文件列表:

.DS_Store (6148, 2022-02-18)
Figs/ (0, 2022-02-18)
Figs/result.png (94910, 2022-02-18)
LICENSE (1061, 2022-02-18)
src/ (0, 2022-02-18)
src/CycleGAN/ (0, 2022-02-18)
src/CycleGAN/datasets.py (2667, 2022-02-18)
src/CycleGAN/datasets_mine_better.py (2569, 2022-02-18)
src/CycleGAN/download_dataset (1123, 2022-02-18)
src/CycleGAN/extract_train.py (6273, 2022-02-18)
src/CycleGAN/fid_score.py (10400, 2022-02-18)
src/CycleGAN/generate_initial_weights.py (2021, 2022-02-18)
src/CycleGAN/inception.py (11623, 2022-02-18)
src/CycleGAN/inference_.run (113, 2022-02-18)
src/CycleGAN/model_clf.py (1749, 2022-02-18)
src/CycleGAN/models.py (4888, 2022-02-18)
src/CycleGAN/models_modified.py (6230, 2022-02-18)
src/CycleGAN/parse_structure.py (864, 2022-02-18)
src/CycleGAN/prun_utils.py (1245, 2022-02-18)
src/CycleGAN/test.py (6332, 2022-02-18)
src/CycleGAN/test_prune.py (8520, 2022-02-18)
src/CycleGAN/train.py (10113, 2022-02-18)
src/CycleGAN/train_imp.py (13250, 2022-02-18)
src/CycleGAN/train_impgd.py (13179, 2022-02-18)
src/CycleGAN/train_oneshot.py (12323, 2022-02-18)
src/CycleGAN/transfer_training_set.py (3544, 2022-02-18)
src/CycleGAN/utils.py (3798, 2022-02-18)
src/CycleGAN/validate.py (7089, 2022-02-18)
src/CycleGAN_cp/ (0, 2022-02-18)
src/CycleGAN_cp/.DS_Store (6148, 2022-02-18)
src/CycleGAN_cp/cp.py (10098, 2022-02-18)
src/CycleGAN_cp/cp.txt (81, 2022-02-18)
src/CycleGAN_cp/cp_finetune.py (10490, 2022-02-18)
src/CycleGAN_cp/cp_finetune2.py (10062, 2022-02-18)
src/CycleGAN_cp/cp_ticket.py (11002, 2022-02-18)
src/CycleGAN_cp/datasets/ (0, 2022-02-18)
... ...

# GANs Can Play Lottery Tickets Too [](https://opensource.org/licenses/MIT) Code for this paper [GANs Can Play Lottery Tickets Too](https://openreview.net/forum?id=1AoMhc_9jER). ## Overview For a range of GANs, we can find matching subnetworks at 67%-74% sparsity. We observe that with or without pruning discriminator has a minor effect on the existence and quality of matching subnetworks, while the initialization used in the discriminator plays a significant role. ## Experiment Results Iterative pruning results on SNGAN  ## Requirements `pytorch==1.4.0` `tensorflow-gpu=1.15.0` `imageio` `scikit-image` `tqdm` `tensorboardx` ## Command ### SNGAN #### Generate Initial Weights ``` mkdir initial_weights python generate_initial_weights.py --model sngan_cifar10 ``` #### Prepare FID statistics Download FID statistics files from [here](https://www.dropbox.com/sh/8xhqxsxnsto18im/AAAkDr-Zf3sgXx1A7RAhlqcva?dl=0) to `fid_stat`. #### Baseline ``` python train.py --model sngan_cifar10 --exp_name sngan_cifar10 --init-path initial_weights ``` Baseline models are also available [here](https://drive.google.com/drive/folders/1-QSfRrVpHSrHppmEf8fuAUn6Nv-N2z2R?usp=sharing). #### Iterative Magnitude Pruning on Generator (IMPG) ``` python train_impg.py --model sngan_cifar10 --exp_name sngan_cifar10 --init-path initial_weights --load-path ``` #### Iterative Magnitude Pruning on Generator (IMPGD) ``` python train_impgd.py --model sngan_cifar10 --exp_name sngan_cifar10 --init-path initial_weights ``` ### Iterative Magnitude Pruning on Generator (IMPGDKD) ``` python train_impgd.py --model sngan_cifar10 --exp_name sngan_cifar10 --init-path initial_weights --use-kd-d ``` ### CycleGAN #### Generate initial weights ``` mkdir initial_weights python generate_initial_weights.py ``` #### Download Data ``` ./download_dataset DATASET_NAME ``` #### Baseline ``` python train.py --dataset DATASET_NAME --rand initial_weights --gpu GPU ``` #### IMPG ``` python train_impg.py --dataset DATASET_NAME --rand initial_weights --gpu GPU --pretrain PRETRAIN ``` #### IMPGD ``` python train_impg.py --dataset DATASET_NAME --rand initial_weights --gpu GPU --pretrain PRETRAIN ``` ## Acknowledgement Inception Score code from OpenAI's Improved GAN (official), and the FID code and CIFAR-10 statistics file from https://github.com/bioinf-jku/TTUR (official).

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：