Fix introduced bug

      inputs, _, _ = self._prepare_batch((inputs, None, None))

      # Invoke the model.

      if len(inputs) == 1:

        inputs = inputs[0]

      if outputs is not None:

          self._output_functions[key] = tf.keras.backend.function(

              self.model.inputs, outputs)

        output_values = self._output_functions[key](inputs)

        output_values = self._compute_model(inputs)

      else:

        output_values = self._compute_model(inputs)

        if isinstance(output_values, tf.Tensor):

        output_values = [t.numpy() for t in output_values]

      # Apply tranformers and record results.

      if uncertainty:

        var = [output_values[i] for i in self._variance_outputs]

        if variances is None:

        results[i].append(t)

    # Concatenate arrays to create the final results.

    final_results = []

    final_variances = []

    final_embeddings = []

    assert scores[metric.name] > 0.9

    # Check that predicting internal layers works.

    pred_logits = np.squeeze(model.predict_on_batch(X, outputs=logits))

    pred_from_logits = 1.0 / (1.0 + np.exp(-pred_logits))

    assert np.allclose(prediction, pred_from_logits, atol=1e-4)