This package is a high-level Rust bindings for wasi-nn example of Mobilenet with Tensorflow Lite backend.

This crate depends on the wasi-nn in the Cargo.toml:

[dependencies]
wasi-nn = "0.4.0"

Compile the application to WebAssembly:

cd rust && cargo build --target=wasm32-wasi --release

The output WASM file will be at rust/target/wasm32-wasi/release/wasmedge-wasinn-example-tflite-bird-image.wasm. To speed up the image processing, we can enable the AOT mode in WasmEdge with:

wasmedge compile rust/target/wasm32-wasi/release/wasmedge-wasinn-example-tflite-bird-image.wasm wasmedge-wasinn-example-tflite-bird-image.wasm

The testing image is located at ./bird.jpg:

The tflite model is located at ./lite-model_aiy_vision_classifier_birds_V1_3.tflite

Users should install the WasmEdge with WASI-NN TensorFlow-Lite backend plug-in.

Execute the WASM with the wasmedge with Tensorflow Lite supporting:

wasmedge --dir .:. wasmedge-wasinn-example-tflite-bird-image.wasm lite-model_aiy_vision_classifier_birds_V1_3.tflite bird.jpg

You will get the output:

Read graph weights, size in bytes: 3561598
Loaded graph into wasi-nn with ID: 0
Created wasi-nn execution context with ID: 0
Read input tensor, size in bytes: 150528
Executed graph inference
   1.) [166](198)Aix galericulata
   2.) [158](2)Coccothraustes coccothraustes
   3.) [34](1)Gallus gallus domesticus
   4.) [778](1)Sitta europaea
   5.) [819](1)Anas platyrhynchos