Python pandas Compare values of the last two rows and perform action-CodePudding

I have a pandas series as shown

I want to compare the value of the last two rows "Total" and "SMA20" to perform an action. (Please note the Pandas dataframe is large, Im pasting only last row)

myData = {'Date':[2022-01-17 15:27:00 05:30], 'total': [200],
      '20SMA': [220]
      }

I tried to store them in two separate variables by doing below

last_sma = myData['SMA20'].iloc[-1:]
last_total = myData['Total'].iloc[-1:]

Then try comparing both of the them,

if last_sma > last_total:
print('SMA is large')

I get the below error.

The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

When I try see the variable content its shown as below

Date 2022-01-17 15:27:00 05:30 215.155 Name: SMA20, dtype: float64

Its pandas series

CodePudding user response：

Welcome on Stackoverflow!

What you are getting with the way you make your slices are pandas Series

last_sma = myData['SMA20'].iloc[-1:]  # last_sma is a single value Series
last_total = myData['Total'].iloc[-1:]  # last_total is a single value Series

myData['SMA20'] returns a Series which holds the data from column SMA20, and the .iloc[-1:] returns a slice of this Series, with only one record, but a Series nonetheless.

Comparing 2 series with > operator returns another Series, holding the true of false value for comparison for each pair of records. In your, case it holds a Series with a single true or false value.

As stated by the error message you get, trying to collapse a Series of boolean values into a single boolean value is not well defined - even if your Series holds a single record.

If you only need to compare the last value, a simple way to make your code work is to remove the : in your slices.

last_sma = myData['SMA20'].iloc[-1]  # returns the single value at location -1
last_total = myData['Total'].iloc[-1] # returns the single value at location -1

if last_sma > last_total:
print('SMA is large')  # should work now!