Cannot use Pandas pct_change with date

Question

I have a data frame:

                 date  value
0 2017-11-30 13:58:57    901
1 2017-11-30 13:59:41    905
2 2017-11-30 13:59:41    925

That was generated by:

import pandas as pd
df = pd.DataFrame.from_items( [('date', ['2017-11-30 13:58:57', '2017-11-30 13:59:41', '2017-11-30 13:59:41']),("value", [901, 905, 925])])
df['date'] =  pd.to_datetime(df['date'])

I want to calculate the percentage change between two consecutive rows, but when I use:

df.pct_change()

I get the error:

ufunc true_divide cannot use operands with types dtype('<M8[ns]') and dtype('<M8[ns]')

How do I make it ignore the date column?

Answer 1

How do I make it ignore the date column?

Here's a solution with select_dtypes that should generalise to any dataframe by ignoring non-numeric columns -

df.select_dtypes(include=['number']).pct_change()

      value
0       NaN
1  0.004440
2  0.022099

Answer 2

我会尝试指定值列。

df[‘pctcng’]=df[‘value’].pct_change()