI(df))
test()

Based on your code, there's a high chance that you actually don't need to loop at all, but its impossible to say without a sample df. Have you thought of giving that a try? Also if you absolutely have to loop you read this [thread](https://stackoverflow.com/questions/16476924/how-to-iterate-over-rows-in-a-dataframe-in-pandas) about how to loop over rows. — Ben Pap, Feb 11 '20 at 23:45
Can you do a df.head() and show us what the dataframe contains ? You can just show 3-4 columns that you are using for ease. — Spandan Brahmbhatt, Feb 11 '20 at 23:50
@BenPap I just added df.head() , I'm interested in optimizing the code, is there a way to do this without looping? — edd, Feb 12 '20 at 00:05
Does this answer your question? [Truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all()](https://stackoverflow.com/questions/36921951/truth-value-of-a-series-is-ambiguous-use-a-empty-a-bool-a-item-a-any-o) — AMC, Feb 12 '20 at 02:18
_I'm interested in optimizing the code, is there a way to do this without looping?_ There probably is, yes. You need to rethink the entire function. — AMC, Feb 12 '20 at 02:20
@AMC I just shared the entire error, and no that didn't answer the question I already had looked into that — edd, Feb 12 '20 at 11:07
Please confirm, all you are tyring to do in the loop is make `updown` 1 if `diff` is positive and -1 if `diff ` is negative? As you set every value to `updown` to 0 initially — Ben Pap, Feb 12 '20 at 23:19
@BenPap that is correct, but **ALSO** it will add `+1` or substract `-1` if `updown` is positive or negative respectively... i.e... if there are two positive diffs in a row, then updown would be 2.... updown basically counts the streak of positive or negative diffs, if current updown is 5 and next diff is negative, then streak resets and starts at -1, here's a screenshot I took [link](http://prntscr.com/r1wjmd) `diff` is the difference between `closing prices` and `updown` would be the `Streak Duration` — edd, Feb 13 '20 at 17:18

score 1 · Accepted Answer · answered Feb 13 '20 at 19:57

data['UpDown'] = (data['Diff'] > 0)
data['UpDown'] = data.groupby((df1['UpDown'] != data['UpDown'].shift()).cumsum()).cumcount() +1
data.loc[data['Diff'] < 0, 'UpDown'] *= -1

These three lines replace your loop. The first makes a bool to check whether diff is positive or negative. The second line counts consecutive increases or decreases, and finally the third makes consecutive decreases negative.

Trying to fix the "Truth value of a Series is Ambiguous" Error on Pandas / Python

1 Answers1