TWNet for spotting the macro- and micro- intervals in the long video sequence

File	Description
requirements.txt	Install the requirements that the code needs.
static_pretraining.py	Static module pre-training model, using revised Fer2013dataset.
dynamic_pretraining.py	Dynamic module pre-training model, using processed CK+ dataset.
TWNet.py	Network architecture of TWNet,and the training function of TWNet.
utils.py	Implementation of some common methods.
peak_detection.py	Improved peak detection algorithm,three filters by height,distance and width.
casme_evaluation.py	On CAS(ME)2 data set, sequence prediction and evaluation to generate TP,FP,FN,F1-score and so on.
samm_evaluation.py	On SAMM data set, sequence prediction and evaluation to generate TP,FP,FN,F1-score and so on.
Data	Some data in this task,read the README.txt in data.