ArangoDB has announced a new machine learning offering. The ArangoML Pipeline Cloud is a fully-hosted, fully-managed common metadata layer for production-grade data science and ML platforms that runs on ArangoDB Oasis.
“ArangoML Pipeline Cloud meets the needs of both data scientists, who are concerned with the quality of the data, feature training, and model results, as well as DevOps, who need to manage which datasets and deployments are in use, their performance, and how they are being deployed,” ArangoDB wrote in a post.
ArangoDB can easily accommodate and unite unstructured, highly-interlinked data, such as inference and model descriptions and allow relationships between them to be stored as a graph that can be managed by a DevOps engineer and be used by a data scientist at the same time, the company explained.
Hyperledger Fabric 2.0 released
Hyperledger Fabric 2.0’s improvements were focused on new ways to manage the chaincode (or smart contract) life cycle to maximize flexibility, remove bottlenecks and create more options to distribute governance.
Security and data privacy were also big development priorities, so many features and enhancements are geared towards delivering more granular control of private channels and minimizing vulnerabilities, according to the team.
Key Hyperledger Fabric 2.0 features include decentralized governance for smart contracts, new chaincode application patterns for collaboration and consensus, an external chaincode launcher, private data enhancements, state database cache for improved performance on CouchDB, and Alpine-based docker images.
Pandas 1.0 announced
Pandas is a open-source library for easy-to-use Python data structures and data analysis. The new release removed a lot of functionality that was deprecated in previous releases and includes enhancements and new features.
Enhancements include the ability to use Numba in ‘rolling.apply’ and ‘expanding.apply,’ the ability to define custom windows for rolling operations, and creating markdown tables.
Experimental new features include NA scalar to denote missing values, a dedicated string data type, and Boolean data types with missing values support.
The full details are available here.
Stoplight closes $6 million series A funding round
API Design Management company Stoplight announced $6 million in a series A funding round, bringing the company’s total funding to $10 million.
Stoplight plans to utilize the funding to augment engineering efforts, accelerate its product roadmap and cover more customer use cases.
“API-first design is becoming a fundamental component of enterprise engineering, and we have full confidence that Stoplight will only deepen their category leadership in the next several years,” said Mike Smerklo, co-founder and managing director of Next Coast Ventures.
The post SD Times news digest: ArrangoML Pipeline Cloud, Hyperledger Fabric 2.0, and Pandas 1.0 appeared first on SD Times.