# rvc4

Python API: `modelconverter.platforms.rvc4`

RVC4 conversion backend.

Implements the RVC4 side of modelconverter: exporting an ONNX or TFLite model to a `.dlc` file with the SNPE tools
(`snpe-onnx-to-dlc` or `snpe-tflite-to-dlc`, then `snpe-dlc-quant` and `snpe-dlc-graph-prepare`), plus inference, benchmarking,
analysis and visualization of the converted model. The classes in this package are picked by platform through the getters in
[modelconverter.platforms](https://docs.luxonis.com/software-v3/ai-inference/conversion/rvc-conversion/offline/modelconverter/modelconverter-api-reference/platforms.md).

> **Note**
> The export runs inside the RVC4 Docker image, which ships the SNPE tools. The host-run `analyze` command calls `snpe-dlc-info` and `snpe-diagview` directly, so it needs them on the host `PATH`. Benchmarking and analysis additionally need a real RVC4 device, reachable over ADB or SSH.

## Child Pages

 * [analyze](https://docs.luxonis.com/software-v3/ai-inference/conversion/rvc-conversion/offline/modelconverter/modelconverter-api-reference/platforms/rvc4/analyze.md)
 * [benchmark](https://docs.luxonis.com/software-v3/ai-inference/conversion/rvc-conversion/offline/modelconverter/modelconverter-api-reference/platforms/rvc4/benchmark.md)
 * [exporter](https://docs.luxonis.com/software-v3/ai-inference/conversion/rvc-conversion/offline/modelconverter/modelconverter-api-reference/platforms/rvc4/exporter.md)
 * [inferer](https://docs.luxonis.com/software-v3/ai-inference/conversion/rvc-conversion/offline/modelconverter/modelconverter-api-reference/platforms/rvc4/inferer.md)
 * [utils](https://docs.luxonis.com/software-v3/ai-inference/conversion/rvc-conversion/offline/modelconverter/modelconverter-api-reference/platforms/rvc4/utils.md)
 * [visualize](https://docs.luxonis.com/software-v3/ai-inference/conversion/rvc-conversion/offline/modelconverter/modelconverter-api-reference/platforms/rvc4/visualize.md)