astar-har's Introduction

A*HAR: A NEW BENCHMARK TOWARDS SEMI-SUPERVISED LEARNING FOR CLASS IMBALANCED HUMAN ACTIVITY RECOGNITION

The tensorflow code for "A*HAR: A NEW BENCHMARK TOWARDS SEMI-SUPERVISED LEARNING FOR CLASS IMBALANCED HUMAN ACTIVITY RECOGNITION " (https://arxiv.org/abs/2101.04859)

The environment can be found in dockerhub: docker pull loklu/mt_tensorflow:tf1.2.1_py35_lib2

The code runs on Python 3. Install the dependencies and prepare the datasets with the following commands:

pip install tensorflow==1.2.1 numpy scipy pandas
./prepare_data.sh

To train the model, run:

'python /experiments/har/har.py --template X' for HAR models. modify arguments in template.py for choosing type of experiment and no: of unlabeled samples , Where X can be defined as suitable template

astar-har's People

Contributors

Stargazers

Watchers

astar-har's Issues

cifar10 data_path error

Error in trials:

file_data = np.load(self.DATA_PATH)
  File "/usr/local/lib/python3.6/dist-packages/numpy/lib/npyio.py", line 428, in load
    fid = open(os_fspath(file), "rb")
FileNotFoundError: [Errno 2] No such file or directory: 'data/images/cifar/cifar10/cifar10_gcn_zca_v2.npz'

in cifar10.py, the path is built as follows:
DATA_PATH = os.path.join('data', 'images', 'cifar', 'cifar10', 'cifar10_gcn_zca_v2.npz')

Is there a missing data folder in this repo?

Recommend Projects