hc/JSCC
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

add imagenet

Browse Source
This commit is contained in:
chun 2024-01-19 20:59:47 +08:00
parent ab409da89e
commit c1c6226f7f
2 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -26,7 +26,7 @@ pip install requirements.txt
## Usage
### Prepare Dataset
The cifar10 dataset can be downloaded automatically by torchvision. But the imagenet dataset should be downloaded manually and put in the right place, refer to [dataset.py](https://github.com/chunbaobao/Deep-JSCC-PyTorch/blob/main/dataset.py#L28). And run:
The cifar10 dataset can be downloaded automatically by torchvision. But the imagenet dataset should be downloaded manually from [ImageNet website](https://image-net.org/) and put in the right place, refer to [dataset.py](https://github.com/chunbaobao/Deep-JSCC-PyTorch/blob/main/dataset.py#L28). And run:
```
python dataset.py
```

5
dataset.py
View File

@ -23,9 +23,10 @@ class Vanilla(Dataset):
def main():
data_path = './Dataset'
os.makedirs(data_path, exist_ok=True)
# ILSVRC2012_img_train.tar and ILSVRC2012_img_val.tar should be downloaded from https://image-net.org/
if not os.path.exists('./Dataset/ILSVRC2012_img_train.tar') or not os.path.exists('./Dataset/ILSVRC2012_img_val.tar'):
print('Please download the dataset from http://www.image-net.org/challenges/LSVRC/2012/downloads and put it in ./Dataset')
print('ILSVRC2012_img_train.tar and ILSVRC2012_img_val.tar should be downloaded from https://image-net.org/')
print('Please download the dataset from https://image-net.org/challenges/LSVRC/2012/2012-downloads and put it in ./Dataset')
raise Exception('not find dataset')
phases = ['train', 'val']
for phase in phases: