The DSVM is currently missing a proper environment to develop enterprise applications for Machine Learning and Deep Learning with Java. Despite of the fact that all the Python frameworks are the best tools for modeling and research, to get ready for production the backend in most case is more suitable to be C++, C# or Java, to address performances, multi-threading and backward interoperability with enterprise systems (distributed or not).
So I would suggest to add Java frameworks for Machine and Deep Learning like

• DeepLearning4J (see benchmarks here: https://github.com/deeplearning4j/dl4j-benchmark/blob/master/README.md)


• Weka: this is a powerful research tool, well known and used in the Research world, that provides powerful tools for clustering and predictive modeling.


• ELKI: another well known application for data clustering that covers the totality of known classic unsupervises clustering techniques most of them used today for data pre-processing and visualization;


• other libraries are Apache Spark Mlib and H2O;

Java BLAS libraries like
- ND4J

I will suggest also to provide GPU support for ND4J. A possibile approach would be to use isolated docker containers with nvidia-docker, that works pretty much fast and can handle multiple and distributed gpus as well.

Conda makes it easy to build custom virtual environments that contain only the packages required for a particular use case. However, in order to leverage Azure services from within Jupyter Notebooks when running a kernel based on a custom environment, there are quite a few Azure-specific packages that should/could be installed as well.

Presently, when I develop a custom environment, I start by cloning the 'base' environment that you include, which already contains the Azure-specific packages that I need, and then I add the additional packages I need. However this 'base' environment is built for Python 3.7 only. For use cases that require other minor versions of Python, I can start from scratch, but then I'm missing the Azure-specific packagage and so cannot leverage other Azure services.

Please publish a set of requirements.txt files that list of Azure specific packages and versions that are compatible with the various minor versions of Python. This would assist users like me in creating custom Conda environments that are capable of leveraging Azure services.

The DSVM is currently missing a proper environment to develop enterprise applications for Machine Learning and Deep Learning with Java. Despite of the fact that all the Python frameworks are the best tools for modeling and research, to get ready for production the backend in most case is more suitable to be C++, C# or Java, to address performances, multi-threading and backward interoperability with enterprise systems (distributed or not).
So I would suggest to add Java frameworks for Machine and Deep Learning like

• DeepLearning4J (see benchmarks here: https://github.com/deeplearning4j/dl4j-benchmark/blob/master/README.md)


• Weka: this is a powerful research tool, well known and used in the Research world, that provides powerful tools for clustering and predictive modeling.


• ELKI: another well known application for data clustering that covers the totality of known classic unsupervises clustering techniques most of them used today for data pre-processing and visualization;


• other libraries are Apache Spark Mlib and H2O;

Java BLAS libraries like
- ND4J

I will suggest also to provide GPU support for ND4J. A possibile approach would be to use isolated docker containers with nvidia-docker, that works pretty much fast and can handle multiple and distributed gpus as well.