Sber.DS - a platform that allows you to create and implement models even without code

Ideas and meetings about what other processes can be automated arise daily in a business of various sizes. But besides the fact that a lot of time can be spent on creating a model, you need to spend it on evaluating it and checking that the result is not random. After implementation, any model must be monitored and periodically checked.


And these are all the stages that you need to go through in any company, regardless of its size. If we talk about the scale and legacy of Sberbank, the number of subtle settings increases significantly. By the end of 2019, more than 2,000 models were already used in Sberbank. It is not enough just to develop a model, it is necessary to integrate with industrial systems, develop data marts for building models, and provide control over its operation on the cluster.





Sber.DS. , , , .


, , โ€” , , Sber.DS. .



Sber.DS , , .




, . , , . .


. . .


. , โ€” (). . , , , , .


, . . .


, . Jupyter Kernel Gateway , ยซ ยป.




Sber.DS . , . , , . REST API. .


, : 4-8 . REST API (Open API). , , API , .


Java Spring Framework. , Red Hat OpenShift (Kubernetes). , ( , AutoML), .


ยซยป , , , . : Hadoop, โ€” OpenShift (Docker). , on-premise . , . โ€“ ยซ ยป .


, - , Python Hadoop , , environment . C/C++ , Python , . , .


, . , . Hadoop Cloudera parcel. Hadoop docker-. python.eggs.


, Linux, , Linux namespace, , , , . . , , .




MVP , Python/R/Java Hadoop. environment Hadoop, .


, , DS- , , , data- . Spark-. , - dataset.


: Linux DevOps, Hadoop Spark, Java Spring, Scala Akka, OpenShift Kubernetes. , , , .


All Articles