Amazon SageMaker

Machine Learning at scale by deploying huge amount of training data

Accelerated data processing for faster outputs and learnings

Kubernetes integration for containerized deployments

Creating API endpoints for use by technical users

The UI can be eased up a bit for use by business analysts and non technical users

For huge amount of data pull from legacy solutions, the platform lags a bit

Considering ML is an emerging topic and would be used by most of the organizations in future, the pipeline integrations can be optimized

Provided an instance of Jupyter notebook for development script, which made it very easy to manage and develop any script.

Our system is cloud-based, and we are charged only for what we use and how long we use it.

We can choose multiple servers for Training, without any headache of distribution.

Most of the libraries are supported.

All training, testing, and models are stored on S3, so it's very easy to access whenever we require it.

It's very good for the hardcore programmer, but a little bit complex for a data scientist or new hire who does not have a strong programming background.

Most of the popular library and ML frameworks are there, but we still have to depend on them for new releases.

Amazon SageMaker is great for visually seeing the development of machine learning models. The process is set up in a logical step-by-step process.

Amazon SageMaker makes training data models about as easy as it gets. It's straight-forward to construct training and test samples.

Amazon SageMaker makes deploying machine learning models much easier than other open-source tools.

Amazon SageMaker is a great tool for a data scientist, although surprisingly, comparing different machine learning models with SageMaker is not as easy as one would think. I think Amazon needs to team up with a data scientist who does ensemble modeling.

Because SageMaker is targeted for machine learning models, other models a data scientist might use require more effort to get them incorporated. My guess is Amazon is moving to make SageMaker a more complete tool.

SageMaker can take a long time to run on larger data sets. That's the case with every big data science tool I've used, but SageMaker doesn't seem to be as quick as other tools.

SageMaker is useful as a managed Jupyter notebook server. Using the notebook instances' IAM roles to grant access to private S3 buckets and other AWS resources is great. Using SageMaker's lifecycle scripts and AWS Secrets Manager to inject connection strings and other secrets is great.

SageMaker is good at serving models. The interface it provides is often clunky, but a managed, auto-scaling model server is powerful.

SageMaker is opinionated about versioning machine learning models and useful if you agree with its opinions.

SageMaker does not allow you to schedule training jobs.

SageMaker does not provide a mechanism for easily tracking metrics logged during training.

We often fit feature extraction and model pipelines. We can inject the model artifacts into AWS-provided containers, but we cannot inject the feature extractors. We could provide our own container to SageMaker instead, but this is tantamount to serving the model ourselves.

Provides the basis for developing algorithms and data without going very deep into the actual development.

Amazon software and so can be used with other Amazon software your organization already uses.

Training and on boarding of the software and customer service was great to work with.

Searching and descriptions can be easier to read and interpret.

Training modules and customer service training representative could make on boarding employees easier.

Provides enough freedom for experienced data scientists and also for those who just need things done without going much deeper into building models.

Customization and easy to alter and change.

If you already are an Amazon user, you do not need to transition over to another software.

I think that although the algorithms are there and you are using one click, there could be more detailed descriptions located in places so that other users are able to easily find the right formula and tools.

Mobile friendly options would be a huge plus, even tracking what employees are using this tool for in regards to reporting.