Problema
Cosine similarity compares the direction of two vectors and ignores their length. Semantic search uses it to find texts with similar embeddings:
cos(a, b) = (a · b) / (|a| × |b|), where |a| = √(a₁² + a₂² + … + aₙ²)
The result is 1 for the same direction, 0 for perpendicular vectors and -1 for opposite directions. If either vector is all zeros, return 0.
Implement cosine_similarity(a, b).
Input
a, b: Two lists of numbers with the same length.
Output
Return the cosine similarity as a decimal number.
Restrições
1 ≤ length ≤ 1,000
-1,000 ≤ each value ≤ 1,000
Exemplos
Entrada:
a = [1.0, 0.0]
b = [0.0, 1.0]
Saída esperada:
expected = 0.0
Entrada:
std::vector<double> a = {1.0, 0.0};
std::vector<double> b = {0.0, 1.0};
Saída esperada:
double expected = 0.0;
Entrada:
double[] a = new double[] {1.0, 0.0};
double[] b = new double[] {0.0, 1.0};
Saída esperada:
double expected = 0.0;
Explicação
The dot product is 0, so the vectors are perpendicular and the similarity is 0.
Entrada:
a = [1.0, 2.0, 3.0]
b = [2.0, 4.0, 6.0]
Saída esperada:
expected = 1.0
Entrada:
std::vector<double> a = {1.0, 2.0, 3.0};
std::vector<double> b = {2.0, 4.0, 6.0};
Saída esperada:
double expected = 1.0;
Entrada:
double[] a = new double[] {1.0, 2.0, 3.0};
double[] b = new double[] {2.0, 4.0, 6.0};
Saída esperada:
double expected = 1.0;
Explicação
b is a times 2: same direction, so the similarity is 1 even though the lengths differ.
Como abordar
Tente sozinho primeiro e depois abra uma dica de cada vez.
Dica 1
The score depends on direction, so first compute the dot product and both vector lengths separately.
Dica 2
If either vector has zero magnitude, the similarity is defined as zero instead of dividing by zero.
Dica 3
Use the same index across both vectors for every term, and remember the denominator is the product of magnitudes.
Complexidade esperada
O(n) time, O(1) extra space
Referências
Este desafio foi adaptado para o CodeUtility. Siga o enunciado e as restrições acima.