jax-ml/jax · error · TypeError

Indexer must have integer or boolean type, got indexer with

Error message

Indexer must have integer or boolean type, got indexer with type {idx.dtype}

What it means

JAX's index-type classifier (IndexType.from_index) rejected an array (jax Array or numpy ndarray) used as an index because its dtype is neither integer nor boolean. JAX only supports advanced indexing with integer or boolean arrays; float or other dtypes are rejected before dispatch. The message reports the offending dtype.

Source

Thrown at jax/_src/numpy/indexing.py:86

  def from_index(cls, idx: Index) -> IndexType:
    """Create an IndexType enum from a supported JAX array index."""
    if idx is None:
      return cls.NONE
    elif idx is Ellipsis:
      return cls.ELLIPSIS
    elif isinstance(idx, slice):
      return cls.SLICE
    elif isinstance(idx, indexing.Slice):
      return cls.DYNAMIC_SLICE
    elif _is_integer_index(idx):
      return cls.INTEGER
    elif _is_boolean_index(idx):
      return cls.BOOLEAN
    elif isinstance(idx, (Array, np.ndarray)):
      if dtypes.issubdtype(idx.dtype, np.integer):
        return cls.ARRAY
      else:
        raise TypeError(
          f"Indexer must have integer or boolean type, got indexer with type {idx.dtype}")
    elif isinstance(idx, str):
      # TODO(jakevdp): this TypeError is for backward compatibility.
      # We should switch to IndexError for consistency.
      raise TypeError(f"JAX does not support string indexing; got {idx=}")
    elif isinstance(idx, Sequence):
      if not idx:  # empty indices default to float, so special-case this.
        return cls.ARRAY
      idx_aval = api.eval_shape(array_constructors.asarray, idx)
      if idx_aval.dtype == bool:
        return cls.BOOLEAN
      elif dtypes.issubdtype(idx_aval.dtype, np.integer):
        return cls.ARRAY
      else:
        raise TypeError(
          f"Indexer must have integer or boolean type, got indexer with type {idx_aval.dtype}")
    elif isinstance(idx, (float, complex, np.generic)):
      raise TypeError(

View on GitHub (pinned to 1e1c6a8fc0)

Solutions

1. Cast the index array to an integer type: x[idx.astype(jnp.int32)]
2. Ensure index-producing computations stay integral (use int dtype in arange/zeros/argwhere results)
3. If the values are truth values, use a boolean mask instead: x[idx > 0]

Example fix

// before
sel = jnp.array([0.5, 1.5])
y = x[sel]
// after
sel = jnp.array([0, 1])
y = x[sel]

def is_valid_index_dtype(a):
    return jnp.issubdtype(jnp.asarray(a).dtype, jnp.integer) or jnp.asarray(a).dtype == jnp.bool_

def is_int_index(idx) -> bool:
    a = jnp.asarray(idx)
    return jnp.issubdtype(a.dtype, jnp.integer)

When it happens

Trigger: Passing a float (or otherwise non-integer/boolean) Array/ndarray as an index, e.g. x[jnp.asarray([0.5, 1.5])] or x[np.array([1.0, 2.0])], inside x[...] on any jax array.

Common situations: Index arrays produced by arithmetic that upcasts to float (e.g. jnp.arange/2), indices loaded from float data, or metrics like argmax results converted through float ops. Common when porting NumPy code that happened to tolerate float indices.

Related errors

• Indexer must have integer or boolean type, got indexer with
• Indexer must have integer or boolean type, got indexer with
• np.delete(arr, obj): got obj.dtype={obj_array.dtype}; must b
• jnp.insert(): index array must be integer typed; got {obj}
• Cannot commute unswizzle and indexer with {aval}, which does

AI-assisted analysis of jax-ml/jax@1e1c6a8fc0 (2026-08-27). Data as JSON: /api/errors/45af7d1ad0494cf0. Report an issue: GitHub.