Use a better formula to calculate sigmoid_bce and logistic_loss

[sxjscience]

Description

Use the same formula to calculate the BCE as TF https://github.com/tensorflow/tensorflow/blob/r1.1/tensorflow/python/ops/nn_impl.py#L128

The formula is simpler than the old one and is more accurate.

import mxnet as mx
import mxnet.ndarray as nd

def sigmoid_bce1(pred, label):
    max_val = nd.relu(-pred)
    loss = pred - pred*label + max_val + nd.log(nd.exp(-max_val)+nd.exp(-pred-max_val))
    return loss.asscalar()

def sigmoid_bce2(pred, label):
    loss = nd.relu(pred) - pred * label + nd.Activation(-nd.abs(pred), act_type='softrelu')
    return loss.asscalar()

for x, label in [(-100, 0), (-50, 0), (-20, 0), (-10, 0), (-1, 0), (1, 1), (10, 1), (20, 1), (50, 1), (100, 1)]:
    x_nd = nd.array([x], ctx=mx.gpu())
    label_nd = nd.array([label], ctx=mx.gpu())
    val1 = sigmoid_bce1(x_nd, label_nd)
    val2 = sigmoid_bce2(x_nd, label_nd)
    print("Formula1=", val1, "Formula2=", val2)

Formula1= 0.0 Formula2= 3.78351e-44
Formula1= 0.0 Formula2= 1.92875e-22
Formula1= 0.0 Formula2= 2.06115e-09
Formula1= 4.54177e-05 Formula2= 4.53989e-05
Formula1= 0.313262 Formula2= 0.313262
Formula1= 0.313262 Formula2= 0.313262
Formula1= 4.54177e-05 Formula2= 4.53989e-05
Formula1= 0.0 Formula2= 2.06115e-09
Formula1= 0.0 Formula2= 1.92875e-22
Formula1= 0.0 Formula2= 3.78351e-44

Checklist

Essentials

• Passed code style checking (make lint)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage:
• Code is well-documented:
• For user-facing API changes, API doc string has been updated.
• For new C++ functions in header files, their functionalities and arguments are documented.
• For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
• To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

• Use max(x, 0) - x * z + log(1 + exp(-abs(x))) to calculate BCE + Test

[szha]

Could you also update the docstring for Input in LogisticLoss? It should have been updated when the label_format flag was introduced.

[sxjscience]

@szha I've updated the description of 'label'.