I have about 35 gb of photos in my amazon cloud drive. I'm trying to run a convolutional neural network using the Deep Learning Linux EC2 AMI on AWS. Is there a way I could use the images in my amazon cloud drive for this purposes? Maybe access them in my python script or something?
Or is there another way of storing the 35 gb of data to be able to be used with AWS?
Interesting question and I could visualize use-case for the same. Note that S3 is in ideal location for enterprise use-cases while Amazon Drive is alternative of Google drive and is more towards customer focussed a b2c solution... Primary purpose of Amazon Drive is to let photos/documents be in sync across different devices of customer and secondly, auto upload of photos taken.
Having said that, ofcourse u can use S3 to store your images and then you have many sdks in all languages through which your ec2 can interact with s3.
But if you want to use amazon drive; you can refer https://developer.amazon.com/public/apis/experience/cloud-drive/content/restful-api
Note that as S3 is focussed towards enterprise; you have loads of options and easy apis as compared to with amazon drive.. You can also try uploading images to Google drive.. they have much better apis and your ec2 can talk to google drive apis.
You cannot access Amazon Cloud Drive. Amazon does not provide public access to its only API. Amazon Cloud Drive does not have WebDav or (S)FTP, the only supported API is poorly documented REST API. And Amazon does not authorize API keys anymore for at least a year.
Even if you got whitelisted access key year ago or earlier, API and service itself are so awkward and buggy that you get more problems than solve them. Constant TooManyRequests errors even if you didnt send any request for whole day, wierd undocumented errors.
