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

Sync v2.0 version of code to github repo

Browse Source
This commit is contained in:
Jiang-Jia-Jun
2025-06-29 23:29:37 +00:00
parent d151496038
commit 92c2cfa2e7
597 changed files with 78776 additions and 22905 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
fastdeploy/model_executor/layers/backends/xpu/utils.py
View File

@@ -16,11 +16,13 @@
!! This file will be deleted after the platform is fully functional
"""
from typing import Tuple
import numpy as np
import paddle
def xpu_clip_and_round(x):
def xpu_clip_and_round(x: np.ndarray) -> np.ndarray:
"""
Clip and round the input array to the range [-127, 127] and convert to int8.
@@ -33,7 +35,8 @@ def xpu_clip_and_round(x):
return np.clip(np.around(x), -127, 127).astype("int8")
def xpu_quant_qkv_weight(weight_np):
def xpu_quant_qkv_weight(
weight_np: np.ndarray) -> Tuple[paddle.Tensor, paddle.Tensor]:
"""
Quantize the query, key, and value weights for the Transformer model.
@@ -61,7 +64,8 @@ def xpu_quant_qkv_weight(weight_np):
return quanted_weight, weight_scales
def xpu_quant_weight(weight_np):
def xpu_quant_weight(
weight_np: np.ndarray) -> Tuple[paddle.Tensor, paddle.Tensor]:
"""
Quantize the weight tensor for XPU devices.