下载
介绍

Keras Applications

Build Status

Keras Applications is the applications module of the Keras deep learning library. It provides model definitions and pre-trained weights for a number of popular archictures, such as VGG16, ResNet50, Xception, MobileNet, and more.

Read the documentation at: https://keras.io/applications/

Keras Applications may be imported directly from an up-to-date installation of Keras:

from keras import applications

Keras Applications is compatible with Python 2.7-3.6 and is distributed under the MIT license.

Performance

  • The top-k accuracies were obtained using Keras Applications with the TensorFlow backend on the 2012 ILSVRC ImageNet validation set and may slightly differ from the original ones.

  • Input: input size fed into models

  • Top-1: single center crop, top-1 accuracy

  • Top-5: single center crop, top-5 accuracy

  • Size: rounded the number of parameters when include_top=True

  • Stem: rounded the number of parameters when include_top=False


InputTop-1Top-5SizeStemReferences
VGG1622471.26890.050138.4M14.7M[paper] [tf-models]
VGG1922471.25689.988143.7M20.0M[paper] [tf-models]
ResNet5022474.92892.06025.6M23.6M[paper] [tf-models] [torch] [caffe]
ResNet10122476.42092.78644.7M42.7M[paper] [tf-models] [torch] [caffe]
ResNet15222476.60493.11860.4M58.4M[paper] [tf-models] [torch] [caffe]
ResNet50V229975.96093.03425.6M23.6M[paper] [tf-models] [torch]
ResNet101V229977.23493.81644.7M42.6M[paper] [tf-models] [torch]
ResNet152V229978.03294.16260.4M58.3M[paper] [tf-models] [torch]
ResNeXt5022477.74093.81025.1M23.0M[paper] [torch]
ResNeXt10122478.73094.29444.3M42.3M[paper] [torch]
InceptionV329977.89893.72023.9M21.8M[paper] [tf-models]
InceptionResNetV229980.25695.25255.9M54.3M[paper] [tf-models]
Xception29979.00694.45222.9M20.9M[paper]
MobileNet(alpha=0.25)22451.58275.7920.5M0.2M[paper] [tf-models]
MobileNet(alpha=0.50)22464.29285.6241.3M0.8M[paper] [tf-models]
MobileNet(alpha=0.75)22468.41288.2422.6M1.8M[paper] [tf-models]
MobileNet(alpha=1.0)22470.42489.5044.3M3.2M[paper] [tf-models]
MobileNetV2(alpha=0.35)22460.08682.4321.7M0.4M[paper] [tf-models]
MobileNetV2(alpha=0.50)22465.19486.0622.0M0.7M[paper] [tf-models]
MobileNetV2(alpha=0.75)22469.53289.1762.7M1.4M[paper] [tf-models]
MobileNetV2(alpha=1.0)22471.33690.1423.5M2.3M[paper] [tf-models]
MobileNetV2(alpha=1.3)22474.68092.1225.4M3.8M[paper] [tf-models]
MobileNetV2(alpha=1.4)22475.23092.4226.2M4.4M[paper] [tf-models]
MobileNetV3(small)22468.07687.8002.6M0.9M[paper] [tf-models]
MobileNetV3(large)22475.55692.7085.5M3.0M[paper] [tf-models]
DenseNet12122474.97292.2588.1M7.0M[paper] [torch]
DenseNet16922476.17693.17614.3M12.6M[paper] [torch]
DenseNet20122477.32093.62020.2M18.3M[paper] [torch]
NASNetLarge33182.49896.00493.5M84.9M[paper] [tf-models]
NASNetMobile22474.36691.8547.7M4.3M[paper] [tf-models]
EfficientNet-B022477.19093.4925.3M4.0M[paper] [tf-tpu]
EfficientNet-B124079.13494.4487.9M6.6M[paper] [tf-tpu]
EfficientNet-B226080.18094.9469.2M7.8M[paper] [tf-tpu]
EfficientNet-B330081.57895.67612.3M10.8M[paper] [tf-tpu]
EfficientNet-B438082.96096.26019.5M17.7M[paper] [tf-tpu]
EfficientNet-B545683.70296.71030.6M28.5M[paper] [tf-tpu]
EfficientNet-B652884.08296.89843.3M41.0M[paper] [tf-tpu]
EfficientNet-B760084.43096.84066.7M64.1M[paper] [tf-tpu]

Reference implementations from the community

Object detection and segmentation

Sequence learning

Reinforcement learning

Generative adversarial networks

相关推荐
tensorflow / tensorflow

一个面向所有人的开源机器学习框架

tensorflow
2021-01-18

opencv / opencv

Open Source Computer Vision Library

opencv
2021-01-26

keras-team / keras

Deep Learning for humans

keras-team
2021-01-19