Question

我要使用此Surface Distance Based Metrics 代码是

import surface_distance as surfdist

surface_distances = surfdist.compute_surface_distances(
    mask_gt, mask_pred, spacing_mm=(1.0, 1.0, 1.0))
hd_dist_95 = surfdist.compute_robust_hausdorff(surface_distances, 95)

如何输入mask_gt和mask_pred？我试图以这种方式做到这一点，

import surface_distance as surfdist

def hausdorff_distance(y_true, y_pred):

  surface_distances = surfdist.compute_surface_distances(
          y_true, y_pred, spacing_mm=(1.0, 1.0, 1.0))
  hd_dist_95 = surfdist.compute_robust_hausdorff(surface_distances, 95)
  return hd_dist_95

但返回

TypeError: Input 'x' of 'LogicalOr' Op has type float32 that does not match expected type of bool.

Answer 1

听起来像您的损失函数希望获得布尔张量，而您的gt张量是“ float”类型（可能是一些值为0,1？的数组）。尝试将张量转换为布尔值，例如：

tf.keras.backend.cast(y_true, 'bool')

如何在Keras中输入遮罩？

1 个答案: