jax-ml/jax · error · RuntimeError

incompatible numbers of samples and fweights

Error message

incompatible numbers of samples and fweights

What it means

When frequency weights are given to jnp.cov, their length must equal the number of observations (X.shape[1]). If np.shape(fweights)[0] != X.shape[1], RuntimeError('incompatible numbers of samples and fweights') is raised.

Source

Thrown at jax/_src/numpy/lax_numpy.py:9214

  if y is not None:
    y_arr = atleast_2d(y)
    if not rowvar and y_arr.shape[0] != 1:
      y_arr = y_arr.T
    X = concatenate((X, y_arr), axis=0)
  if X.shape[1] == 0:
    cov_shape = () if X.shape[0] == 1 else (X.shape[0], X.shape[0])
    return array_creation.full(cov_shape, np.nan, dtype=X.dtype)

  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

View on GitHub (pinned to 1e1c6a8fc0)

Solutions

1. Resize weights to match the observation count: fweights = fweights[:m.shape[1]]
2. Transpose m or set rowvar=False if the orientation is flipped
3. Verify fweights.shape[0] == m.shape[-1] before calling

Example fix

// before
jnp.cov(m, fweights=w)  # len(w) != m.shape[1]
// after
assert w.shape[0] == m.shape[-1]
jnp.cov(m, fweights=w)

n_obs = m.shape[-1]
assert jnp.shape(fweights)[0] == n_obs, f'fweights len {jnp.shape(fweights)[0]} != {n_obs} samples'
jnp.cov(m, fweights=fweights)

When it happens

Trigger: jnp.cov(m, fweights=np.arange(5)) when m has 3 observations (m.shape[1] == 3), or weights sized to the number of variables instead of samples.

Common situations: Confusing variables x observations orientation (rowvar semantics); computing weights over a filtered subset of the data while passing the full m.

Related errors

• incompatible numbers of samples and aweights
• cannot handle multidimensional fweights
• fweights must be integer.
• cannot handle multidimensional aweights
• expected w and y to have the same length

AI-assisted analysis of jax-ml/jax@1e1c6a8fc0 (2026-08-27). Data as JSON: /api/errors/7376920f509c8b5c. Report an issue: GitHub.