assign_unique_ids()

assign_unique_ids() is a helper function used to add an ID column to your dataset if it does not already have one, before creating a SequenceData object.

Function Usage

df_with_ids = assign_unique_ids(
   df,
   id_col_name="Entity ID"  # optional, default = "Entity ID"
)

Entry Parameters

Parameter	Required	Type	Description
df	✓	DataFrame	Input dataset. Each row represents an entity (e.g., a person, a case, etc.)
id_col_name	✗	str	Name of the new ID column. Default = "Entity ID".

What It Does

• Creates a new column at the first position of your DataFrame.
• Fills it with unique integer IDs, starting from 0 up to number of rows − 1.
• Raises an error if a column with the same name already exists (to avoid overwriting).
• Returns a copy of your DataFrame with the new ID column included.

Key Features

• Automatic IDs: no need to provide IDs yourself.
• Safe design: prevents overwriting if the column name is already present.
• Customizable: you can rename the ID column via id_col_name.
• Beginner-friendly: ensures your dataset always has a proper unique identifier, which is especially useful before creating SequenceData.

Examples

1. Add IDs to a dataset without identifiers

import pandas as pd

data = pd.DataFrame({
    "Y1": ["EDU", "EDU", "FT"],
    "Y2": ["EDU", "UNEMP", "FT"]
})

df_with_ids = assign_unique_ids(data)
print(df_with_ids)

Output:

   Entity ID    Y1     Y2
0          0   EDU    EDU
1          1   EDU  UNEMP
2          2    FT     FT

2. Use a custom column name

df_with_ids = assign_unique_ids(data, id_col_name="Person")
print(df_with_ids)

Output:

   Person    Y1     Y2
0       0   EDU    EDU
1       1   EDU  UNEMP
2       2    FT     FT

Authors

Code: Yuqi Liang

Documentation: Yuqi Liang

Edited by: Yuqi Liang