Skip to main content

quantize_weights

god_graph::transformer::quantization::weight_quantization

Function quantize_weights 

Source 
pub fn quantize_weights(weights: &DenseTensor) -> QuantizedTensor
Expand description

Quantize model weights to INT8