Lattice sensAI Studio User Guide Application Note
■The Lattice sensAI™ Studio is a user interface-based tool for training, validating, and compiling Machine Learning (ML) models optimized for Lattice FPGAs. The tool makes it easy to take advantage of transfer learning to deploy ML models. It allows you to select from a range of models pre-trained to cover popular use cases, bringing your own data for additional training, validating the quality of training using Tensor Board, and compiling for Lattice FPGAs.
■The features of the Lattice sensAI Studio are:
▲Model Zoo with variety of models based on multiple architecture
▲Option to bring your own model and environment into framework
▲Easy to use dataset pre-processing such as labeling, annotation conversion, augmentation
▲Model training, freezing, TensorBoard
▲Different types of transfer learning
▲Storage management
▲Informative dashboard for better visualization
▲Support to compile model for various Lattice provided FPGAs
▲JupyterLab interface for code editing
▲User management and collaboration
▲DIY installation and setup with docker container on CPU/GPU PC/Server
■This document describes the installation/configuration, usage, and troubleshooting of the Lattice sensAI Studio
|
|
Application note & Design Guide |
|
|
|
Please see the document for details |
|
|
|
|
|
|
|
English Chinese Chinese and English Japanese |
|
May 2021 |
|
Revision 1.0 |
|
FPGA-AN-02039-1.0 |
|
3.8 MB |
- +1 Like
- Add to Favorites
Recommend
- Why GaN in Space?
- Innoscience Built The First World-class 8-inch Wafer FAB with Gold-free and CMOS Compatible Process
- Two kinds of structure of unit in SiC power MOSFET: plane structure and groove structure
- Successful Demonstration of Next-Gen Ka-band Rx Beamforming IC Portfolio with 1.5dB NF LNA
- Futureway Has Developed a Variety of Silicone Solutions for Battery Pack
- UnitedSiC Rediscovers The Perfect Switch with SiC FETs
- These Kinds of MOS Tubes “Breakdown“, How Many Do You Know?
- Low-Frequency Noise Modeling When Quantum Chips Are Getting Noiser
All reproduced articles on this site are for the purpose of conveying more information and clearly indicate the source. If media or individuals who do not want to be reproduced can contact us, which will be deleted.