# median_absolute_error

### cuml.metrics.median_absolute_error(y_true, y_pred, , sample_weight=None, multioutput='uniform_average')[[source]](https://github.com/NVIDIA/cuml/blob/93418baf5/python/cuml/cuml/metrics/regression.py#L326)

Median absolute error regression loss.

Median absolute error output is non-negative floating point. The best value
is 0.0.

* **Parameters:**
  **y_true**
  : or (n_samples, n_outputs)
    Ground truth (correct) target values.

  **y_pred**
  : or (n_samples, n_outputs)
    Estimated target values.

  **sample_weight**
  : Sample weights.

  **multioutput**
  : or array-like of shape (n_outputs)
    Defines aggregating of multiple output values.
    Array-like value defines weights used to average errors.
    ‘raw_values’ :
    Returns a full set of errors in case of multioutput input.
    ‘uniform_average’ :
    Errors of all outputs are averaged with uniform weight.
* **Returns:**
  **loss**
  : If multioutput is ‘raw_values’, then median absolute error is returned
    for each output separately.
    If multioutput is ‘uniform_average’ or an ndarray of weights, then the
    weighted average of all output errors is returned.

<!-- !! processed by numpydoc !! -->