Keras-NASNet:Keras 2.0+中带有权重的“ NASNet”模型

  • R3_843493
    了解作者
  • 3.5MB
    文件大小
  • zip
    文件格式
  • 0
    收藏次数
  • VIP专享
    资源类型
  • 0
    下载次数
  • 2022-04-10 18:20
    上传日期
Keras神经架构搜索网络(NASNet) Keras 2.0+中的论文“”中“ NASNet”模型的实现。 基于描述的模型从和一些模块 权重已从正式的移植。 由于未提供CIFAR权重,并且我没有资源在CIFAR上训练如此大的模型,因此不会提供这些权重。 外部的帮助表示赞赏。 用法 可以构建所有类型的NASNet模型。 此外,还NASNet Large - NASNet (6 @ 4032)和NASNet Mobile - NASNet (4 @ 1056) ,并以NASNetLarge和NASNetMobile 。 建立专门的NASNet模型 from nasnet import NASNet # the parameters for NASNetLarge model = NASNet ( input_shape = ( 331 , 331 , 3 ),
Keras-NASNet-master.zip
  • Keras-NASNet-master
  • logs
  • events.out.tfevents.1510943462.DESKTOP-8LOD6L4
    12.9MB
  • events.out.tfevents.1510943894.DESKTOP-8LOD6L4
    14.8MB
  • events.out.tfevents.1510944261.DESKTOP-8LOD6L4
    14MB
  • images
  • cheetah.jpg
    1011.9KB
  • nasnet_mobile.png
    241.7KB
  • elephant.jpg
    51.9KB
  • weight_translation
  • weight_load_large.py
    11.8KB
  • nasnet.py
    36.8KB
  • extract_weights.py
    2.4KB
  • weight_load_mobile.py
    11.7KB
  • cifar10_nasnet.py
    4.2KB
  • cutout.py
    1.3KB
  • nasnet.py
    36.2KB
  • tf_hub_compare.py
    1.6KB
  • predict_imagenet.py
    1.3KB
  • LICENSE
    1KB
  • README.md
    1.7KB
  • imagenet_validation.py
    2.2KB
  • ImageNet_Val.txt
    1.6MB
  • .gitignore
    1.2KB
内容介绍
# Keras Neural Architecture Search Network (NASNet) An implementation of "NASNet" models from the paper [Learning Transferable Architectures for Scalable Image Recognitio](https://arxiv.org/abs/1707.07012) in Keras 2.0+. Based on the models described in the [TFSlim implementation](https://github.com/tensorflow/models/tree/master/research/slim/nets/nasnet) and some modules from the [TensorNets implementation](https://github.com/taehoonlee/tensornets/blob/master/tensornets/nasnets.py) Weights have been ported over from the official [NASNet Tensorflow repository](https://github.com/tensorflow/models/tree/master/research/slim/nets/nasnet). 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 ```python 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 ```python from nasnet import NASNetLarge, NASNetMobile model = NASNetLarge(input_shape=(331, 331, 3), dropout=0.5) ``` # Network Overview <img src="https://github.com/titu1994/Keras-NASNet/blob/master/images/nasnet_mobile.png?raw=true" height=100% width=100%>
评论
    相关推荐