/STSNet_SW

Codes and models for the manuscript submitted.

Primary LanguagePython

  • how to use folder 1:
    To train the model on all split files in the ./splits directory run this command:

     python main.py --train

    Results, including a copy of the split and python files, will be stored in ./data directory. You can specify different directory with a parameter -o <directory_name> This is convenient if you are running a number of experiments and want to preserve the results and configuration.

    The final results will be recorded in ./data/results.txt with corresponding models in the ./data/models directory.

    By default, the training is done with split files in ./splits directory. These files were created with create_split.py. For example, to create 5 fold split file for the dataset run the following command:

     python create_split.py -d datasets/expression.h5 --save-dir splits --save-name exp_splits --num-splits 5

    The split file will be saved as ./splits/exp_splits.json

  • how to use folder 2:
    first： set list_file_train and list_file_test in main.py properly, each of them is a list file, contents in file like this:
    /home/XXX/fold/1/anger1_1/1 5 1
    ...
    where /home/XXX/fold/1/anger1_1/1 is a fold which contain a image sequence of a expression, 5 is the len of the clips, 1 is the label
    second: set premodel in main.py if you have the pretrained model
    third: run python main.py in your terminal.