Mohamed Moursi

Hi, I am not an expert but just out of curiosity is normal for it to try to allocate 10.3 TiB???

Could this be due to CIFAR100, not ImgNet?

AFAIK, draw a box is not classification rather object detection {Classification+Localization}

Doesn't it at least produce a warning that a keyword argument is passed but not used?

AKAIK there is no quantized softmax layer but you can use it from torch it's just a mathematical layer.

Hi @ardeal yes it could be passed directly.

Hi, This depends on which bias quantizer you use, most bias quantizers require the input to be quantized e.g. `Int8Bias, Int16Bias, Int32Bias` if you do not want to quantize the...

Hi, I am not sure what is the current state of the documentation as I am not part of the team behind brevitas. But there are a couple of notebooks...

Hi @ardeal, In this case, have a look at [this](https://arxiv.org/abs/1712.05877) paper and [this](https://www.tensorflow.org/lite/performance/quantization_spec) doc.

Hi, AFAIK, brevitas follows the same quantization scheme as the paper I shared previously. Regarding your questions: 1. In each layer you can define, `weight_quant, bias_quant, input_quant, output_quant`. Some layers...