Skip to content

jatinchowdhury18/RTNeural

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RTNeural

Tests Bench Examples RADSan codecov arXiv License

A lightweight neural network inferencing engine written in C++. This library was designed with the intention of being used in real-time systems, specifically real-time audio processing.

Currently supported layers:

  • Dense
  • GRU
  • LSTM
  • Conv1D
  • Conv2D
  • MaxPooling
  • BatchNorm1D
  • BatchNorm2D

Currently supported activations:

  • tanh
  • ReLU
  • Sigmoid
  • SoftMax
  • ELu
  • PReLU

Additional resources:

Citation

If you are using RTNeural as part of an academic work, please cite the library as follows:

@article{chowdhury2021rtneural,
        title={RTNeural: Fast Neural Inferencing for Real-Time Systems},
        author={Jatin Chowdhury},
        year={2021},
        journal={arXiv preprint arXiv:2106.03037}
}

How To Use

RTNeural is capable of taking a neural network that has already been trained, loading the weights from that network, and running inference. Some simple examples are available in the examples/ directory.

Exporting weights from a trained network

Neural networks are typically trained using Python libraries including Tensorflow or PyTorch. Once you have trained a neural network using one of these frameworks, you can "export" the network weights to a json file, so that RTNeural can read them. An implementation of the export process for a "sequential" Tensorflow model is provided in python/model_utils.py, and can be used as follows.

# import dependencies
import tensorflow as tf
from tensorflow import keras
from model_utils import save_model

# create Tensrflow model
model = keras.Sequential()
...

# train model
model.train()

# export model weights
save_model(model, 'model_weights.json')

For an example of exporting a model from PyTorch, see this example script.

Creating a model

Next, you can create an inferencing engine in C++ directly from the exported json file:

#include <RTNeural.h>
...
std::ifstream jsonStream("model_weights.json", std::ifstream::binary);
auto model = RTNeural::json_parser::parseJson<double>(jsonStream);

Running inference

Before running inference, it is recommended to "reset" the state of your model (if the model has state).

model->reset();

Then, you may run inference as follows:

double input[] = { 1.0, 0.5, -0.1 }; // set up input vector
double output = model->forward(input); // compute output

Compile-Time API

The code shown above will create the inferencing engine dynamically at run-time. If the model architecture is fixed at compile-time, it may be preferable to use RTNeural's API for defining an inferencing engine type at compile-time, which can significantly improve performance.

// define model type
RTNeural::ModelT<double, 8, 1
    RTNeural::DenseT<double, 8, 8>,
    RTNeural::TanhActivationT<double, 8>,
    RTNeural::DenseT<double, 8, 1>
> modelT;

// load model weights from json
std::ifstream jsonStream("model_weights.json", std::ifstream::binary);
modelT.parseJson(jsonStream);

modelT.reset(); // reset state

double input[] = { 1.0, 0.5, -0.1 }; // set up input vector
double output = modelT.forward(input); // compute output

Loading Layers from PyTorch

The above example code assumes that the trained model has been exported from TensorFlow. For loading PyTorch models, the RTNeural namespace RTNeural::torch_helpers, provides helper functions for loading layers exported from PyTorch.

// load model weights from json
std::ifstream jsonStream("model_weights.json", std::ifstream::binary);
nlohmann::json modelJson;
jsonStream >> modelJson;

// load a layer from a static model
RTNeural::ModelT<float, 1, 1, RTNeural::DenseT<float, 1, 1>> model;
RTNeural::torch_helpers::loadDense(modelJson, "name_of_layer.", model.get<0>());

For more examples, see the examples/torch directory.

Building with CMake

RTNeural is built with CMake, and the easiest way to link is to include RTNeural as a submodule:

...
add_subdirectory(RTNeural)
target_link_libraries(MyCMakeProject LINK_PUBLIC RTNeural)

If you are trying to use RTNeural in a project that does not use CMake, please see the instructions below.

Choosing a Backend

RTNeural supports three backends, Eigen, xsimd, or the C++ STL. You can choose your backend by passing either -DRTNEURAL_EIGEN=ON, -DRTNEURAL_XSIMD=ON, or -DRTNEURAL_STL=ON to your CMake configuration. By default, the Eigen backend will be used. Alternatively, you may select your choice of backends in your CMake configuration as follows:

set(RTNEURAL_XSIMD ON CACHE BOOL "Use RTNeural with this backend" FORCE)
add_subdirectory(modules/RTNeural)

In general, the Eigen backend typically has the best performance for larger networks, while smaller networks may perform better with XSIMD. However, it is recommended to measure the performance of your network with all the backends that are available on your target platform to ensure optimal performance. For more information see the benchmark results.

Note that you must abide by the licensing rules of whichever backend library you choose.

Other configuration flags

If you would like to build RTNeural with the AVX SIMD extensions, you may run CMake with the -DRTNEURAL_USE_AVX=ON. Note that this flag will have no effect when compiling for platforms that do not support AVX instructions.

Building the test suite

To build RTNeural's test suite, run cmake -Bbuild -DBUILD_TESTS=ON, followed by cmake --build build. To run the full testing suite, run ctest from the build folder. For more information, see tests/README.md.

Building the Performance Benchmarks

To build the performance benchmarks, run cmake -Bbuild -DBUILD_BENCH=ON, followed by cmake --build build --config Release. To run the layer benchmarks, run ./build/rtneural_layer_bench <layer> <length> <in_size> <out_size>. To run the model benchmark, run ./build/rtneural_model_bench.

Building the Examples

To build the RTNeural examples run:

cmake -Bbuild -DBUILD_EXAMPLES=ON
cmake --build build --config Release

The example programs will then be located in build/examples_out/, and may be run from there.

An example of using RTNeural within a real-time audio plugin can be found on GitHub here.

Building without CMake

If you wish to use RTNeural in a project that doesn't use CMake, RTNeural can be included as a header-only library, along with a few extra steps.

  1. Add a compile-time definition to define a default byte alignment for RTNeural. For most cases this definition will be one of either:

    • RTNEURAL_DEFAULT_ALIGNMENT=16
    • RTNEURAL_DEFAULT_ALIGNMENT=32
  2. Add a compile-time definition to select a backend. If you wish to use the STL backend, then no definition is required. This definition should be one of the following:

    • RTNEURAL_USE_EIGEN=1
    • RTNEURAL_USE_XSIMD=1
  3. Add the necessary include paths for your chosen backend. This path will be one of either:

    • <repo>/modules/Eigen
    • <repo>/modules/xsimd/include/xsimd

It may also be worth checking out the example Makefile.

Contributing

Contributions to this project are most welcome! Currently, there is a need for the following improvements:

  • Improved support for 2-dimensional input/output data.
  • Improved support for "stateless" Conv1D layers.
  • More robust support for exporting/loading models.
  • Support for more activation layers.
  • Any changes that improve overall performance.

General code maintenance and documentation is always appreciated as well! Note that if you are implementing a new layer type, it is not required to provide support for all the backends, though it is recommended to at least provide a "fallback" implementation using the STL backend.

Contributors

Please thank the following individuals for their important contributions:

Powered by RTNeural

RTNeural is currently being used by several audio plugins and other projects:

  • 4000DB-NeuralAmp: Neural emulation of the pre-amp section from the Akai 4000DB tape machine.
  • AIDA-X: An AU/CLAP/LV2/VST2/VST3 audio plugin that loads RTNeural models and cabinet IRs.
  • BYOD: A guitar distortion plugin containing several machine learning-based effects.
  • Chow Centaur: A guitar pedal emulation plugin, using a real-time recurrent neural network.
  • Chow Tape Model: An analog tape emulation, using a real-time dense neural network.
  • cppTimbreID: An audio feature extraction library.
  • guitarix: A guitarix effects suite, including neural network amplifier models.
  • GuitarML: GuitarML plugins use machine learning to model guitar amplifiers and effects.
  • MLTerror15: Deeply learned simulator for the Orange Tiny Terror with Recurrent Neural Networks.
  • NeuralNote: An audio-to-MIDI transcription plugin using Spotify's basic-pitch model.
  • rt-neural-lv2: A headless lv2 plugin using RTNeural to model guitar pedals and amplifiers.
  • Tone Empire plugins:
    • LVL - 01: An A.I./M.L.-based compressor effect.
    • TM700: A machine learning tape emulation effect.
    • Neural Q: An analog emulation 2-band EQ, using recurrent neural networks.
  • ToobAmp: Guitar effect plugins for the Raspberry Pi.

If you are using RTNeural in one of your projects, let us know and we will add it to this list!

License

RTNeural is open source, and is licensed under the BSD 3-clause license.

Enjoy!