5 Best Ways to Assign IDs to Each Unique Value in a Python List

Rate this post

πŸ’‘ Problem Formulation: In many programming scenarios, it’s necessary to uniquely identify each distinct element in a list. Let’s say you have a list of values such as ['apple', 'banana', 'apple', 'orange'] and you want to translate this into a list of unique IDs such as [0, 1, 0, 2], where each unique value is represented by a different ID. This article explores five effective ways to accomplish this task in Python.

Method 1: Using Enumerate with Dictionary Comprehension

This method leverages the power of dictionary comprehension and the built-in function enumerate() to map each unique value to an identifier. This solution is both elegant and efficient, offering a good balance between readability and performance in most cases.

Here’s an example:

values = ['apple', 'banana', 'apple', 'orange']
unique_values = {value: id for id, value in enumerate(sorted(set(values)))}
ids = [unique_values[value] for value in values]

Output: [0, 1, 0, 2]

This code snippet starts by creating a set from the list to eliminate duplicates, sorts it, then uses a dictionary comprehension with the enumerate() function to assign a unique ID to each value. The list comprehension then creates the ID list, mapping each original value to its corresponding ID.

Method 2: Using Defaultdict

The collections.defaultdict class can be used to automatically handle new keys and assign a unique ID to them. This is useful when processing the list element by element in a single pass without having to sort or deduplicate the list first.

Here’s an example:

from collections import defaultdict

values = ['apple', 'banana', 'apple', 'orange']
unique_ids = defaultdict(lambda: len(unique_ids))
result = [unique_ids[value] for value in values]

Output: [0, 1, 0, 2]

In this snippet, we initialize defaultdict with a lambda function that returns the current length of the defaultdict, effectively giving each new key a new ID. We then iterate through the list to build the result.

Method 3: Using Pandas Factorize

For data-heavy applications, the pandas library provides a handy function called factorize(). This method is incredibly efficient when dealing with large datasets and integrates well into a data analysis workflow.

Here’s an example:

import pandas as pd

values = ['apple', 'banana', 'apple', 'orange']
ids = pd.factorize(values)[0]

Output: [0, 1, 0, 2]

The factorize() function in the above code maps each unique value to a consecutive integer starting from 0. It returns a tuple where the first element is the array of encoded labels (which we extract using [0]).

Method 4: Using a Counter

Python’s collections.Counter class can be used to not only count the number of occurrences of each element but also to establish unique IDs based on the occurrences.

Here’s an example:

from collections import Counter

values = ['apple', 'banana', 'apple', 'orange']
counter = Counter(values)
lookup = {key: id for id, key in enumerate(counter)}
ids = [lookup[value] for value in values]

Output: [0, 1, 0, 2]

This code creates a Counter object, which is essentially a dictionary where keys are list elements and values are their counts. We then enumerate this counter to assign unique IDs and finally, we generate the ID list.

Bonus One-Liner Method 5: Using Dictionary Setdefault

The setdefault() method of a dictionary can also be utilized to achieve our goal in a concise one-liner. While this method is very Pythonic, readability might suffer for those unfamiliar with setdefault().

Here’s an example:

values = ['apple', 'banana', 'apple', 'orange']
ids = [dict().setdefault(v, len(dict())) for v in values]

Output: [0, 0, 0, 0] (Note: This is incorrect, illustrating the issue with using two separate dict() calls within the comprehension.)

This attempt to create a one-liner uses setdefault() within a list comprehension; however, it incorrectly creates a new empty dictionary on each iteration, failing to preserve the state and assign correct IDs. This illustrates the importance of correctly managing state within one-liners.


  • Method 1: Enumerate with Dictionary Comprehension. Suitable for most situations. Efficient and readable.
  • Method 2: Defaultdict. Ideal for on-the-fly ID assignments. Less beginner-friendly due to default factory usage.
  • Method 3: Pandas Factorize. Best when working within the pandas ecosystem, especially for large data sets.
  • Method 4: Using a Counter. Good for cases where you may also need the count of each element. Slightly more verbose.
  • Bonus Method 5: Dictionary Setdefault One-Liner. Not recommended due to incorrect use in example, but could theoretically be done correctly with careful state management.