GitLab

PiperOrigin-RevId: 362548055

PiperOrigin-RevId: 361199986

PiperOrigin-RevId: 341966519

PiperOrigin-RevId: 341966519

fix the last layer of mobilenet v2 and add testing for complete classification with mobilenet backbone

move Conv2DBNBlock to mobilenet.py; change the reference name for InvertedBottleNeckBlock to align with paper

* factor make_divisible function and move round_filters to nn_layers

* modify SqueezeExcitation to add two additional parameter: divisible_by and gating_activation

* modify the InvertedBottleneckBlock to include 1. use_depthwise, 2. use_residual, 3. regularize_depthwise additional boolean flag; Add control for depthwise activation and regularizer; remove expand_ratio from SqueezeExcitation

* add Conv2DBNBlock definition

* add mobilenet v2, v3 implementation

* add mobilenet v1

* put mobilenet_base into class body

* fix a type hint error

* the invertedbottlenetblock is different for mobilenet and efficientnet. Made necessary changes to cope both.

* add target_backbone while call invertedbottleneckblock

* add relu6 and hard_sigmoid

* add test for mobilenet

* add mobilenet to factory

* fix some typo; link the reference to the architectures

* remove future import
Co-authored-by: Shixin Luo <luoshixin@google.com>