Data and machine-learning analytics are becoming more widespread, but they grow in complexity with larger datasets requiring much time for configuration. Researchers spend less time actually in data science than getting their systems up to date which can prove difficult at times.
IBM has open-sourced CodeFlare, a machine learning framework that will allow developers to train their models more efficiently onto the hybrid cloud. This new framework is an exciting concept for those who are looking to simplify their workflow and shorten the time it takes. The idea behind this design is that when users have 10,000 work pipelines running, they wait up to 4 hours before receiving a result. While using this new framework, its implementation into these machines will require only 15 minutes.
To optimize machine learning models, developers must first train and clean the data. CodeFlare aims to unify a multi-platform workflow process for data scientists, so they don’t need to learn new languages when accessing different frameworks or platforms.
CodeFlare pipelines will run with ease on IBM’s new serverless platform, the IBM Cloud Code Engine. It allows users to deploy it just about anywhere and extend the benefits of serverless technology for data scientists and AI researchers alike. With its new features, it will be easier to integrate and bridge with other cloud-native ecosystems. You can now use event triggers such as the arrival of a new file or load data from sources like cloud object storage, data lakes, and distributed filesystems.
IBM hopes to take the complexity out of data science by providing a platform that allows innovation and focus. Data scientists will utilize richer tools with more consistency, allowing them to delve deeper into their research while IBM takes care of configuration and deployment challenges.