/tfgarden

CNN models for sensor-based human activity recognition built in TensorFlow

Primary LanguagePythonMIT LicenseMIT

TensorFlow model Garden for Human Activity Recognition

The TensorFlow model Garden for Human Activity Recognition (tfgarden) is the repository of CNN models implemented for sensor-based human activity recognition, like tensorflow.keras.applications.

The models implemented here can also be used as a source domain for sensor-based task (e.g. sidewalk surface type estimation).

Modules

  • densenet module: DenseNet models for Keras.
  • efficientnet module : EfficientNet models for Keras.
  • efficientnet_v2 module: EfficientNetV2 models for Keras.
  • inception_resnet_v2 module: Inception-ResNet V2 model for Keras.
  • inception_v3 module: Inception V3 model for Keras.
  • mobilenet module : MobileNet v1 models for Keras.
  • mobilenet_v2 module: MobileNet v2 models for Keras.
  • mobilenet_v3 module: MobileNet v3 models for Keras.
  • nasnet module: NASNet-A models for Keras.
  • resnet module : ResNet models for Keras.
  • resnet_v2 module: ResNet v2 models for Keras, Not implemented yet.
  • vgg11 module : VGG11 model for Keras.
  • vgg13 module : VGG13 model for Keras.
  • vgg16 module : VGG16 model for Keras.
  • vgg19 module : VGG19 model for Keras.
  • xception module: Xception V1 model for Keras.
  • mnasnet module: MnasNet-A1 model for Keras.
  • pyramidnet module : PyramidNet models for Keras.
  • efficientnet_lite module : EfficientNet-Lite models for Keras.

Performance

Please refer to tfmars

Install

pip install git+https://github.com/Shakshi3104/tfgarden.git

Dependency

  • tensorflow >= 2.0