Pandas How to use a column value as an index to another row

Question

I have the following line of code


df["high_int"] = df.Slope  * (df.index - df.max_idx) + df,loc['max_idx', 'High]

max_idx contains the indexes of the highest highs over a period eg: 15 or 30.

I have tried .loc, .iloc, .at, .iat .get, .shift(), as well as df['max_idx'].map(df['High'])

Most errors seem to be related to using a series rather than an int (in the case of .iloc) or similar. It seems to me that this should a trivial task. Am I missing something obvious?

Thanks in advance

Asked By: BTT

||

Source

Answer 1

Last part doesn’t really make sense, df.loc[index, columns] takes index filters, and column, or list of columns, not 2 columns. Another thing – assuming you wanted to write df[["max_id", "High"]] – it would also fail, since you cannot force 2 columns into one in this way.

Consider the below as example of what you can and cannot do:

>>> df =pd.DataFrame({"Slope": [1,3,2, -5, -23.3], "max_id": [1,1,1,2,2], "High": [3,4,4,4,3]})
>>> df["high_int"] = df.Slope * (df.index - df.max_id)
>>> df
   Slope  max_id  High  high_int
0    1.0       1     3      -1.0
1    3.0       1     4       0.0
2    2.0       1     4       2.0
3   -5.0       2     4      -5.0
4  -23.3       2     3     -46.6

>>> df["high_int"] = df.Slope * (df.index - df.max_id) + df[["max_id", "High"]]
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/9skibi2/miniconda3/envs/airflow2/lib/python3.9/site-packages/pandas/core/frame.py", line 3967, in __setitem__
    self._set_item_frame_value(key, value)
  File "/home/9skibi2/miniconda3/envs/airflow2/lib/python3.9/site-packages/pandas/core/frame.py", line 4097, in _set_item_frame_value
    raise ValueError("Columns must be same length as key")
ValueError: Columns must be same length as key

Answered By: Grzegorz Skibinski

Pandas How to use a column value as an index to another row

Question:

Answers: