CODEWITHSUNDEEP
pythonpandasnumpyindexing
nilsinelabore
nilsinelabore

Reputation: 5115

How to get integer location for index using pandas.Index.get_loc

The docs of pandas.Index.get_loc states it is used to "Get integer location".

Index.get_loc(self, key, method=None, tolerance=None)[source]

Get integer location, slice or boolean mask for requested label.

I could apply it to a normal column in a dataframe but am not sure how to do it on an index column of dataframe df. Would it be correct to change it to time_index=df2.columns.get_loc(T_index.index)? Thanks

Edit: Let's say perviously for dataframe df, Timestamp is a normal column, and I used time_index=df2.columns.get_loc("Timestamp") to get the integer location. If Timestamp is converted into index format, how should the code change accordingly?

       Timestamp    Temperature 
2020-02-06 08:23:04 18.5    0.0
2020-02-06 08:23:05 18.5    0.0
2020-02-06 08:23:06 18.5    0.0
2020-02-06 08:23:07 18.5    0.0
2020-02-06 08:23:08 18.5    0.0
...

Thanks

Upvotes: 1

Views: 1608

Answers (1)

Scott Boston
Scott Boston

Reputation: 153460

Use df.columns.get_loc():

Given df, 

df = pd.DataFrame(np.random.randint(0,100,(5,5)), index=[*'abcde'], columns=[*'VWXYZ'])

|    |   V |   W |   X |   Y |   Z |
|:---|----:|----:|----:|----:|----:|
| a  |  27 |  50 |  12 |  96 |  47 |
| b  |  63 |  93 |  27 |  37 |   7 |
| c  |  77 |  29 |  57 |  85 |  65 |
| d  |  84 |   6 |  98 |  32 |  85 |
| e  |  38 |   8 |  69 |  66 |  15 |

df.columns.get_loc('X')

Output:

2

The integer location of the 'X' column is 2.

Note:

type(df.index)
pandas.core.indexes.base.Index

and 

type(df.columns)
pandas.core.indexes.base.Index

Both the index and column headers of a dataframe are pd.Index.

Upvotes: 3

Related Questions