TypeError: ("unsupported operand type(s) for -: 'decimal.Decimal' and 'float'", 'occurred at index growth(%)')

Question

这是我的数据框 --

    c2_name Q1_GMV      Q2_GMV     growth(%)
0   A       1170727260  221801763   -81
1   B       1604716749  829186592   -48
2   C       661473481   553698141   -16

我正在尝试使用 pandas 样式将 CSS 添加到数据帧输出。

# Set colormap equal to seaborns light green color palette
cm = sns.light_palette("green", as_cmap=True)

(df.style
  .background_gradient(cmap=cm, subset=['growth(%)'])
  .set_caption('This is a custom caption.')
  .set_table_styles(styles))

但是出现这个错误

TypeError: ("unsupported operand type(s) for -: 'decimal.Decimal' and 'float'", 'occurred at index growth(%)')

试着让它看起来像这样

here

Answer 1

您使用 Decimal 而不是浮动有什么特殊原因吗？这似乎是你问题的根源。在上面的示例中，鉴于该列中的值，完全没有必要。您可以通过以下方式解决您的问题：

df['growth(%)'] = df['growth(%)'].astype('float')

示例：

from decimal import Decimal
import seaborn as sns
cm = sns.light_palette("green", as_cmap=True)

print(df)
#  c2_name      Q1_GMV     Q2_GMV  growth(%)
#0       A  1170727260  221801763      -81.0
#1       B  1604716749  829186592      -48.0
#2       C   661473481  553698141      -16.0

df.dtypes
#c2_name      object
#Q1_GMV        int64
#Q2_GMV        int64
#growth(%)   float64

#Add a decimal type to the `df` to reproduce your error.
df.loc[2, 'growth(%)'] = Decimal(2.1511231)

# Now styling will throw an error:
(df.style
  .background_gradient(cmap=cm, subset=['growth(%)'])
  .set_caption('This is a custom caption.'))

TypeError: ("unsupported operand type(s) for -: 'decimal.Decimal' and 'float'", 'occurred at index growth(%)')

# Make the Decimals into floats
df['growth(%)'] = df['growth(%)'].astype('float')
(df.style
  .background_gradient(cmap=cm, subset=['growth(%)'])
  .set_caption('This is a custom caption.'))

如果您需要保留 Decimal 类型，请考虑编写一个仅转换类型以进行样式设置和显示的函数。

TypeError: ("unsupported operand type(s) for -: 'decimal.Decimal' and 'float'", 'occurred at index growth(%)')

TypeError: ("unsupported operand type(s) for -: 'decimal.Decimal' and 'float'", 'occurred at index growth(%)')

styling

pandas

jupyter-notebook

pandas-styles