issues
search
alibaba
/
MNN
MNN is a blazing fast, lightweight deep learning framework, battle-tested by business-critical use cases in Alibaba
http://www.mnn.zone/
8.34k
stars
1.62k
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
OpenCLBackend.cpp里调用createOpenCLSymbolsOperatorSingleInstance()生成了gInstance，在哪里释放
#2869
xxxxxxLD
opened
18 hours ago
0
部署在安卓端的多分类分割任务，为什么输出的是一个一维的元组呀，有没有什么好方法能够实现分割结果的mask
#2868
MrBroWinter
opened
20 hours ago
0
pymnn inference quality is unstable
#2867
kmn1024
opened
1 day ago
2
Added support to build pymnn with openmp
#2866
avinash31d
opened
2 days ago
0
Added support to build pymnn with openmp
#2865
avinash31d
closed
2 days ago
1
环境变量设置推理线程数
#2864
Zzzz1111
opened
2 days ago
0
MNN:Sync Sync Internal 2.9.0
#2863
jxt1234
closed
4 days ago
1
从.onnx转成.mnn之后不能训练
#2862
teslanow
opened
5 days ago
0
Update module.md
#2861
DestinyOfLove
closed
4 days ago
1
MNN1.2 MNN_CODEGEN_REGISTER是什么？可以关了吗？
#2860
xin486946
opened
6 days ago
2
Demo multiPose fail, report inputChannel: 225, batch:1, width:225, height:3. Input data channel may be mismatch with filter channel count
#2859
bin913
opened
1 week ago
2
Does MNN supports intra_op_num_threads to configure number of physical cores? numThread is not helping in this.
#2858
avinash31d
opened
1 week ago
2
Remove libc++ setting for Linux clang builds
#2857
EduMenges
opened
1 week ago
1
鲲鹏920环境Benchmark跑量化模型crash
#2856
zhenjing
opened
2 weeks ago
2
鲲鹏920环境session推理采用fp16推理结果错误
#2855
zhenjing
opened
2 weeks ago
2
多session时(多算法) CPU计算场景，内部线程池性能比openMP线程池差50%
#2854
zhenjing
opened
2 weeks ago
4
纯GPU网络推理耗时
#2853
danpeng2
opened
2 weeks ago
1
Android平台下OpenCL使用setCacheFile()，模型的初始化时间未得到优化
#2852
CloudGuardian
opened
2 weeks ago
5
鲲鹏920支持fp16，yolov8模型使用fp16模型相比fp32速度没提升
#2851
zhenjing
closed
2 weeks ago
1
mnn推理时间异常
#2850
jamesdod
opened
2 weeks ago
5
debian系统下mnnquant量化出错：
#2849
LimingA1
opened
2 weeks ago
0
几乎必现的heap-use-after-free
#2848
VincentZhaoBing
opened
2 weeks ago
3
请教关于mnncompress的使用问题
#2847
stricklandye
opened
2 weeks ago
3
Android使用Vulkan推理模型，好像依然是跑在CPU上
#2846
Jverson
opened
2 weeks ago
1
MNN 在 iOS平台上编译mnn_2.8.0_ios_llm.zip release版本的方法
#2845
xingjinglu
opened
2 weeks ago
2
请问mnn框架在英伟达的orinX上可以用吗(使用GPU作为backend)？
#2844
Moxoo
opened
3 weeks ago
0
提个bug， Windows 使用 MSVC 编译时， 如果配置为 MT/MTd ， BenchMark 的doBench函数在 析构智能指针崩溃
#2843
sg0771
closed
4 days ago
1
[MNN:Bugfix] Fix bug for resize opt bug, support llama3 8b
#2842
jxt1234
closed
3 weeks ago
1
如何调用自带的 vulkan ，经试验，无论MNN_USE_SYSTEM_LIB怎么设置都还是去找系统的库
#2841
deahhh
opened
3 weeks ago
5
AMD64 上运行正常的程序到Arm64上出现段错误，出现在创建session时
#2840
deahhh
closed
3 weeks ago
2
[MNN:Sync] Sync Internal 2.8.4
#2839
jxt1234
closed
3 weeks ago
1
Error for concat size of op
#2838
jtyan123
closed
3 weeks ago
3
ONNX模型成功转为MNN后，通过几个测试工具如MNNV2Basic.out测试均未发现错误，使用python加载运行出现段错误
#2837
deahhh
closed
3 weeks ago
2
交叉编译报错
#2836
jamesdod
closed
2 weeks ago
3
mnn推断比pytorch推断耗时长
#2835
jtyan123
opened
3 weeks ago
8
【Mac上运行llm_demo报错：./llm_demo model/qwen-1.8b-int4
#2834
luocf
opened
3 weeks ago
3
windows下是否支持GPU推理？
#2833
puxuntu
opened
4 weeks ago
2
convert onnx model of decode part of SAM failed.
#2832
AminGeng
closed
4 days ago
3
ImageProcess，官方文档中的示例应用于 灰度图，从Tensor.print()出来的数值无法用matplotlib绘制出图像
#2831
deahhh
closed
3 weeks ago
10
TorchScript to MNN转换出错
#2830
puxuntu
opened
4 weeks ago
1
onnx转mnn，resize报错
#2829
123liluky
closed
4 weeks ago
3
图片推理mnn耗时是pytorch的两倍多近似3倍。
#2828
jtyan123
closed
4 weeks ago
8
相同图片推测。mnn输出与pytorch差距太大
#2827
jtyan123
closed
4 weeks ago
2
MNN是否查看权重？
#2826
cvv-student
opened
1 month ago
1
onnx和mnn推理结果不一致，使用onnxruntime推理onnx结果是正常的，但使用mnn就不正常了
#2825
lewis-v
opened
1 month ago
13
module API 输出为空
#2824
jtyan123
closed
1 month ago
5
Android手机上shell仿真测试，.so可以调起vulkan但.a失败
#2823
CloudGuardian
opened
1 month ago
6
c++ mNN release lib so 版本选择
#2822
linlongrd
closed
3 weeks ago
2
pip install MNN fail
#2821
xlsd
opened
1 month ago
3
包含5维的gridsample模型，cpu和opencl的输出结果不一致
#2820
liuyingbin123
opened
1 month ago
6
Next