Use cloud storage for models and training data
Open the Amazon S3 Console https://console.aws.amazon.com/s3/
S3 console appears as
The window contains buckets available and their information.
There is a possibility of having the old console of S3. If so, there will be a dialogue box in the window as shown below.
Switch to the improved version by clicking on Switch to the new console.
An s3 bucket can be created using Create bucket option.
Name the bucket. An S3 bucket name should be unique in each region.
The bucket name should be
Start with a lowercase letter or number.
For more information, see Rules for S3 bucket naming.
S3 bucket resides in the selected region. Select the Frankfurt (EU (Frankfurt) eu-central-1) region since SageMaker notebook instance can only access the S3 buckets in the same region.
Block public access
It is better to block public access to an S3 bucket unless it is created for public use. By blocking the public access only the user has access to the bucket data.
Bucket versioning, tag, and encryption (optional)
Use the default settings.
Object lock can be enabled only when you create a new bucket. Once enabled, it cannot be disabled.
For more information, see Object lock - S3 bucket
Finally, Create Bucket.
The newly created bucket appears in the Amazon S3.
Clicking the bucket name the following window. (Amazon S3 → ****).
Overview contains the contents in the bucket. Creation of new folder, deletion of files and modifications of files are performed in Overview.
To upload files to S3 bucket either use the Upload icon or drag the files to S3 bucket.
Upload → Add files → ****
The properties of a file in an S3 bucket can be obtained by selecting the file.
Modify a file/folder in S3 bucket by Select file → Actions S3 bucket
In order to delete or empty the bucket, select the bucket in AWS S3. Then Empty, Delete options are visible above.