Amazon Sagemaker Ground Truth Reviews & Product Details

From a user's perspective, almost all the features provided by Amazon Machine Learning services are awesome, In that list if we have a look at SageMaker Ground Truth it reduces the significant cost of running a Machine Learning model in the cloud by its advanced automatic labeling capacity. Also, it reduces human interaction when it comes to data labeling by a significant amount. Review collected by and hosted on G2.com.

As per our observation while using SageMaker Ground Truth is, it performs significantly well than any other ML Solution. So nothing to complain about at all. Review collected by and hosted on G2.com.

Very easy to use and deploy the Machine learning models and API managment is very easy. Very easy to connect to Docker. Review collected by and hosted on G2.com.

Pricing is a little on the high side when running high amount of labels. Endpoint cannot be turned off and thus wastes memory. Review collected by and hosted on G2.com.

Great understable UX for the labeler, with time restrictions and being able to pick up wherever left off Review collected by and hosted on G2.com.

The labelers cannot track how many objects are pending in the project if they don't have access to the AWS console. Hard to match PII to Labeler ID to track per labeler contribution/progress. Review collected by and hosted on G2.com.

Easy with which we can deploy and monitor our machine learning models using sagemaker endpoint. Review collected by and hosted on G2.com.

Sagemaker endpoint cannot be turned off, thus consumes unncessary resource and money at times. Review collected by and hosted on G2.com.

I enjoyed the ease of use of the product. Good amount of options; I like how it supports Docker containers and API management. It is very easy to set up and integrate with other services. Also very nice that it supports multiple frameworks. Probably the best tool on the market right now used for machine learning. Review collected by and hosted on G2.com.

When running the product on a large amount of labels; the price gets really steep. It takes a while to set up and figure out on your own when you are not used to Amazon products. The integration with AWS can take a while to get used to if you are trying to use other products other than AWS. Very limited options when choosing something other than the AWS platform. Review collected by and hosted on G2.com.

AWS SageMaker Ground Truth service is the best and only solution about it. It uses AI service to predict lables from data.

Service, doesn't promise to solve all labels. Unable to labled data can use for crowdcourcing. As in Amazon Turk service you can assign your data to people in all around the world who try to predict labels.

SageMaker Ground Truth can lower your labeling costs by up to 70% using automatic labeling. Review collected by and hosted on G2.com.

Because of it is very new and unique service we can not say any negative feedback Review collected by and hosted on G2.com.

This is a great tool that you can use to organize data for the task of machine learning. If you have a lot of data and don't have the necessary tools or computing power to add all of the data; then AWS SageMaker does a great job in allowing you to do so. I use this on a daily basis at the moment to complete daily tasks. Review collected by and hosted on G2.com.

One thing I do not like is the pricing of the tool. Once you have a lot of data and tons of objects to input; the pricing can go up substantially. It is affordable; but for a small business can be a bit expensive once you start dealing with vast amount of data. Review collected by and hosted on G2.com.

This tool resolve one of the most popular problems into data mining and machine learning. It allows users to tag every data set content like audio, databases with large texts, images. This tasks make life much more easy for a data scientist. Review collected by and hosted on G2.com.

The configuration is to hard and confuse, the time to set up the tool working on production and the price are the things that i dislike. For a large data the tool is a bit slow, but is understandable. Review collected by and hosted on G2.com.

Amazon Web Services, the cloud provider service of Amazon, announced the release of AWS SageMaker Ground Truth during the 2019 AWS Re:Invent and with this introduction has moved further in the AI and Machine learning space.

What this service promises to do is help in data labelling - thus reducing cost and time

I like this service because:

1. Traditionally Machine learning requires good size data set which one has to label or acquire. Both are time taking processes. With AWS SageMaker Ground Truth, it will become easy to give the unlabelled dataset with small labelled dataset and it will do the rest. No more using human workers and explain them the process, add their bias into the process. I have done this task before and I know it's not an easy one.

2. Cost: Cost is another one of the issues especially when it comes to hiring those workers for labelling.

This service can build datasets for text, image, object detection etc. This can optionally use active learning to automatically label the data as well. Review collected by and hosted on G2.com.

This is a new service and not many tutorials are there on the website. I had to learn myself using AWS blogposts and self learning so there is a high learning curve.

I don't how much accurate this service might be. Review collected by and hosted on G2.com.

Most of the time is spent labeling data used for training while creating a machine learning model but AWS sage maker cloud truth makes it easier to label this data and use them for training the model. The time taken to label the data reduces significantly. Review collected by and hosted on G2.com.

It can be hard to use for people who do not have any machine learning experience. After the data is labeled by AWS SageMaker Ground Truth it has to be validated that the data has been labeled correctly. Some of the labels is inaccurate and they have to be corrected. This might take more time than expected sometimes. Review collected by and hosted on G2.com.