jax-ml/jax · error · ValueError

method argument to `gamma` must be one of {'exact', 'approxi

Error message

method argument to `gamma` must be one of {'exact', 'approximate'}, got {method!r}

What it means

jax.random.gamma accepts only method='exact' (Marsaglia-Tsang rejection sampling, used under jit) or method='approximate' (a faster, less accurate transform). Any other string (including typos or None) is rejected with ValueError because there is no fallback sampling algorithm.

Source

Thrown at jax/_src/random/core.py:1686

      across devices in multi-device computation. Can be a
      :class:`~jax.sharding.NamedSharding`, a :class:`~jax.sharding.PartitionSpec`
      (``P``), or ``None`` (default). When specified, the output will be sharded
      according to the given sharding specification. Primarily used in explicit
      sharding mode.
      See the `explicit sharding tutorial <https://docs.jax.dev/en/latest/parallel.html>`_
      for more details.

  Returns:
    A random array with the specified dtype and with shape given by ``shape`` if
    ``shape`` is not None, or else by ``a.shape``.

  See Also:
    loggamma : sample gamma values in log-space, which can provide improved
      accuracy for small values of ``a``.
  """
  key, _ = _check_prng_key("gamma", key)
  if method not in {'exact', 'approximate'}:
    raise ValueError("method argument to `gamma` must be one of "
                     f"{{'exact', 'approximate'}}, got {method!r}")
  dtype = dtypes.check_and_canonicalize_user_dtype(
      float if dtype is None else dtype)
  if not dtypes.issubdtype(dtype, np.floating):
    raise ValueError(f"dtype argument to `gamma` must be a float "
                     f"dtype, got {dtype}")
  if shape is not None:
    shape = core.canonicalize_shape(shape)
  out_sharding = canonicalize_sharding_for_samplers(out_sharding, "gamma", shape)
  if method == 'approximate':
    return maybe_auto_axes(_gamma_approx, out_sharding,
                           shape=shape, dtype=dtype)(key, a)
  return maybe_auto_axes(_gamma, out_sharding, shape=shape, dtype=dtype)(key, a)


def loggamma(key: ArrayLike,
             a: RealArray,
             shape: Shape | None = None,

View on GitHub (pinned to 1e1c6a8fc0)

Solutions

1. Use method='exact' (default) or method='approximate', matching exactly including case.
2. If the method comes from a config, validate it against {'exact','approximate'} at load time with a clear error listing valid values.
3. Check the jax.random.gamma docstring for your JAX version to confirm supported methods.

Example fix

// before
samples = jax.random.gamma(key, 2.0, method='fast')

// after
samples = jax.random.gamma(key, 2.0, method='approximate')

method = method or 'exact'
assert method in {'exact', 'approximate'}, f"invalid gamma method: {method!r}"

def is_valid_gamma_method(method: str) -> bool:
    return method in {'exact', 'approximate'}

When it happens

Trigger: Calling jax.random.gamma(key, a, method='fast'), method=None, method=' Exact' (case/whitespace typo), or any string outside {'exact','approximate'}.

Common situations: Typos in hyperparameter configs; passing a method intended for a different library (e.g. numpy Generator method names); a version change where a previously tolerated value was removed or the argument was newly added.

Related errors

• dtype argument to `gamma` must be a float dtype, got {dtype}
• method argument to `loggamma` must be one of {'exact', 'appr
• method argument to `poisson` must be one of {'exact', 'appro
• method argument to `chisquare` must be one of {'exact', 'app
• dirichlet requires alpha.ndim >= 1, got alpha.ndim == {}

AI-assisted analysis of jax-ml/jax@1e1c6a8fc0 (2026-08-27). Data as JSON: /api/errors/99e69c727a999a1b. Report an issue: GitHub.