The State of Machine Learning Frameworks in 2019

Re:The State of Machine Learning Frameworks in 2019 by oceansky on Saturday October 25, 2025 @10:56AM (#45704393)

In 2019 I delivered a instance segmentation project and I used Mask RCNN and tensorflow.

Nowadays it looks like yolo absolutely dominates this segment. Any data scientists can chime in?

[ Parent | on HN ]

Re:The State of Machine Learning Frameworks in 2019 by bonoboTP on Saturday October 25, 2025 @12:08PM (#45704921)

SAM (Segment Anything Model) by Meta is a popular go-to choice for off the shelf segmentation.

But the exciting new research is moving beyond the narrow task of segmentation. It's not just about having new models that get better scores but building larger multimodal systems, broader task definitions etc.

[ Parent | on HN ]

Re:The State of Machine Learning Frameworks in 2019 by deepsquirrelnet on Saturday October 25, 2025 @11:52AM (#45704790)

I haven’t used RCNN, but trained a custom YOLOv5 model maybe 3-4 years ago and was very happy with the results.

I think people have continued to work on it. There’s no single lab or developer, it mostly appears that the metrics for comparison are usually focused on the speed/MAP plane.

One nice thing is that even with modest hardware, it’s low enough latency to process video in real time.

[ Parent | on HN ]