Evaluate.md

Evaluate Model

Evaluate test dataset

cd ./evaluate
# change model name when init_point_model
python evaluate.py

You can output results on the test dataset with batch input.

The results will show in the terminal.

Evaluate looping all h5 files

# change exp_name and init_point_model for each evaluation
python evaluate_all_h5.py

You can save an .xlsx file which containes each result for the .h5 file in this folder.

Evaluate on a single file

# change sub, session, mov for the specific movement, as well as exp_name and init_point_model
python evaluate_video_h5.py

You can save the gif and videos in this folder.