- Notifications
You must be signed in to change notification settings - Fork1
License
NotificationsYou must be signed in to change notification settings
Bernardo1998/FairGRAPE
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
This repo presents an official implementation of FairGRAPE: Fairness-aware GRAdient Pruning mEthod for Face Attribute Classification
The code has been tested on the following environment:
python 3.9pytorch 1.11.0dlib 19.22.0opencv2 4.5.5Use Anaconda and the following command to replicate the full environment:
conda env create -f environment.ymlThis code automatically downloads the following datasets for trianing, cross validation and testing:FairFace,UTKFace andCelebA. Request demographics labels of the Imagenet person subtree through the officaldatabase, save the annotation file under /csv and images under /Images/Imagenet.
UTKFace experiments:
python main_test.py --sensitive_group 'gender' --loss_type 'race' --prune_type 'FairGRAPE' --network 'resnet34' --dataset 'UTKFace' --prune_rate 0.9 --keep_per_iter 0.975FairFace experiments
python main_test.py --sensitive_group race--loss_type 'race' --prune_type 'FairGRAPE' --network 'resnet34' --dataset 'FairFace' --prune_rate 0.99CelebA experiments
python main_test.py --sensitive_group 'gender' --loss_type 'classes' --prune_type 'FairGRAPE' --network 'mobilenetv2' --dataset 'CelebA' --prune_rate 0.9Imagenet experiments:
python main_test.py --sensitive_group 'gender' --loss_type 'classes' --prune_type 'FairGRAPE' --network 'resnet34' --dataset 'Imagenet' --prune_rate 0.5Download trained modelshere.
Loading a trained model:
python main_test.py --sensitive_group 'gender' --loss_type 'race' --prune_type 'FairGRAPE' --network 'mobilenetv2'--dataset 'UTKFace' --prune_rate 0.9 --keep_per_iter 0.975 --checkpoint "UTKFace_FairGRAPE_race_bygender_resnet34_0.9_0.pt" --init_train 0 --retrain 0 --print_acc 1