quantized_ops source code

[hanihanizyh]

Thanks for your work on this project and the paper "Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networks". Thesedays, I also want to use uniform symmetric quantizer limited by hardware platform while training models in tensorflow. I am wondering if it possible to change the quantization method in tensorflow source code of fake-nodes and realize uniform symmetric quantizer inference and back propagation? And if it's possible to send me the source code of kernels (quantize_ops.so,quantize_ops_cuda.so) for reference? Here is my e-mail : [email protected] Thanks very much~