Frequency encoding replaces each categorical value with its relative frequency (proportion) in the dataset. A category appearing 30 times out of 100 entries is encoded as 0.3. This creates a numerical feature that captures how common each category is, often useful because rare categories behave differently from common ones.
Given a list of categorical values, replace each value with the proportion of times it appears in the list.
Count the occurrences of each unique value, then divide each count by the total number of values to get the proportion.
Input:
values = ["a", "b", "a", "c", "a"]
Output:
[0.6, 0.2, 0.6, 0.2, 0.6]
"a" appears 3/5 = 0.6, "b" appears 1/5 = 0.2, "c" appears 1/5 = 0.2. Each position gets its category's frequency.
Input:
values = ["cat", "dog", "cat", "cat", "dog"]
Output:
[0.6, 0.4, 0.6, 0.6, 0.4]
"cat" has frequency 3/5 = 0.6 and "dog" has frequency 2/5 = 0.4.
First count occurrences using a dictionary: for each value, increment its count. Then divide each count by len(values) to get proportions.
Build the counts dict, then return [counts[v] / len(values) for v in values]. Each position gets its category's proportion.
Sign in to take notes on this problem
Accepts: array
Frequency encoding replaces each categorical value with its relative frequency (proportion) in the dataset. A category appearing 30 times out of 100 entries is encoded as 0.3. This creates a numerical feature that captures how common each category is, often useful because rare categories behave differently from common ones.
Given a list of categorical values, replace each value with the proportion of times it appears in the list.
Count the occurrences of each unique value, then divide each count by the total number of values to get the proportion.
Input:
values = ["a", "b", "a", "c", "a"]
Output:
[0.6, 0.2, 0.6, 0.2, 0.6]
"a" appears 3/5 = 0.6, "b" appears 1/5 = 0.2, "c" appears 1/5 = 0.2. Each position gets its category's frequency.
Input:
values = ["cat", "dog", "cat", "cat", "dog"]
Output:
[0.6, 0.4, 0.6, 0.6, 0.4]
"cat" has frequency 3/5 = 0.6 and "dog" has frequency 2/5 = 0.4.
First count occurrences using a dictionary: for each value, increment its count. Then divide each count by len(values) to get proportions.
Build the counts dict, then return [counts[v] / len(values) for v in values]. Each position gets its category's proportion.
Sign in to take notes on this problem
Accepts: array