Empowering developers with extensions for optimized ML experiences
SiMa.ai, a software-centric embedded edge machine learning system-on-chip company, announced the latest release of its Palette™ SDK (version 1.3), which introduces several enhancements designed to streamline the machine learning (ML) developer experience. This release marks a major step forward in making ML application development more accessible and efficient for developers worldwide.
Palette is a low-code command line environment designed for ML application development on SiMa.ai's Machine Learning System-on-Chip (MLSoC) silicon. It supports the entire application pipeline from creation to deployment in minutes using Python scripts. Integrated caching facilitates automatic partitioning and compilation across the MLA and quad-core Arm subsystems. Additionally, C/C++ APIs allow developers to integrate C/C++ host applications, libraries, or functions to rapidly achieve a cohesive production environment.
Palette 1.3 enhances SiMa.ai's C++ API to improve integration of the MLSoC platform into existing C++ applications. New features include acceleration pipeline status updates and improved error reporting, providing better visibility for tracing and debugging, and simplifying porting of C++ applications with HOST+GPU/Accelerators.
Palette 1.3 also introduces int16 quantization to improve the accuracy of models that are difficult to quantize. This feature balances accuracy and computational efficiency, making it ideal for applications with moderate memory and computational constraints. Developers can start with int8 quantization, evaluate the accuracy using different calibration methods, switch to int16 quantization when the int8 metric is insufficient, and fine-tune the calibration parameters for optimal performance.
Krishna Rangasayee, CEO and Founder, SiMa.ai, commented, “At SiMa.ai, our goal is to support developers at every stage of ML. We are continuously adding new features, scripts, and models to ensure a seamless, hassle-free developer experience. With Palette 1.3, we are excited to further empower developers in accelerating their application pipelines at the edge.”
Continuing our work to expand and optimize model support, Palette 1.3 now includes MaskRCNN and YOLOv8 4 camera support over Ethernet. This addition joins over 350 models that are already fully compatible with SiMa.ai MLSoCs.
About SiMa.ai
SiMa.ai is a software-centric embedded edge machine learning system-on-chip (MLSoC) company. SiMa.ai's hardware-to-software stack is flexible to adapt to any framework, network, model, sensor, and modality on one platform. Edge ML applications running fully on SiMa.ai MLSoCs experience 10x performance and energy efficiency, bringing high-fidelity intelligence to ML use cases from computer vision to generative AI in minutes. SiMa.ai enables customers to unlock new paths to revenue, realize significant cost savings, and innovate at the edge in industrial manufacturing, retail, aerospace and defense, agriculture, and healthcare. Founded in 2018, SiMa.ai has raised $270M and is backed by Fidelity Management & Research Company, Maverick Capital, Point72, MSD Partners, VentureTech Alliance, and others.
