We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
PyTorch implementation of "Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference"