I would like quantization-aware training to be fully supported in TF2 as in TF1 for object detection model
3. Additional context
The object detection API in TF1 allows to perform quantization-aware training using the graph_rewriter in the config pipeline. It would be nice to have this option in TF2
With the graph rewriter you practice fake quantization as I know only one model is fully quantized. I guess there's no support in TF2 for the quantization aware training.
Prerequisites
Please answer the following question for yourself before submitting an issue.
1. The entire URL of the file you are using
https://github.com/tensorflow/models/tree/master/research/object_detection
2. Describe the feature you request
I would like quantization-aware training to be fully supported in TF2 as in TF1 for object detection model
3. Additional context
The object detection API in TF1 allows to perform quantization-aware training using the graph_rewriter in the config pipeline. It would be nice to have this option in TF2
4. Are you willing to contribute it? (Yes or No)
No