Implement **max pooling over a set of embedding vectors** to produce a single fixed-size representation, a common step in embeddings-and-retrieval pipelines. Given multiple embeddings (e.g., from tokens, passages, or chunks), you’ll return a pooled embedding by taking the maximum value per dimension.

The max-pooled embedding is defined as:
$$
p_j = \max_{i \in \{1,\dots,n\}} E_{i,j}
$$
where $E$ is an $n \times d$ matrix of embeddings and $p$ is a length-$d$ vector.

## Requirements

Implement the function
```python
import numpy as np

def max_pool_embeddings(embeddings):
    \"""
    Computes element-wise max pooling over a matrix of embedding vectors.

    Args:
        embeddings (np.ndarray): A matrix of n embeddings, shape (n, d).

    Returns:
        np.ndarray: A single pooled embedding of length d.
    \"""
    # Your solution below
```

**Rules:**
- Perform element-wise max across all input embeddings (pool over the first dimension).
- Return the pooled embedding as a NumPy array.
- Do not use any prebuilt pooling helpers beyond basic NumPy ops (e.g., no deep learning frameworks).
- Keep the implementation in a single Python function.

## Example

```python
embeddings = np.array([
    [0.1, -1.0,  2.5],
    [0.3, -2.0,  1.2],
    [0.2, -0.5,  2.0],
])

max_pool_embeddings(embeddings)
```

Output:
```python
array([0.3, -0.5, 2.5])
```

Longest Repeating Character Replacement

Remove Duplicates from Sorted Array

Max pooling of embeddings

Solve this easy level ML coding question from General. The premier LeetCode for ML platform to master Python, SQL, and Machine Learning algorithms.

1117. Max pooling of embeddings

Examples

Constraints

Similar Questions