Inconsistencies in Gluon Model Zoo API

[ChaiBapchya]

Currently, if I were to use Gluon Model Zoo and use any model for eg DenseNet, I face 2 issues regarding APIs a. Lack of clarity about differences in the APIs (Titlecase vs lowercase) E.g. mxnet.gluon.model_zoo.vision.AlexNet(classes=1000, **kwargs) and mxnet.gluon.model_zoo.vision.alexnet(pretrained=False, ctx=cpu(0), root='/home/jenkins_slave/.mxnet/models', **kwargs)

b. Even if that were clear, APIs themselves weren't consistent with function definitions for few networks

By inconsistency I mean, **kwargs instead of mentioning all the parameters that are required by the function. (e.g. pretrained=False, ctx=cpu(0))

Case in point - mobile nets, densenets, resnets, squeezenets (note the case sensitivity of these APIs)

• mxnet.gluon.model_zoo.vision.mobilenet_v2_0_25(**kwargs)
• mxnet.gluon.model_zoo.vision.densenet161(**kwargs)
• mxnet.gluon.model_zoo.vision.resnet152_v1(**kwargs)
• mxnet.gluon.model_zoo.vision.squeezenet1_0(**kwargs)
• mxnet.gluon.model_zoo.vision.vgg13(**kwargs)

Compared to the likes of alexnet and inceptions

• mxnet.gluon.model_zoo.vision.alexnet(pretrained=False, ctx=cpu(0), root='/home/jenkins_slave/.mxnet/models', **kwargs)
• mxnet.gluon.model_zoo.vision.inception_v3(pretrained=False, ctx=cpu(0), root='/home/jenkins_slave/.mxnet/models', **kwargs)

[mxnet-label-bot]

Hey, this is the MXNet Label Bot. Thank you for submitting the issue! I will try and suggest some labels so that the appropriate MXNet community members can help resolve it.

[ChaiBapchya]

@mxnet-label-bot add [API change]