Multinomial

Returns a tensor where each row contains num_samples indices sampled from the multinomial probability distribution.

Abstract Signature:

Multinomial(input: Tensor, num_samples: int, replacement: bool = False)

PyTorch

API: torch.multinomial
Strategy: Direct Mapping
Official Docs ↗

JAX (Core)

API: jax.random.categorical
Strategy: Plugin (inject_prng)

NumPy

API: numpy.random.mtrand.multinomial
Strategy: Direct Mapping
Official Docs ↗

Apple MLX

API: mlx.random.categorical
Strategy: Direct Mapping
Official Docs ↗

Flax NNX

API: nnx.rnglib.Rngs.multinomial
Strategy: Direct Mapping
Official Docs ↗