auphelia

When exporting operations that use quantized tensors (floating point numbers with a scaling factor), the operations are split. First, a division block is created, which divides the input by the...

Contains fixes for VitisBuild flow

With the growth of more and more RTL variants of common FINN hardware building blocks, there is a need to better organize the integration of these components. The purpose of...

## Quick summary A standalone `BatchNormalization` node with certain settings (see .onnx file in .zip folder: [bn_model.zip](https://github.com/fastmachinelearning/qonnx/files/11102623/bn_model.zip)) changes its functional behavior, when transformed with the `BatchNormToAffine` transformation. ### Steps to...

Hi, I would like to reproduce your results from the paper "Integer-only Zero-shot Quantization for Efficient Speech Recognition" for int8 (or even int4 if possible) QuartzNet 15x5 on an A10...

The PR updates the Vitis/Vivado and FINN version and it adds an additional row in the overview table for the GTSRB benchmark. Open to-dos before it can get merged: -...

This PR is an extension of previous changes done by @mmrahorovic .

- Update commit hash in fetch-repos.sh - Update code if necessary

This PR contains changes from @mmrahorovic and others to enable double pumped compute and memory for the MVAU layer.