-
sparseml
Libraries for applying sparsification recipes to neural networks with a few lines of code, enabling faster and smaller models
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
We’ve noticed YOLOv3 runs pretty slowly on CPUs restricting its use for real-time requests. Given that, we looked into combining pruning and quantization using the Ultralytics YOLOv3 model, and the results turned out well, over 5X faster over a dense FP32 baseline! We open sourced the integration and models on GitHub for anyone to play around with; if you’re interested, please check it out and give us feedback.