BTDA

KO | EN

gitlite — search

#python #react #ai #typescript #go #android #machine-learning #deep-learning #linux #docker #macos #windows #awesome-list #vue

(★ 38)

implement for paper Rethinking Domain Adaptation Blending target Domain Adaptation by Adversarial Meta Adaptation Network

BTDA 최신버젼 다운로드

// repository documentation

★ 0 (0 ratings)

# BTDA
implement for paper Rethinking Domain Adaptation Blending target Domain Adaptation by Adversarial Meta Adaptation Network

## Requirements
* python 2.7
* PyTorch 0.4.1
* cuda 8.0
* numpy
* tqdm
* visdom
* PIL
* argparse
* sklearn
* torchvision

## download dataset

cd BTDA

bash download_dataset.sh

## digit dataset

cd BTDA/Digit

python train.py --source_name mnist --gpu_id 0

## office31 dataset

cd BTDA/Office

python alexnet_train.py --dataset_name Office31 --source_name amazon --gpu_id 0

python resnet_train.py  --dataset_name Office31 --source_name amazon --gpu_id 0

## officeHome dataset

cd BTDA/Office

python alexnet_train.py --dataset_name OfficeHome --source_name A --gpu_id 0

python resnet_train.py  --dataset_name OfficeHome --source_name A --gpu_id 0

## some result modles

alexnet amazon->webcam,dslr acc:74.6 https://drive.google.com/open?id=1_6KGKGfhNJkOo5j0p154FQKCIavrrrpr

resnet  amazon->webcam,dslr acc:90.1 https://drive.google.com/open?id=1UGMjGaaHKQvEAGLGym0R-fQyQIeOkjOa