vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-05-04 22:51:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

[fix](test): fix import kt-kernel (#1728)

Browse source
This commit is contained in:
ErvinXie 2025-12-17 19:46:32 +08:00 • committed by GitHub
parent 6fc4080a7d
commit a8667ddb58
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
33 changed files with 1063 additions and 1151 deletions
Download patch file Download diff file Expand all files Collapse all files

39
kt-kernel/examples/test_linear.py
View file

@ -1,26 +1,27 @@
#!/usr/bin/env python
# coding=utf-8
'''
Description :
"""
Description :
Author : chenht2022
Date : 2024-07-25 10:32:05
Version : 1.0.0
LastEditors : chenht2022
LastEditors : chenht2022
LastEditTime : 2024-08-06 10:36:59
Copyright (c) 2024 by KVCache.AI, All Rights Reserved.
'''
Copyright (c) 2024 by KVCache.AI, All Rights Reserved.
"""
import os, sys
import time
sys.path.append(os.path.dirname(__file__) + '/../build')
import kt_kernel_ext
sys.path.append(os.path.dirname(__file__) + "/../build")
from kt_kernel import kt_kernel_ext
import torch
input_size = 16384
output_size = 5120
stride = 32
group_max_len = 1024
proj_type = 1 # ggml_type::GGML_TYPE_F16
hidden_type = 1 # ggml_type::GGML_TYPE_F16
proj_type = 1 # ggml_type::GGML_TYPE_F16
hidden_type = 1 # ggml_type::GGML_TYPE_F16
qlen = 30
layer_num = 10
CPUInfer = kt_kernel_ext.CPUInfer(48)
@ -30,8 +31,10 @@ with torch.inference_mode(mode=True):
linears = []
projs = []
for _ in range(layer_num):
proj = torch.randn((output_size, input_size), dtype=torch.float16, device = "cuda").to("cpu").contiguous()
config = kt_kernel_ext.linear.LinearConfig(input_size, output_size, stride, group_max_len, proj.data_ptr(), proj_type, hidden_type)
proj = torch.randn((output_size, input_size), dtype=torch.float16, device="cuda").to("cpu").contiguous()
config = kt_kernel_ext.linear.LinearConfig(
input_size, output_size, stride, group_max_len, proj.data_ptr(), proj_type, hidden_type
)
linear = kt_kernel_ext.linear.Linear(config)
projs.append(proj)
linears.append(linear)
@ -43,20 +46,14 @@ with torch.inference_mode(mode=True):
output = torch.empty((qlen, output_size), dtype=torch.float16).contiguous()
input = input / 100
CPUInfer.submit(
linear.forward(
qlen,
input.data_ptr(),
output.data_ptr()
)
)
CPUInfer.submit(linear.forward(qlen, input.data_ptr(), output.data_ptr()))
CPUInfer.sync()
# print('cpuinfer output', output)
proj = projs[i%layer_num]
proj = projs[i % layer_num]
t_output = torch.mm(input, proj.t())
# print('torch output', t_output)
diff = torch.mean(torch.abs(output - t_output)) / torch.mean(torch.abs(t_output))
print('diff = ', diff)
assert(diff < 0.001)
print("diff = ", diff)
assert diff < 0.001