Improve the adaptability of your pipeline
In our previous blogs on pipeline operators, we saw how we could use the pipeline operators to speed up our inference, to add unit tests to the input of our machine learning model and to conditionally routing the pipeline request to one way or the other. With UbiOps latest release, v2.22.0, we introduced a new […]
UbiOps Operators #2 Testing and conditional post processing
Exciting things you can build with the new UbiOps operators! In our first blog on pipeline operators, we used them to speed up a video processing pipeline. Specifically, we used the create subrequests and the collect subrequests operators to analyze multiple frames of the video simultaneously, by introducing parallelization. This time we will highlight three […]
Data pipelines: what, why and which ones
It can be quite confusing keeping track of what all these different pipelines are and how they differ from one another. In this article, we will map out and compare a few common pipelines, as well as clarify where UbiOps pipelines fit in the general picture.
Combining R and Python in the same pipeline: the prediction of house prices
Combining R and Python in the same pipeline While hundreds of programming languages exist, Python and R remain the most popular ones to use in the world of data science. R is a great language to make visualizations and graphs, furthermore, it has many functionalities for data analysis. Python is a general-purpose language that is […]
How to create a data pipeline in UbiOps?
(2020 version) In a previous article I explained how to deploy a single data science deployment onto UbiOps that could be used for object recognition. However, in the real world, it is often not that simple. In reality, you usually don’t have a single model with perfect input and output. Rather, there is a chain […]