Pandas: Reset index for non unique index

Question

I have a dataframe like this:

	Name	Value
0	x1	0.20
0	x1	0.40
0	x1	0.34
3	x2	0.12
3	x2	0.13
4	x3	0.19
4	x3	0.23

df = pd.DataFrame([["x1", 0.2], ["x1", 0.4], ["x1", 0.34], ["x2", 0.12], ["x2", 0.13], ["x3", 0.19], ["x3", 0.239]],  index = [0, 0, 0, 3, 3, 4, 4], columns= ["Name", "Value"])

I would like to reset the index to get the following DataFrame:

	Name	Value
0	x1	0.20
0	x1	0.40
0	x1	0.34
1	x2	0.12
1	x2	0.13
2	x3	0.19
2	x3	0.23

df = pd.DataFrame([["x1", 0.2], ["x1", 0.4], ["x1", 0.34], ["x2", 0.12], ["x2", 0.13], ["x3", 0.19], ["x3", 0.239]],  index = [0, 0, 0, 1, 1, 2, 2], columns= ["Name", "Value"])

As this does not seem possible with the reset_index() function, what would be best way to achieve this reindexing?

The DataFrame results from using explode() and filtering NaN values. I want to reset the index after filtering to make it easier to loop over the exploded values later on. Notice that I don’t want to use the "Name" column as the index, as indexing by string seems to be extremly slow for larger data sets.

Asked By: Nils F.

||

Source

Answer 1

Use:

df = df.set_axis(pd.factorize(df['Name'])[0].tolist())

Or:

df.index = pd.factorize(df['Name'])[0]

  Name  Value
0   x1  0.200
0   x1  0.400
0   x1  0.340
1   x2  0.120
1   x2  0.130
2   x3  0.190
2   x3  0.239

Answered By: ansev

Pandas: Reset index for non unique index

Question:

Answers: