brevitas
brevitas copied to clipboard
Xilinx
Brevitas: neural network quantization in PyTorch
Currently we are not running CI in a GPU enviroment. Best way to do might be through a self hosted runner on our own servers. For security reasons, running code...
Hello, I'm trying to quantize the BatchNormd2d Layer but when I'm using BatchNorm2dToQuantScaleBias the loss isn't converging(stays at the same value). When using regular nn.BatchNorm2d with the rest of the...
Hello everyone, I am trying to define a conv layer where input & output quantizers use the same scale factor so the fractional bit_width is equal. I tried setting `tie_input_output_quant=True`...
Thanks for releasing the code and your effort. I like this project! My goal is to compile the Brevitas MobileNet model with my own dataset trained then deploy it to...
Hello Brevitas developers, It fails when I load the MobileNet model from brevitas_examples with `pretrained=True`, but it works perfectly when I change the `pretrained` flag to `False`. ```python from brevitas_examples.imagenet_classification...
I want to use the brevitas library for quantization training, and I want to set the quantization precision of the gradient, what should I do? I only construct the forward...
Hi, I want to QAT MobileNetv1 Model using my Quantization Algorithm, and export FINN ONNX to compile QAT trained model. So I set MobileNetv1's ConvBlock as follows, As you can...
Among these, per-channel scaling, tracking of the norm at training time and other minor stuff
Need more commits to setup download from release files
Documentation files need to be re-generated, will be done once everything else has been reviewed. A general rebase will be performed once all the files have been reviewed and approved.