Remove header index in pandas Dataframe-CodePudding

I have the following dataframe :


                            0                        1         2    ... 630 631 632
0                         index                MATRICULE    ID_UEV  ...            
1       9936-25-3989-4-000-0000  9936-25-3989-4-000-0000  01045406  ...            
2       9739-83-9737-8-001-0302  9739-83-9737-8-001-0302  01038232  ...            
3       9754-37-9664-9-000-0000  9754-37-9664-9-000-0000  02004842  ...            
4       8134-96-8810-1-000-0000  8134-96-8810-1-000-0000  04007065  ...

How can I remove the first row/index so I can have index MATRICULE, ID_UEV as a header


0                         index                MATRICULE    ID_UEV  ...            
1       9936-25-3989-4-000-0000  9936-25-3989-4-000-0000  01045406  ...            
2       9739-83-9737-8-001-0302  9739-83-9737-8-001-0302  01038232  ...            
3       9754-37-9664-9-000-0000  9754-37-9664-9-000-0000  02004842  ...            
4       8134-96-8810-1-000-0000  8134-96-8810-1-000-0000  04007065  ...

CodePudding user response：

You can use

df.columns = df.loc[0]
df = df.drop(0)

This sets the columns to the items in the first row, then drops the first row.

CodePudding user response：

If you are reading this data using pd.read_csv or pd.read_excel APIs, then it has a skiprows argument which you can use to skip the line numbers.

import pandas as pd

df = pd.read_csv(
    r"your_path",
    skiprows=lambda x: x in [0], # Skip the first line
)
print(df)

CodePudding user response：

You can use it either way when reading the file:

Start header from the first index:

data = pd.read_csv("file.csv", header = 1)

Remove the first row

data = pd.read_csv("file.csv", skiprows=1)

CodePudding user response：

Alternative way is:

df.rename(columns=df.iloc[0]).drop(df.index[0])

or use this if don't want your index to be missing anything:

df.rename(columns=df.iloc[0]).drop(df.index[0]).reset_index(drop=True)