jax-ml/jax · error · RuntimeError

cannot handle multidimensional aweights

Error message

cannot handle multidimensional aweights

What it means

jnp.cov uses aweights as per-observation scaling factors, which must be one weight per observation. It fires when the aweights array has more than one dimension and cannot be lined up with the samples axis.

Solutions

  1. Flatten: aweights=W.ravel()
  2. Check aweights.ndim == 1 and length equals observation count before calling
  3. Recompute weights as a 1D per-observation vector

Example fix

// before
jnp.cov(m, aweights=w_2d)
// after
jnp.cov(m, aweights=w_2d.ravel())
Defensive patterns

Strategy: validation

Validate before calling

if aweights is not None:
    aweights = jnp.asarray(aweights).ravel()
jnp.cov(m, aweights=aweights)

Prevention

When it happens

Trigger: jnp.cov(m, aweights=W) where W is 2D, e.g. a full weight matrix or a (n, 1) array from slicing with keepdims.

Common situations: Passing a covariance weighting matrix instead of per-observation weights; weights retaining extra axes after reductions.

Related errors


AI-assisted analysis of jax-ml/jax@1e1c6a8fc0 (2026-08-27). Data as JSON: /api/errors/8df6a760734f7f87. Report an issue: GitHub.

Appendix: source

Thrown at jax/_src/numpy/lax_numpy.py:9222

  if ddof is None:
    ddof = 1 if bias == 0 else 0

  w: Array | None = None
  if fweights is not None:
    fweights = util.ensure_arraylike("cov", fweights)
    if np.ndim(fweights) > 1:
      raise RuntimeError("cannot handle multidimensional fweights")
    if np.shape(fweights)[0] != X.shape[1]:
      raise RuntimeError("incompatible numbers of samples and fweights")
    if not issubdtype(fweights.dtype, np.integer):
      raise TypeError("fweights must be integer.")
    # Ensure positive fweights; note that numpy raises an error on negative fweights.
    w = abs(fweights)
  if aweights is not None:
    aweights = util.ensure_arraylike("cov", aweights)
    if np.ndim(aweights) > 1:
      raise RuntimeError("cannot handle multidimensional aweights")
    if np.shape(aweights)[0] != X.shape[1]:
      raise RuntimeError("incompatible numbers of samples and aweights")
    # Ensure positive aweights: note that numpy raises an error for negative aweights.
    aweights = abs(aweights)
    w = asarray(aweights if w is None else w * aweights)

  if dtype is not None:
    X = X.astype(dtype)
    w = w.astype(dtype) if w is not None else w

  avg, w_sum = reductions.average(X, axis=1, weights=w, returned=True)
  w_sum = w_sum[0]

  if w is None:
    f = X.shape[1] - ddof
  elif ddof == 0:
    f = w_sum
  elif aweights is None:

View on GitHub (pinned to 1e1c6a8fc0)