Python backend.min方法代码示例

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def _get_min_max_exponents(non_sign_bits, need_exponent_sign_bit,
                           quadratic_approximation):
  """Given a bitwidth, gets min and max exponents that it can represent.

  Args:
    non_sign_bits: An integer representing the bitwidth of the exponent.
    need_exponent_sign_bit: An integer representing whether it needs sign bit
      in exponent. (1: need sign bit. 0: sign bit is not needed.)
    quadratic_approximation: A boolean representing whether the quadratic
      approximiation method is enforced.

  Returns:
    A tuple of integers: min_exp, max_exp
  """
  effect_bits = non_sign_bits - need_exponent_sign_bit
  min_exp = -2**(effect_bits)
  max_exp = 2**(effect_bits) - 1
  if quadratic_approximation:
    max_exp = 2 * (max_exp // 2)
  return min_exp, max_exp 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def min(self):
    """Get minimum value that quantized_bits class can represent."""
    if not self.keep_negative:
      return 0.0
    unsigned_bits = self.bits - self.keep_negative
    if unsigned_bits > 0:
      return -max(1.0, np.power(2.0, self.integer))
    else:
      return -1.0 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def pn_loss(margin=1):
    def _pn_loss(y_true, y_pred):    
        anchor, positive, negative = tf.unstack(y_pred)

        anchor_positive_distance = _euclidean_distance(anchor, positive)
        anchor_negative_distance = _euclidean_distance(anchor, negative)
        positive_negative_distance = _euclidean_distance(positive, negative)

        minimum_distance = K.min(K.concatenate([anchor_negative_distance, positive_negative_distance]), axis=-1, keepdims=True)

        return K.mean(K.maximum(anchor_positive_distance - minimum_distance + margin, 0))

    return _pn_loss 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def manhattan_pn_loss(margin=1):
    def _pn_loss(y_true, y_pred):    
        anchor, positive, negative = tf.unstack(y_pred)

        anchor_positive_distance = _manhattan_distance(anchor, positive)
        anchor_negative_distance = _manhattan_distance(anchor, negative)
        positive_negative_distance = _manhattan_distance(positive, negative)

        minimum_distance = K.min(K.concatenate([anchor_negative_distance, positive_negative_distance]), axis=-1, keepdims=True)

        return K.mean(K.maximum(anchor_positive_distance - minimum_distance + margin, 0))

    return _pn_loss 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def chebyshev_pn_loss(margin=1):
    def _pn_loss(y_true, y_pred):
        anchor, positive, negative = tf.unstack(y_pred)

        anchor_positive_distance = _chebyshev_distance(anchor, positive)
        anchor_negative_distance = _chebyshev_distance(anchor, negative)
        positive_negative_distance = _chebyshev_distance(positive, negative)

        minimum_distance = K.min(K.concatenate([anchor_negative_distance, positive_negative_distance]), axis=-1, keepdims=True)

        return K.mean(K.maximum(anchor_positive_distance - minimum_distance + margin, 0))

    return _pn_loss 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def cosine_pn_loss(margin=1):
    def _pn_loss(y_true, y_pred):
        anchor, positive, negative = tf.unstack(y_pred)

        anchor_positive_distance = _cosine_distance(anchor, positive)
        anchor_negative_distance = _cosine_distance(anchor, negative)
        positive_negative_distance = _cosine_distance(positive, negative)

        minimum_distance = K.min(tf.stack([anchor_negative_distance, positive_negative_distance]), axis=0, keepdims=True)

        return K.mean(K.maximum(anchor_positive_distance - minimum_distance + margin, 0))

    return _pn_loss 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def softmax_ratio_pn(y_true, y_pred):
    anchor, positive, negative = tf.unstack(y_pred)

    anchor_positive_distance = _euclidean_distance(anchor, positive)
    anchor_negative_distance = _euclidean_distance(anchor, negative)
    positive_negative_distance = _euclidean_distance(positive, negative)

    minimum_distance = K.min(K.concatenate([anchor_negative_distance, positive_negative_distance]), axis=-1, keepdims=True)

    softmax = K.softmax(K.concatenate([anchor_positive_distance, minimum_distance]))
    ideal_distance = K.variable([0, 1])
    return K.mean(K.maximum(softmax - ideal_distance, 0)) 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def call(self, inputs, **kwargs):
        return K.min(inputs, axis=1) 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def _check_series_length(self, X):
        """Ensures that time series in X matches the following requirements:
        
        - their length is greater than the size of the longest shapelet
        - (at predict time) their length is lower than the maximum allowed 
        length, as set by self.max_size
        """
        sizes = numpy.array([ts_size(Xi) for Xi in X])
        self._min_sz_fit = sizes.min()

        if self.n_shapelets_per_size is not None:
            max_sz_shp = max(self.n_shapelets_per_size.keys())
            if max_sz_shp > self._min_sz_fit:
                raise ValueError("Sizes in X do not match maximum "
                                 "shapelet size: there is at least one "
                                 "series in X that is shorter than one of the "
                                 "shapelets. Shortest time series is of "
                                 "length {} and longest shapelet is of length "
                                 "{}".format(self._min_sz_fit, max_sz_shp))

        if hasattr(self, 'model_') or self.max_size is not None:
            # Model is already fitted
            max_sz_X = sizes.max()

            if hasattr(self, 'model_'):
                max_size = self._X_fit_dims[1]
            else:
                max_size = self.max_size
            if max_size < max_sz_X:
                raise ValueError("Sizes in X do not match maximum allowed "
                                 "size as set by max_size. "
                                 "Longest time series is of "
                                 "length {} and max_size is "
                                 "{}".format(max_sz_X, max_size)) 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def _batch_hard_triplet_loss(self, y_true: Tensor, pairwise_dist: Tensor) -> Tensor:
        mask_anchor_positive = self._get_anchor_positive_triplet_mask(y_true, pairwise_dist)
        anchor_positive_dist = mask_anchor_positive * pairwise_dist
        hardest_positive_dist = K.max(anchor_positive_dist, axis=1, keepdims=True)
        mask_anchor_negative = self._get_anchor_negative_triplet_mask(y_true, pairwise_dist)
        anchor_negative_dist = mask_anchor_negative * pairwise_dist
        mask_anchor_negative = self._get_semihard_anchor_negative_triplet_mask(anchor_negative_dist,
                                                                               hardest_positive_dist,
                                                                               mask_anchor_negative)
        max_anchor_negative_dist = K.max(pairwise_dist, axis=1, keepdims=True)
        anchor_negative_dist = pairwise_dist + max_anchor_negative_dist * (1.0 - mask_anchor_negative)
        hardest_negative_dist = K.min(anchor_negative_dist, axis=1, keepdims=True)
        triplet_loss = K.clip(hardest_positive_dist - hardest_negative_dist + self.margin, 0.0, None)
        triplet_loss = K.mean(triplet_loss)
        return triplet_loss 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def yolo3_correct_boxes(box_xy, box_wh, input_shape, image_shape):
    '''Get corrected boxes'''
    input_shape = K.cast(input_shape, K.dtype(box_xy))
    image_shape = K.cast(image_shape, K.dtype(box_xy))

    #reshape the image_shape tensor to align with boxes dimension
    image_shape = K.reshape(image_shape, [-1, 1, 1, 1, 2])

    new_shape = K.round(image_shape * K.min(input_shape/image_shape))
    offset = (input_shape-new_shape)/2./input_shape
    scale = input_shape/new_shape
    # reverse offset/scale to match (w,h) order
    offset = offset[..., ::-1]
    scale = scale[..., ::-1]

    box_xy = (box_xy - offset) * scale
    box_wh *= scale

    box_mins = box_xy - (box_wh / 2.)
    box_maxes = box_xy + (box_wh / 2.)
    boxes =  K.concatenate([
        box_mins[..., 0:1],  # x_min
        box_mins[..., 1:2],  # y_min
        box_maxes[..., 0:1],  # x_max
        box_maxes[..., 1:2]  # y_max
    ])

    # Scale boxes back to original image shape.
    image_wh = image_shape[..., ::-1]
    boxes *= K.concatenate([image_wh, image_wh])
    return boxes 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def yolo2_correct_boxes(box_xy, box_wh, input_shape, image_shape):
    '''Get corrected boxes'''
    input_shape = K.cast(input_shape, K.dtype(box_xy))
    image_shape = K.cast(image_shape, K.dtype(box_xy))

    #reshape the image_shape tensor to align with boxes dimension
    image_shape = K.reshape(image_shape, [-1, 1, 1, 1, 2])

    new_shape = K.round(image_shape * K.min(input_shape/image_shape))
    offset = (input_shape-new_shape)/2./input_shape
    scale = input_shape/new_shape
    # reverse offset/scale to match (w,h) order
    offset = offset[..., ::-1]
    scale = scale[..., ::-1]

    box_xy = (box_xy - offset) * scale
    box_wh *= scale

    box_mins = box_xy - (box_wh / 2.)
    box_maxes = box_xy + (box_wh / 2.)
    boxes =  K.concatenate([
        box_mins[..., 0:1],  # x_min
        box_mins[..., 1:2],  # y_min
        box_maxes[..., 0:1],  # x_max
        box_maxes[..., 1:2]  # y_max
    ])

    # Scale boxes back to original image shape.
    image_wh = image_shape[..., ::-1]
    boxes *= K.concatenate([image_wh, image_wh])
    return boxes 

# 需要导入模块: from tensorflow.keras import backend [as 别名]
# 或者: from tensorflow.keras.backend import min [as 别名]
def grabocka_params_to_shapelet_size_dict(n_ts, ts_sz, n_classes, l, r):
    """Compute number and length of shapelets.

     This function uses the heuristic from [1]_.

    Parameters
    ----------
    n_ts: int
        Number of time series in the dataset
    ts_sz: int
        Length of time series in the dataset
    n_classes: int
        Number of classes in the dataset
    l: float
        Fraction of the length of time series to be used for base shapelet
        length
    r: int
        Number of different shapelet lengths to use

    Returns
    -------
    dict
        Dictionary giving, for each shapelet length, the number of such
        shapelets to be generated

    Examples
    --------
    >>> d = grabocka_params_to_shapelet_size_dict(
    ...         n_ts=100, ts_sz=100, n_classes=3, l=0.1, r=2)
    >>> keys = sorted(d.keys())
    >>> print(keys)
    [10, 20]
    >>> print([d[k] for k in keys])
    [4, 4]


    References
    ----------
    .. [1] J. Grabocka et al. Learning Time-Series Shapelets. SIGKDD 2014.
    """
    base_size = int(l * ts_sz)
    base_size = max(base_size, 1)
    r = min(r, ts_sz)
    d = {}
    for sz_idx in range(r):
        shp_sz = base_size * (sz_idx + 1)
        n_shapelets = int(numpy.log10(n_ts *
                                      (ts_sz - shp_sz + 1) *
                                      (n_classes - 1)))
        n_shapelets = max(1, n_shapelets)
        d[shp_sz] = n_shapelets
    return d