Segmenting a df based on condition shows nan values in jupyter

Question

Using df[columns][df[colums]<0.5] to segment a DataFrame in Pandas usually only showed me the rows matching this condition.

Just recently I end up getting NaN - Values shown:

Was there an update I'm missing or what is causing this behaviour usually?

Dropping all NaN with .dropna() is obviously a fast solution to this issue, but doesn't explain the change in "behaviour".

Fix: Don't put columns in double brackets..

Answer 1

如果需要按一列进行比较并按多列进行过滤，请使用带有boolean indexing DataFrame.loc ：

df.loc[df['district-heating'] < 0.5, ['id','district-heating']]