"The process of incorporating machine learning into an application often involves a team of experts tuning and tinkering for months with inconsistent setups. Businesses and developers want an end-to-end, development to production pipeline for machine learning"
This seems to be a complete pipeline, including deployment to production. Edit: DataRobot too, I misread. Edit2: DataRobot seems to offer a "codeless" approach. Thanks for sharing, I'll check it out.
I was wondering at the difference with Microsoft Azure Machine Learning Studio (which I haven't used yet).
If anyone from AWS is in this forum, could you comment if the custom Docker training in sagemaker can also be used for general optimisation of any dockerised objective function, e.g. bayesian hyper parameter training?
In the blog post example there is this python code:
Would I also write some kind of similar function for scoring the result of the training?
To provide some context, I work in bioinformatics where some of our algorithms have 100s of parameters. This is not ML where we want to classify or predict but rather optimise the parameters for a given objective function. If sagemaker allows general optimisation in an AWS lambda like way, that would be very useful.
There are no restrictions on the types of algorithms that you can optimize using the HyperParameterOptimization service.
SageMaker is designed for machine learning which means it's optimized for algorithms that process a lot of data to develop a model where each run of the algorithm may generate an objective function value (or potentially many such as the value may change during training).
If this structure fits your problem, SageMaker could be useful to you even if your problem isn't strictly "machine learning."
The hyperparameter optimization feature is still in preview (though the rest of SageMaker is in general availability). We'll be putting up a page within the next week or so for you to request access.
9 comments
[ 0.20 ms ] story [ 25.5 ms ] thread{1G}
Human Druid - Sage
{G}, Tap: Create 0/1 Plant Token named Seed of Knowledge
Sacrifice {X} Plants: Look at the top X cards of opponent's library
1/1
This seems to be a complete pipeline, including deployment to production. Edit: DataRobot too, I misread. Edit2: DataRobot seems to offer a "codeless" approach. Thanks for sharing, I'll check it out.
I was wondering at the difference with Microsoft Azure Machine Learning Studio (which I haven't used yet).
In the blog post example there is this python code:
Would I also write some kind of similar function for scoring the result of the training?To provide some context, I work in bioinformatics where some of our algorithms have 100s of parameters. This is not ML where we want to classify or predict but rather optimise the parameters for a given objective function. If sagemaker allows general optimisation in an AWS lambda like way, that would be very useful.
There are no restrictions on the types of algorithms that you can optimize using the HyperParameterOptimization service.
SageMaker is designed for machine learning which means it's optimized for algorithms that process a lot of data to develop a model where each run of the algorithm may generate an objective function value (or potentially many such as the value may change during training).
If this structure fits your problem, SageMaker could be useful to you even if your problem isn't strictly "machine learning."
I took a quick search on the documentation and I could see anything on a cursory pass:
http://docs.aws.amazon.com/search/doc-search.html?searchPath...
The hyperparameter optimization feature is still in preview (though the rest of SageMaker is in general availability). We'll be putting up a page within the next week or so for you to request access.