I would like to try the Quantization Aware Training tool proposed by Transfer Learning Toolkit.
I have a few questions:
- is QAT only available in TLT 2 and with TensorRT 7 ?
- in the documentation
there is a new parameter “enable_qat” for most object detection models, except Faster RCNN. Is QAT training available for Faster RCNN ?
Thanks for your help