keras-team/keras · error · TypeError

Invalid input type. Expected `float32` or `float64`. Receive

Error message

Invalid input type. Expected `float32` or `float64`. Received: input type={x.dtype}

What it means

Error "Invalid input type. Expected `float32` or `float64`. Received: input type={x.dtype}" thrown in keras-team/keras.

Source

Thrown at keras/src/backend/numpy/ops/math.py:201

    return (
        np.real(complex_output).astype(x.dtype),
        np.imag(complex_output).astype(x.dtype),
    )


def irfft(x, fft_length=None):
    complex_input = _get_complex_tensor_from_tuple(x)
    # numpy always outputs float64, so we need to recast the dtype
    return np.fft.irfft(
        complex_input, n=fft_length, axis=-1, norm="backward"
    ).astype(x[0].dtype)


def stft(
    x, sequence_length, sequence_stride, fft_length, window="hann", center=True
):
    if standardize_dtype(x.dtype) not in {"float32", "float64"}:
        raise TypeError(
            "Invalid input type. Expected `float32` or `float64`. "
            f"Received: input type={x.dtype}"
        )
    if fft_length < sequence_length:
        raise ValueError(
            "`fft_length` must equal or larger than `sequence_length`. "
            f"Received: sequence_length={sequence_length}, "
            f"fft_length={fft_length}"
        )
    if isinstance(window, str):
        if window not in {"hann", "hamming"}:
            raise ValueError(
                "If a string is passed to `window`, it must be one of "
                f'`"hann"`, `"hamming"`. Received: window={window}'
            )
    x = convert_to_tensor(x)
    ori_dtype = x.dtype

View on GitHub (pinned to 7a34a03db6)

When it happens

Trigger: Thrown at keras/src/backend/numpy/ops/math.py:201 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/c072b2492b56db4e. Report an issue: GitHub.