How to insert a new row to a Pandas Dataframe?

In this post, we will learn to insert/add a new row to an existing Pandas Dataframe using pandas.DataFrame.loc, pandas.concat() and numpy.insert(). Using these methods you can add multiple rows/lists to an existing or an empty Pandas DataFrame.

Creating a Pandas Dataframe from Pandas Series objects.

First of all, we will create a Pandas Dataframe from two Pandas Series objects as under :-

s1 = pd.Series([1, 2, 3])
s2 = pd.Series([4, 5, 6])
cols =  ["A", "B", "C"]

df = pd.DataFrame([list(s1), list(s2)],  columns =  cols)

   A  B  C
0  1  2  3
1  4  5  6

[2 rows x 3 columns]

Now assume that we need to append the following list as a new row to the Pandas Dataframe.

new_row = [7, 8, 9]

Inserting a new row to a Pandas Dataframe using .loc

We can insert a new row as the last row to a Pandas Dataframe using pandas.DataFrame.loc as shown in the following code:-

df.loc[-1] = new_row

   A  B  C
0  1  2  3
1  4  5  6
-1 7  8  9

[3 rows x 3 columns]

But, here the index of the last row is -1, you can reset the index of a Pandas DataFrame using df.reset_index():-

df.reset_index(drop=True)
   A  B  C
0  1  2  3
1  4  5  6
2  7  8  9

[3 rows x 3 columns]

Inserting a new row to a Pandas Dataframe using pandas.concat()

You can also add a new row to a Pandas Dataframe using pandas.concat() by converting the row into a Dataframe and then concatenating it to the existing Dataframe as under:-

new_df = pd.DataFrame([new_row], columns=cols)

   A  B  C
0  7  8  9

[1 row x 3 columns]

df = pd.concat([df, new_df])

   A  B  C
0  1  2  3
1  4  5  6
0  7  8  9

[3 rows x 3 columns]

Here, the new Pandas Dataframe has preserved the index of the row, but you can reset the index while concatenating by using the argument ‘ignore_index = True’ in pandas.concat().

df = pd.concat([df, new_df], ignore_index = True)

   A  B  C
0  1  2  3
1  4  5  6
2  7  8  9

[3 rows x 3 columns]

You can also add the new row to the top of the existing Dataframe by switching the order while using pandas.concat().

df = pd.concat([new_df, df], ignore_index = True)

   A  B  C
0  7  8  9
1  1  2  3
2  4  5  6

[3 rows x 3 columns]

Inserting a new row to a Pandas Dataframe using numpy.insert()

You can also insert a new row to an existing Pandas Dataframe using numpy.insert(). The major advantage of using numpy.insert() to insert a new row to the Pandas Dataframe is that you can insert the new row at an arbitrary or a desired position/index in the dataframe by declaring the desired index of the row in np.insert().

import numpy as np

df = pd.DataFrame(np.insert(df.values, 1, new_row, axis=0)) # 1 is the index at which we need to insert the new row.

   A  B  C
0  1  2  3
1  7  8  9
2  4  5  6

[3 rows x 3 columns]

Inserting a new row at a desired index to a Pandas Dataframe using pandas.concat()

Alternatively, you can also insert a new row at an arbitrary or desired index using pandas.concat() using slicing (df.iloc) as under:-

df = pd.concat

df = pd.concat([df.iloc[:1], new_df, df.iloc[1:]], ignore_index = True)

   A  B  C
0  1  2  3
1  7  8  9
2  4  5  6

[3 rows x 3 columns]

How to add a new row to a Pandas Dataframe? - Video Tutorial