when i run the command "python3 -m vit_jax.train --name ViT-B_16-cifar10_`date +%F_%H%M%S` --model ViT-B_16 --logdir /tmp/vit_logs --dataset cifar10",I encountered the error"Error executing an HTTP request: libcurl code 6 meaning 'Couldn't resolve host name', error details: Couldn't resolve host 'metadata'Error executing an HTTP request: libcurl code 6 meaning 'Couldn't resolve host name', error details: Couldn't resolve host 'metadata'"."." about vision_transformer HOT 3 OPEN

Maybe you run into problems connecting to a server for downloading the cifar10 dataset? If that is the reason the the command python -c 'import tensorflow_datasets as tfds; tfds.builder("cifar10").download_and_prepare()' should result in the same error message. Make sure to have internet connection from the computer where you're running the code, or prepare the dataset on another computer and then copy the ~/tensorflow_datasets directory.

from vision_transformer.

Maybe you run into problems connecting to a server for downloading the cifar10 dataset? If that is the reason the the command python -c 'import tensorflow_datasets as tfds; tfds.builder("cifar10").download_and_prepare()' should result in the same error message. Make sure to have internet connection from the computer where you're running the code, or prepare the dataset on another computer and then copy the ~/tensorflow_datasets directory.

I have prepared cifar10 in local, but have no idea to prevent the program connecting tensorflow_datasets. Does any args tell the program getting data from local? Or any operation to function get_data of input_pipeline.py can achieve it ?

from vision_transformer.

I met exactly the same problem, and I found it is a mistake of version. I replace the version to tensorflow-datasets=2.1.0 and tensorflow=2.3.0. Problem solved!

from vision_transformer.