apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

polish code with new pre-commit rule (#2923)

Browse Source
This commit is contained in:
Zero Rains
2025-07-19 23:19:27 +08:00
committed by GitHub
parent b8676d71a8
commit 25698d56d1
424 changed files with 14307 additions and 13518 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
fastdeploy/model_executor/layers/backends/xpu/utils.py
View File

@@ -36,7 +36,8 @@ def xpu_clip_and_round(x: np.ndarray) -> np.ndarray:
def xpu_quant_qkv_weight(
weight_np: np.ndarray) -> Tuple[paddle.Tensor, paddle.Tensor]:
weight_np: np.ndarray,
) -> Tuple[paddle.Tensor, paddle.Tensor]:
"""
Quantize the query, key, and value weights for the Transformer model.
@@ -65,7 +66,8 @@ def xpu_quant_qkv_weight(
def xpu_quant_weight(
weight_np: np.ndarray) -> Tuple[paddle.Tensor, paddle.Tensor]:
weight_np: np.ndarray,
) -> Tuple[paddle.Tensor, paddle.Tensor]:
"""
Quantize the weight tensor for XPU devices.