Hi,
How can I use model_maker with sagemaker?
I do’nt manage to retrieve the data with DataLoader and I wonder if there is another way to work with s3?
Thanks
Hello @user50_1
Thank you for using TensorFlow,
In this case, using aws CLI we can copy dataset to s3 bucket from any machine,
aws s3 cp local-data-path s3://your-bucket-name/path/to/data/ --recursive
once the dataset is entirely available on s3 bucket, we can setup data in sagemaker using
s3_input_train = sagemaker.inputs.TrainingInput(
s3_data='s3://your-bucket-name/path/to/data/train/',
content_type='application/x-image'
)
Then we can continue to load data with s3fs package, continue with training using your logic and models.