import sagemaker | |
from sagemaker.tensorflow import TensorFlow | |
from sagemaker import Session | |
sess = sagemaker.Session() | |
role = sagemaker.get_execution_role() | |
# Create the apropriate folder struture, no need to be the same as I use | |
train_dir = 'data/train' | |
test_dir = 'data/test' | |
np.save(os.path.join(train_dir, 'train_images.npy'), train_images) | |
np.save(os.path.join(train_dir, 'train_labels.npy'), train_labels) | |
np.save(os.path.join(test_dir, 'test_images.npy'), test_images) | |
np.save(os.path.join(test_dir, 'test_labels.npy'), test_labels) | |
print(f"Sagemaker bucket: {sess.default_bucket()}") | |
s3_prefix = 'fashion-mnist' | |
traindata_s3_prefix = '{}/data/train'.format(s3_prefix) | |
testdata_s3_prefix = '{}/data/test'.format(s3_prefix) | |
train_s3 = sagemaker.Session().upload_data(path='./data/train/', key_prefix=traindata_s3_prefix) | |
test_s3 = sagemaker.Session().upload_data(path='./data/test/', key_prefix=testdata_s3_prefix) | |
inputs = {'train':train_s3, 'test': test_s3} | |
print(inputs) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment