Add paddle.nn.SmoothL1Loss

[yghstill]

PR types

New features

PR changes

APIs

Describe

add paddle.nn.SmoothL1Loss. This operator is calculate smooth_l1_loss. Creates a criterion that uses a squared term if the absolute element-wise error falls below 1 and an L1 term otherwise. In some cases it can prevent exploding gradients. Also known as the Huber loss.

class:

class paddle.nn.SmoothL1Loss(reduction='mean', delta=1.0, name=None)

functioanl:

def paddle.nn.functioanl.smooth_l1_loss(input, label, reduction='mean', delta=1.0, name=None)

docs

Examples:

           import paddle
            import numpy as np
            paddle.disable_static()
            input_data = np.random.rand(3,3).astype("float32")
            label_data = np.random.rand(3,3).astype("float32")
            input = paddle.to_tensor(input_data)
            label = paddle.to_tensor(label_data)
            loss = paddle.nn.SmoothL1Loss()
            output = loss(input, label)
            print(output.numpy())

[paddle-bot-old]

Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

[yaoxuefeng6]

LGTM

[jzhang533]

lgtm
great to know no c++ op is introduced and huber_loss api is reused.

[XiaoguangHu01]

LGTM