EDA@TUM's Projects
An Open Source Machine Learning Framework for Everyone
TensorFlow Lite for Microcontrollers
Packing Utils for TFLite Models
generate tflite micro code which bypasses the interpreter (directly calls into kernels)
This project runs the TFLM interpreter off-target to produce static code that is able to run inference on the processed model. This avoids interpretation and code size overheads on a target that does not need to process different models dynamically.
Collection of TinyML benchmark NN models (in the making)
MicroSpeech example from the TinyML book
TUM Speech Recording is a small web application that collects short snippets of speech. It's designed to help gather open speech data sets to train machine learning systems.
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Contains files for TVM-Workshops offered by the chair.
Collection of Python scripts wrapping around the TVM framework.
AI inference library based on mbed and TensorFlow
This project contains a code generator that produces static C NN inference deployment code targeting tiny micro-controllers (TinyML) as replacement for other ยตTVM runtimes. This tools generates a runtime, which statically executes the compiled model. This reduces the overhead in terms of code size and execution time compared to having a dynamic on-device runtime.
Library of example SystemC/TLM peripherals for various SoCs based on the SCS library
vRTLmod modifies Verilator generated RTL simulation code for faul-injection purposes. It transforms source code with the help of LLVM/Clang-Tools and generates a fault injection API.