How to Batch Embed a List of Strings in Python

Batch embed a list of strings into deterministic pseudo-random vectors using a mock encoder class.

Easy Python 3.9+ Aug 9, 2026 AI & LLM integration patterns 13 views 0 copies

embedding batch-processing mock-encoder random ai-integration

Python code

25 lines

Python 3.9+

class MockEncoder:
    def __init__(self, dim=8, seed=42):
        self.dim = dim
        self.seed = seed

    def embed(self, text):
        # Deterministic pseudo-random embedding based on text content
        hash_val = hash(text)
        import random
        rng = random.Random(hash_val + self.seed)
        return [rng.random() for _ in range(self.dim)]


def batch_embed(texts, encoder):
    """Embed a list of strings, returning a list of embedding vectors."""
    return [encoder.embed(text) for text in texts]


if __name__ == "__main__":
    encoder = MockEncoder(dim=4, seed=42)
    texts = ["hello world", "python coding", "data science"]
    embeddings = batch_embed(texts, encoder)

    for text, emb in zip(texts, embeddings):
        print(f"{text!r}: {[round(x, 3) for x in emb]}")

Output

stdout

'hello world': [0.577, 0.165, 0.39, 0.061]
'python coding': [0.901, 0.713, 0.01, 0.412]
'data science': [0.091, 0.466, 0.941, 0.585]

How it works

The MockEncoder generates a deterministic pseudo-random embedding for each text by seeding a random.Random instance with a hash of the text plus a fixed seed. This ensures the same text always produces the same vector, which is crucial for reproducibility in testing. The batch_embed function simply applies the encoder's embed method to each string in the list, returning a list of vectors. This pattern mirrors real embedding APIs like OpenAI's text-embedding-3-small, where you often process multiple strings in one call for efficiency.

Common mistakes

Using Python's built-in `hash()` which is salted per process, leading to non-deterministic results across runs.
Forgetting that `random.Random` must be reseeded for each text to avoid sharing state.
Assuming the output is meaningful semantically; it's random, not a real semantic embedding.

Variations

Replace `MockEncoder` with a real embedding client like `openai.Embedding.create` for actual embeddings.
Use NumPy to generate vectors via `numpy.random.default_rng(hash_val).random(dim)` for faster vectorized generation.

Real-world use cases

Mocking an embedding service in unit tests to avoid network calls and ensure deterministic results.
Generating feature vectors for a small set of strings in a prototype before using a full-fledged embedding API.
Creating synthetic embeddings to benchmark or test downstream workflows like clustering or search.

How to Batch Embed a List of Strings in Python

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from AI & LLM integration patterns

Tutorials

Quizzes

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from AI & LLM integration patterns

Keep learning

Tutorials

Quizzes