Skip to content

deepaklunked/Keras-NASNet

 
 

Repository files navigation

Keras Neural Architecture Search Network (NASNet)

An implementation of "NASNet" models from the paper Learning Transferable Architectures for Scalable Image Recognitio in Keras 2.0+.

Based on the models described in the TFSlim implementation and some modules from the TensorNets implementation

Weights have been ported over from the official NASNet Tensorflow repository.

Since CIFAR weights are not provided, and I don't have the resources to train such large models on CIFAR, those weights will not be provided. External help is appreciated.

Usage

All types of NASNet models can be built. In addition, NASNet Large - NASNet (6 @ 4032) and NASNet Mobile - NASNet (4 @ 1056) are prebuilt and provided as NASNetLarge and NASNetMobile.

Building a speficific NASNet model

from nasnet import NASNet

# the parameters for NASNetLarge
model = NASNet(input_shape=(331, 331, 3),
           penultimate_filters=4032,
           nb_blocks=6,
           stem_filters=96,
           skip_reduction=True,
           use_auxilary_branch=False,
           filters_multiplier=2,
           dropout=0.5,
           classes=1000)

Using Pre-built NASNet models

from nasnet import NASNetLarge, NASNetMobile

model = NASNetLarge(input_shape=(331, 331, 3), dropout=0.5)

Network Overview

About

"NASNet" models in Keras 2.0+ with weights

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 100.0%