Python backend.maximum方法代码示例

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def batch_iou(boxes, box):
  """Compute the Intersection-Over-Union of a batch of boxes with another
  box.

  Args:
    box1: 2D array of [cx, cy, width, height].
    box2: a single array of [cx, cy, width, height]
  Returns:
    ious: array of a float number in range [0, 1].
  """
  lr = np.maximum(
      np.minimum(boxes[:,0]+0.5*boxes[:,2], box[0]+0.5*box[2]) - \
      np.maximum(boxes[:,0]-0.5*boxes[:,2], box[0]-0.5*box[2]),
      0
  )
  tb = np.maximum(
      np.minimum(boxes[:,1]+0.5*boxes[:,3], box[1]+0.5*box[3]) - \
      np.maximum(boxes[:,1]-0.5*boxes[:,3], box[1]-0.5*box[3]),
      0
  )
  inter = lr*tb
  union = boxes[:,2]*boxes[:,3] + box[2]*box[3] - inter
  return inter/union 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def call(self, input_tensor, mask=None):
        z_s = input_tensor[0]
        z_n = input_tensor[1]
        r_s = input_tensor[2]

        z_s = K.l2_normalize(z_s, axis=-1)
        z_n = K.l2_normalize(z_n, axis=-1)
        r_s = K.l2_normalize(r_s, axis=-1)

        steps = z_n.shape[1]

        pos = K.sum(z_s * r_s, axis=-1, keepdims=True)
        pos = K.repeat_elements(pos, steps, axis=1)
        r_s = K.expand_dims(r_s, axis=-2)
        r_s = K.repeat_elements(r_s, steps, axis=1)
        neg = K.sum(z_n * r_s, axis=-1)

        loss = K.cast(K.sum(K.maximum(0., (1. - pos + neg)), axis=-1, keepdims=True), K.floatx())
        return loss 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def __init__(self, lr=1e-1, beta_1=0.9, beta_2=0.999,
                 epsilon=1e-8, decay=0., amsgrad=False, partial=1. / 8., **kwargs):
        if partial < 0 or partial > 0.5:
            raise ValueError(
                "Padam: 'partial' must be a positive float with a maximum "
                "value of `0.5`, since higher values will cause divergence "
                "during training."
            )
        super(Padam, self).__init__(**kwargs)
        with K.name_scope(self.__class__.__name__):
            self.iterations = K.variable(0, dtype='int64', name='iterations')
            self.lr = K.variable(lr, name='lr')
            self.beta_1 = K.variable(beta_1, name='beta_1')
            self.beta_2 = K.variable(beta_2, name='beta_2')
            self.decay = K.variable(decay, name='decay')
        if epsilon is None:
            epsilon = K.epsilon()
        self.epsilon = epsilon
        self.partial = partial
        self.initial_decay = decay
        self.amsgrad = amsgrad 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def ranking_loss_with_margin(y_pred, y_true):
    """
    Using this loss trains the model to give scores to all correct elements in y_true that are
    higher than all scores it gives to incorrect elements in y_true, plus a margin.

    For example, let ``y_true = [0, 0, 1, 1, 0]``, and let ``y_pred = [-1, 1, 2, 0, -2]``.  We will
    find the lowest score assigned to correct elements in ``y_true`` (``0`` in this case), and the
    highest score assigned to incorrect elements in ``y_true`` (``1`` in this case).  We will then
    compute a hinge loss given these values: ``K.maximum(0.0, 1 + 1 - 0)``.

    Note that the way we do this uses ``K.max()`` and ``K.min()`` over the elements in ``y_true``,
    which means that if you have a lot of values in here, you'll only get gradients backpropping
    through two of them (the ones on the margin).  This could be an inefficient use of your
    computation time.  Think carefully about the data that you're using with this loss function.

    Because of the way masking works with Keras loss functions, also, you need to be sure that any
    masked elements in ``y_pred`` have very negative values before they get passed into this loss
    function.
    """
    correct_elements = y_pred + (1.0 - y_true) * VERY_LARGE_NUMBER
    lowest_scoring_correct = K.min(correct_elements, axis=-1)
    incorrect_elements = y_pred + y_true * VERY_NEGATIVE_NUMBER
    highest_scoring_incorrect = K.max(incorrect_elements, axis=-1)
    return K.mean(K.maximum(0.0, 1.0 + highest_scoring_incorrect - lowest_scoring_correct)) 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def mix_gaussian_loss(x, mu, log_sig, w):
    '''
    Combine the mixture of gaussian distribution and the loss into a single function
    so that we can do the log sum exp trick for numerical stability...
    '''
    if K.backend() == "tensorflow":
        x.set_shape([None, 1])
    gauss = log_norm_pdf(K.repeat_elements(x=x, rep=mu.shape[1], axis=1), mu, log_sig)
    # TODO: get rid of clipping.
    gauss = K.clip(gauss, -40, 40)
    max_gauss = K.maximum((0.), K.max(gauss))
    # log sum exp trick...
    gauss = gauss - max_gauss
    out = K.sum(w * K.exp(gauss), axis=1)
    loss = K.mean(-K.log(out) + max_gauss)
    return loss 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def margin_loss(y, pred):
    """
    For the first part of loss(classification loss)
    """
    return K.mean(K.sum(y * K.square(K.maximum(0.9 - pred, 0)) + \
        0.5 *  K.square((1 - y) * K.maximum(pred - 0.1, 0)), axis=1)) 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def box_iou(b1, b2):
    '''Return iou tensor

    Parameters
    ----------
    b1: tensor, shape=(i1,...,iN, 4), xywh
    b2: tensor, shape=(j, 4), xywh

    Returns
    -------
    iou: tensor, shape=(i1,...,iN, j)

    '''

    # Expand dim to apply broadcasting.
    b1 = K.expand_dims(b1, -2)
    b1_xy = b1[..., :2]
    b1_wh = b1[..., 2:4]
    b1_wh_half = b1_wh/2.
    b1_mins = b1_xy - b1_wh_half
    b1_maxes = b1_xy + b1_wh_half

    # Expand dim to apply broadcasting.
    b2 = K.expand_dims(b2, 0)
    b2_xy = b2[..., :2]
    b2_wh = b2[..., 2:4]
    b2_wh_half = b2_wh/2.
    b2_mins = b2_xy - b2_wh_half
    b2_maxes = b2_xy + b2_wh_half

    intersect_mins = K.maximum(b1_mins, b2_mins)
    intersect_maxes = K.minimum(b1_maxes, b2_maxes)
    intersect_wh = K.maximum(intersect_maxes - intersect_mins, 0.)
    intersect_area = intersect_wh[..., 0] * intersect_wh[..., 1]
    b1_area = b1_wh[..., 0] * b1_wh[..., 1]
    b2_area = b2_wh[..., 0] * b2_wh[..., 1]
    iou = intersect_area / (b1_area + b2_area - intersect_area)

    return iou 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def compute_euclidean_match_score(l_r):
    l, r = l_r
    denominator = 1. + K.sqrt(
        -2 * K.batch_dot(l, r, axes=[2, 2]) +
        K.expand_dims(K.sum(K.square(l), axis=2), 2) +
        K.expand_dims(K.sum(K.square(r), axis=2), 1)
    )
    denominator = K.maximum(denominator, K.epsilon())
    return 1. / denominator 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def nrlu(x):
   std = K.mean(K.sigmoid(x))
   eta = K.random_normal(shape=x.shape, std=std)
   y = K.maximum(x + eta, 0)
   return y 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def tensor_iou(box1, box2, input_mask, config):
    """Computes pairwise IOU of two lists of boxes
    
    Arguments:
        box1 {[type]} -- First list of boxes
        box2 {[type]} -- Second list of boxes
        input_mask {[type]} -- Zero-One indicating which boxes to compute
        config {[type]} -- dict containing hyperparameters
    
    Returns:
        [type] -- [description]
    """

    
    xmin = K.maximum(box1[0], box2[0])
    ymin = K.maximum(box1[1], box2[1])
    xmax = K.minimum(box1[2], box2[2])
    ymax = K.minimum(box1[3], box2[3])

    w = K.maximum(0.0, xmax - xmin)
    h = K.maximum(0.0, ymax - ymin)

    intersection = w * h

    w1 = box1[2] - box1[0]
    h1 = box1[3] - box1[1]
    w2 = box2[2] - box2[0]
    h2 = box2[3] - box2[1]

    union = w1 * h1 + w2 * h2 - intersection

    return intersection / (union + config.EPSILON) * K.reshape(input_mask, [config.BATCH_SIZE, config.ANCHORS]) 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def box_iou(b1, b2):
    # 13,13,3,1,4
    # 计算左上角的坐标和右下角的坐标
    b1 = K.expand_dims(b1, -2)
    b1_xy = b1[..., :2]
    b1_wh = b1[..., 2:4]
    b1_wh_half = b1_wh/2.
    b1_mins = b1_xy - b1_wh_half
    b1_maxes = b1_xy + b1_wh_half

    # 1,n,4
    # 计算左上角和右下角的坐标
    b2 = K.expand_dims(b2, 0)
    b2_xy = b2[..., :2]
    b2_wh = b2[..., 2:4]
    b2_wh_half = b2_wh/2.
    b2_mins = b2_xy - b2_wh_half
    b2_maxes = b2_xy + b2_wh_half

    # 计算重合面积
    intersect_mins = K.maximum(b1_mins, b2_mins)
    intersect_maxes = K.minimum(b1_maxes, b2_maxes)
    intersect_wh = K.maximum(intersect_maxes - intersect_mins, 0.)
    intersect_area = intersect_wh[..., 0] * intersect_wh[..., 1]
    b1_area = b1_wh[..., 0] * b1_wh[..., 1]
    b2_area = b2_wh[..., 0] * b2_wh[..., 1]
    iou = intersect_area / (b1_area + b2_area - intersect_area)

    return iou

#---------------------------------------------------#
#   loss值计算
#---------------------------------------------------# 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def triplet_loss(y_true, y_pred):
    batch = batch_num
    ref1 = y_pred[0:batch,:]
    pos1 = y_pred[batch:batch+batch,:]
    neg1 = y_pred[batch+batch:3*batch,:]
    dis_pos = K.sum(K.square(ref1 - pos1), axis=1, keepdims=True)
    dis_neg = K.sum(K.square(ref1 - neg1), axis=1, keepdims=True)
    #dis_pos = K.sqrt(dis_pos)
    #dis_neg = K.sqrt(dis_neg)
    a1pha = 0.2
    d1 = K.maximum(0.0,(dis_pos-dis_neg)+a1pha)
    d2 = K.maximum(0.0,(dis_pos-dis_neg)+a1pha) 
    d = d1+d2
    return K.mean(d) 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def triplet_loss(y_true, y_pred):
    batch = batch_num
    ref1 = y_pred[0:batch,:]
    pos1 = y_pred[batch:batch+batch,:]
    neg1 = y_pred[batch+batch:3*batch,:]
    dis_pos = K.sum(K.square(ref1 - pos1), axis=1, keepdims=True)
    dis_neg = K.sum(K.square(ref1 - neg1), axis=1, keepdims=True)
    #dis_pos = K.sqrt(dis_pos)
    #dis_neg = K.sqrt(dis_neg)
    a1pha = 0.2
    d1 = K.maximum(0.0,(dis_pos-dis_neg)+a1pha)
    d2 = K.maximum(0.0,(dis_pos-dis_neg)+a1pha) 
    d = d1 + d2
    return K.mean(d) 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def binary_dice(y_true, y_pred):
    """
    N-D dice for binary segmentation
    """
    ndims = len(y_pred.get_shape().as_list()) - 2
    vol_axes = list(range(1, ndims+1))

    top = 2 * tf.reduce_sum(y_true * y_pred, vol_axes)
    bottom = tf.maximum(tf.reduce_sum(y_true + y_pred, vol_axes), 1e-5)
    dice = tf.reduce_mean(top/bottom)
    return -dice 

# 需要导入模块: from keras import backend [as 别名]
# 或者: from keras.backend import maximum [as 别名]
def conv_block(self,data, mask, conv_layer, mask_conv_layer, core_name):
        '''
        data is the data tensor
        mask is a binary tensor the same size as data

        steps:
        - set empty voxels in data using data *= mask
        - conv data and mask with the conv conv_layer
        - re-weight data
        - binarize mask
        '''

        # mask.dtype
        # data.dtype
        # make sure the data is sparse according to the mask
        wt_data = keras.layers.Lambda(lambda x: x[0] * x[1], name='%s_pre_wmult' % core_name)([data, mask])
        # convolve data
        conv_data = conv_layer(wt_data)  
    
        # convolve mask
        conv_mask = mask_conv_layer(mask)
        zero_mask = keras.layers.Lambda(lambda x:x*0+1)(mask)
        conv_mask_allones = mask_conv_layer(zero_mask) # all_ones mask to get the edge counts right.
        mask_conv_layer.trainable = False
        o = np.ones(mask_conv_layer.get_weights()[0].shape)
        mask_conv_layer.set_weights([o])
    
        # re-weight data (this is what makes the conv makes sense)
        data_norm = lambda x: x[0] / (x[1] + 1e-2)
        # data_norm = lambda x: x[0] / K.maximum(x[1]/x[2], 1)
        out_data = keras.layers.Lambda(data_norm, name='%s_norm_im' % core_name)([conv_data, conv_mask])
        mask_norm = lambda x: tf.cast(x > 0, tf.float32)
        out_mask = keras.layers.Lambda(mask_norm, name='%s_norm_wt' % core_name)(conv_mask)

        return (out_data, out_mask, conv_data, conv_mask)