Skip to content

LarryUESTC/GATE

Repository files navigation

GATE

Contents

  1. Installation
  2. Preparation
  3. Testing
  4. Training

Installation

pip install -r requirements.txt

Preparation

Pls see Configs >>>here Lines 528-544<<<.

Configs Custom key note
dataset ABIDE/FTD dataset
seed random seeds default is 2022
train train or test (bool) None
label_rate 0.2 (flaot) None
knn the top k in KNN (int) default is 5
SAMA use SA+MA (bool) default is True
beta the trade-off coefficient in Eq. (2) default is 0.2
lastdim dimensions in SSL 256 and 512

other args:

  • --lastdim2 dimension in fine-tune stage
  • --epoch_all1 Epoch in SSL
  • --epoch_all2 Epoch in fine-tune stage
  • --lr1 learning rate in SSL
  • --lr2 learning rate in fine-tune stage
  • --weight_decay1 weight decay in SSL
  • --weight_decay2 weight decay in fine-tune stage

Data: The pre-processed data are uploaded into Baidu Cloud Drive, and you can also generate your owns (see Pre_time_MA.py and Pre_time_SA.py).

Link:https://pan.baidu.com/s/1r4bLcJn0PM1ThDKZ2IFOVA?pwd=g559 code:g559

Testing

  • Classification Task ( Default dataset is ABIDE )
python main.py --train False

Training

  • Classification Task ( Default dataset is ABIDE )
python main.py --train True

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages