PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度
tiny change
wenhuach21f92ed2c72959
diff --git a/README.md b/README.mdindex 331e11d96..2b5a27c7f 100644--- a/README.md+++ b/README.md@@ -33,7 +33,7 @@ autoround.export_to_itrex(output_dir=output_dir)## export to autogptq for gpu deployment-# please install auto-gptq https://github.com/AutoGPTQ/AutoGPTQ+## please install auto-gptq https://github.com/AutoGPTQ/AutoGPTQ# output_dir = "/PATH/TO/SAVE/COMPRESSED/MODEL/"# autoround.export_to_autogptq(output_dir, use_triton=True) ## Utilizing Triton for 2-bit and 4-bit scenarios## then follow autogptq to load the model