datasets.md
June 27, 2022 ยท View on GitHub
Instructions for downloading datasets (modified from https://github.com/mlfoundations/wise-ft)
Step 1: Download
export DATA_LOCATION=~/data # feel free to change.
cd $DATA_LOCATION
ImageNet-A
wget https://people.eecs.berkeley.edu/~hendrycks/imagenet-a.tar
tar xvf imagenet-a.tar
rm imagenet-a.tar
ImageNet-R
wget https://people.eecs.berkeley.edu/~hendrycks/imagenet-r.tar
tar xvf imagenet-r.tar
rm imagenet-r.tar
ImageNet Sketch
Download links:
- from Google Drive
- from Kaggle
ImageNet V2
wget https://s3-us-west-2.amazonaws.com/imagenetv2public/imagenetv2-matched-frequency.tar.gz
tar -xvf imagenetv2-matched-frequency.tar.gz
rm imagenetv2-matched-frequency.tar.gz
ObjectNet
wget https://objectnet.dev/downloads/objectnet-1.0.zip
unzip objectnet-1.0.zip
rm objectnet-1.0.zip
ImageNet
Can be downloaded via https://www.image-net.org/download.php. Please format for PyTorch, e.g., via https://github.com/pytorch/examples/blob/main/imagenet/extract_ILSVRC.sh.
Step 2: Check that datasts are downloaded
When running:
cd $DATA_LOCATION
ls
you should see (at least):
imagenet # containing train and val subfolders
imagenetv2-matched-frequency-format-val
imagenet-r
imagenet-a
sketch # imagenet-sketch
objectnet-1.0