Error 403, Problem with importing image data in s3 to SageMaker Notebook

0

I am working with SageMaker Notebook and image data in S3 bucket with name s3://<BucketName>/train/<class0-4> and validate data in other dir.
I create an IAM Role and put previous specific bucket, in the notebook I load this bucket with:

s3_train = 's3://<BucketName>/train'
train_data = sagemaker.session.s3_input(s3_train, distribution='FullyReplicated', 
                        content_type='application/x-image', s3_data_type='S3Prefix')

The same for train lst file, validation data and validation lst data, after create data channels with this:

data_channels = {'train': train_data, 'validation': validation_data, 
                 'train_lst': train_data_lst, 'validation_lst': validation_data_lst}

After create a TensorFlow estimator, and finally in fit pass the data with this:

tf_estimator.fit(inputs=data_channels, logs=True)

And return this menssage error:
An error occurred (403) when calling the HeadObject operation: Forbidden

Inntech
質問済み 4年前813ビュー
3回答
0
承認された回答

An empty bucket policy is fine. You should only make your objects public if you want everyone in the world to have access.

If you’re using a SageMaker notebook server, then when you created it you would have created (or used an existing) IAM role for it. You should check to see it’s S3 permissions, and make sure that it has access to the bucket in question.

Hope that helps.

回答済み 4年前
0

Looks like an issue with a policy attached to your role. Do you want to paste it in here so we can take a look?

回答済み 4年前
0

The bucket policy its empty, change the bucket to public access, but the files aren't public, think that the directories are public and files not. What suggest put in bucket policy??

Inntech
回答済み 4年前

ログインしていません。 ログイン 回答を投稿する。

優れた回答とは、質問に明確に答え、建設的なフィードバックを提供し、質問者の専門分野におけるスキルの向上を促すものです。

質問に答えるためのガイドライン

関連するコンテンツ