Features & Scope of support

NetsPresso Model Launcher (beta)

Accelerated models, simplified deployment

Key Feature

Quantization

• Quantization enables to run and accelerate an AI model on the target device.

Convert and package

• Convert and compile AI models so that hardware can understand and run them.
• Package AI models with processing codes to be ready for the deployment

Device farm

• Actual hardwares are already installed in NetsPresso Device Farm to provide hardware-aware features.

Workflow

Scope of support

Converting options

• Models built with Model Searcher → TensorRT, TensorFlow Lite, OpenVINO
• ONNX → TensorRT, TensorFlow Lite, OpenVINO (Coming soon)
• TensorFlow → TensorFlow Lite (Coming soon)

Package formats

• Python wheel
• Packaging is available for models converted with Model Launcher

Hardware

• NVIDIA Jetson family (Nano, TX2, Xavier NX, AGX Xavier, AGX Orin)
• Raspberry Pi series (3B+, 4B)
• Intel Xeon W-2223