def test_gather_pynative_fp16_14(): context.set_context(mode=context.PYNATIVE_MODE, device_target="GPU") error = 1e-4 x = Tensor(np.array([1., 2., 3., 4.]), ms.float16) expect = np.array(15, np.float16) output = P.L2Loss()(x) diff = output.asnumpy() - expect assert np.all(diff < error)
def __init__(self): super(openpose_loss, self).__init__() self.expand_dims = P.ExpandDims() self.tile = P.Tile() self.mul = P.Mul() self.l2_loss = P.L2Loss() self.square = P.Square() self.reduceMean = P.ReduceMean() self.reduceSum = P.ReduceSum() self.print = P.Print() self.shape = P.Shape() self.maxoftensor = P.ArgMaxWithValue(-1)
def __init__(self, label, mask, weight_decay, param): super(Loss, self).__init__(auto_prefix=False) self.label = Tensor(label) self.mask = Tensor(mask) self.loss = P.SoftmaxCrossEntropyWithLogits() self.one = Tensor(1.0, mstype.float32) self.zero = Tensor(0.0, mstype.float32) self.mean = P.ReduceMean() self.cast = P.Cast() self.l2_loss = P.L2Loss() self.reduce_sum = P.ReduceSum() self.weight_decay = weight_decay self.param = param
def construct(self, logits): """calc l2 loss""" l2_loss = 0 for i in range(self.num_params): l2_loss = l2_loss + self.l2_coeff * P.L2Loss()(self.params[i]) logits = P.Reshape()(logits, (-1, self.num_class)) label = P.Reshape()(self.label, (-1, self.num_class)) mask = P.Reshape()(self.mask, (-1,)) logits = self.cast(logits, mstype.float32) loss = self.softmax(logits, label)[0] mask /= self.reduce_mean(mask) loss *= mask loss = self.reduce_mean(loss) l2_loss = P.Cast()(l2_loss, mstype.float32) return loss+l2_loss
def __init__(self, neg_item_num, l2_embed, dist_reg): super(BGCFLoss, self).__init__() self.neg_item_num = neg_item_num self.l2_embed = l2_embed self.dist_reg = dist_reg self.log = P.Log() self.pow = P.Pow() self.cast = P.Cast() self.tile = P.Tile() self.shape = P.Shape() self.reshape = P.Reshape() self.concat = P.Concat(1) self.concat2 = P.Concat(2) self.split = P.Split(0, 2) self.reduce_sum = P.ReduceSum() self.expand_dims = P.ExpandDims() self.multiply = P.Mul() self.matmul = P.BatchMatMul() self.squeeze = P.Squeeze(1) self.transpose = P.Transpose() self.l2_loss = P.L2Loss() self.sigmoid = P.Sigmoid()
'desc_bprop': [3, 3], 'skip': ['backward']}), ('ApplyRMSProp', { 'block': P.ApplyRMSProp(), 'desc_const': [0.9, 0.0, 1e-10, 0.001], 'desc_inputs': [[3, 3], [3, 3], [3, 3], [3, 3]], 'desc_bprop': [3, 3], 'skip': ['backward']}), ('ApplyCenteredRMSProp', { 'block': P.ApplyCenteredRMSProp(), 'desc_const': [0.9, 0.0, 1e-10, 0.001], 'desc_inputs': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'desc_bprop': [3, 3], 'skip': ['backward']}), ('L2Loss_1', { 'block': P.L2Loss(), 'desc_inputs': [Tensor(np.array([1, 2, 3, 4]), mstype.float16)], 'desc_bprop': []}), ('L2Loss_2', { 'block': P.L2Loss(), 'desc_inputs': [Tensor(np.array([[1, 1], [2, 2], [3, 3], [4, 4]]), mstype.float16)], 'desc_bprop': []}), ] test_case_array_ops = [ ('SpaceToDepth', { 'block': P.SpaceToDepth(2), 'desc_inputs': [[1, 3, 2, 2]], 'desc_bprop': [[1, 12, 1, 1]]}), ('DepthToSpace', { 'block': P.DepthToSpace(2),
def __init__(self): super(L2LossNet, self).__init__() self.l2_loss = P.L2Loss()