📘

Token Embedding Lookup

Implement a token embedding lookup table.

Given a vocabulary size V, embedding dimension D, and a sequence of token IDs, create a random embedding matrix (using numpy seed 42) and return the embeddings for each token.

Input:

Line 1: V D (vocab size, embedding dimension)
Line 2: space-separated token IDs

Output: The embedding matrix for the input tokens, one row per token, values rounded to 4 decimal places.

Example:

Input:

5 3
0 2 4

Output:

[[ 0.4967 -0.1383  0.6477]
 [ 0.5426 -0.4634 -0.4657]
 [-0.2349  0.2767 -0.3539]]

Reasoning:

We create a random embedding matrix of size $V \times D$ (5 x 3 in this case) using numpy with seed 42.
The input token IDs are used to index into this embedding matrix: tokens 0, 2, and 4 correspond to rows 0, 2, and 4 of the matrix.
Since token ID 4 is out of range for the given vocabulary size 5, we consider the actual indexing to be modulo V, so token ID 4 corresponds to row 4 % 5 = 4, which is the last row of the matrix.
The embeddings for the input tokens are retrieved from the embedding matrix and rounded to 4 decimal places, resulting in the output [[ 0.4967 -0.1383 0.6477], [ 0.5426 -0.4634 -0.4657], [-0.2349 0.2767 -0.3539]].

Constraints:

Use numpy with seed 42: np.random.seed(42); embeddings = np.random.randn(V, D)
1 <= V <= 100, 1 <= D <= 10
Token IDs are in range [0, V-1]
Round output to 4 decimal places

Editor

Python 3.13.1

Test Results

0/0

Run code to see test results.

📘