Amazon SageMaker Reviews & Product Details

Sagemaker lets you build, train and deploy your models all in one place. I like how it lets you choose different machine types for each phase of development so no resources are wasted. Sagemaker also supports distributed training. I've used Sagemaker to build, train and deploy deep learning models using PyTorch, Tensorflow as well as Keras. It provides lots of custom conda environments to support development using any framework, like p36, p32 etc. Sagemaker instances comes preloaded with some sample notebooks. There's also a large repository of materials available to help you get started. Review collected by and hosted on G2.com.

It's bit pricey and the UI doesn't exactly tell you if you've go unused instances or deployed models lying around unlike the GCP equivalent. Amazon sneaks up on you with unexpected bills if you aren't careful. It's a bit difficult to get started with as the whole user creation vs instance creation kinda trips you up. As soon as you click on preview, amazon directs you to the user creation part and you've a whole lot of unneccesary set up when all you probably need is to create a single notebook instance. Review collected by and hosted on G2.com.

Easy deployment of ML Model as HTTP endpoints. Review collected by and hosted on G2.com.

Not enough documentation available on latest features like Batch Transform Review collected by and hosted on G2.com.

you can Integration with other AWS ecosystem with very easily, one of the best features of Amazon SageMaker. Plus you will get charged based on your usage. it has jupyter notebook which facilitates developer efficiency and it is also very scalable and customization is also available at the fingertip. All Machine learning algorithms are very well defined and understandable and any machine learning engineer with naive experience can make it possible to work with Sagemaker. Most of the machine learning, as well as deep learning libraries, are supported. Review collected by and hosted on G2.com.

It demands good coding knowledge to use it. if you dint have any prior experience of coding or coding language you might feel hard to use it. features like drag and drop are missing which are available with other tools. on a larger dataset sometime it takes a longer time than expected. Review collected by and hosted on G2.com.

Best one is this is online machine learning platform very easy to understand and train developers, it is so easy to implement all our machine learning models, one can easily do with out much expertise Review collected by and hosted on G2.com.

I believe all good, may be some sort of UI experience need to improve. All are good only Review collected by and hosted on G2.com.

The overall simplicity it provides as a tool. Review collected by and hosted on G2.com.

Truth is i haven't found anything negative for now Review collected by and hosted on G2.com.

I like the concept of having an infrastructure where you can train the model and deploy to the public for seamlessly. Review collected by and hosted on G2.com.

I feel there could be more helpful resources available for the getting started that would help. Review collected by and hosted on G2.com.

I like the AWS environment and automated access to Jupyter notebooks. Review collected by and hosted on G2.com.

It restricts the way I am willing to do my work. Review collected by and hosted on G2.com.

I have worked with DeepLens by AWS for various face recognition application. It's very easy to use and configure. You don't need your own NVIDIA machine to train the model. If you have more data and need a powerful machine to train. Just upgrade your cloud and start training. Review collected by and hosted on G2.com.

Sometimes AWS doesn't say some of the configurations explicitly, you have to be familiar with AWS. There are very less good resources for that. Our company has some help from AWS marketing developer Review collected by and hosted on G2.com.

I love this platform is so simple because it allows me as a developer to create data in a simple way I do not have to work in a tedious way implementing algorithms because it gives me everything is so simple it saves me many hours of work because with it I can implement up to my own framework I work without any problem, I can also have notebooks and connect directly in S3 .

To be able to verify if the model I have generated I like I can do integrated A / B tests and in this way I achieve better results and success in what I look for . its simplicity has allowed me to work very efficiently and very quickly acquiring very complete and successful works and models Review collected by and hosted on G2.com.

for beginners who wish to use this platform the only disadvantage is that it requires very high payments because this platform implements the GPU computation to provide a better training of the GPU which makes them faster Review collected by and hosted on G2.com.

- Lots of Amazon examples of use cases where you will use this kind of tools

- The configuration for your environment is super easy

- The SDK is simple to use Review collected by and hosted on G2.com.

- Since it's "new," you won't find much about it on the Internet.

- I think it would be super helpful to have other languages in the Development Kit besides Sparks and Python (probably Ruby or Java or even JavaScript). Review collected by and hosted on G2.com.