issue with model last layer

[speeding-motor]

in keras_darknet19.py

def darknet19(inputs):
    """Generate Darknet-19 model for Imagenet classification."""
    body = darknet_body()(inputs)
    logits = DarknetConv2D(1000, (1, 1), activation='softmax')(body)
    return Model(inputs, logits)

question : 1、here 1000 = anchor_num * (5 + class_num) in other dataset? 2、and why use activation softmax here?

I'm really confused with that, any help is appreciate ,thanks in advance