Comments (3)
用 GetMNNInfo model.mnn可以查看三个输入对应的dimension format
from mnn.
用 GetMNNInfo model.mnn可以查看三个输入对应的dimension format
hw.cpufamily: 3660830781 , size = 4
The device support i8sdot:1, support fp16:1, support i8mm: 0
Model default dimensionFormat is NCHW
Model Inputs:
[ source_img ]: dimensionFormat: NC4HW4, size: [ 1,3,416,320 ], type is float
[ ref_img ]: dimensionFormat: NC4HW4, size: [ 1,15,416,320 ], type is float
[ audio_feature ]: dimensionFormat: NC4HW4, size: [ 1,29,5 ], type is float
Model Outputs:
[ 774 ]
Model Version: 2.8.3
使用GetMNNInfo model.mnn获取到的信息,但是这个跟我下面代码读到信息不一致
`auto input = net->getSessionInput(session, "source_img");
MNN_PRINT("source_img shape size: %d, ", input->shape().size());
for (int i = 0; i < input->shape().size(); ++i) {
MNN_PRINT("%d --> dim:%d, ", i, input->shape()[i]);
}
MNN_PRINT("N: %d, C:%d, H:%d, W:%d, ",input->batch(), input->channel(), input->height(), input->width());
MNN_PRINT("type:%d (0: NHWC, 1:NCHW, 2:NC4HW4)", input->getDimensionType());
MNN_PRINT("\n\n");
auto input2 = net->getSessionInput(session, "ref_img");
MNN_PRINT("ref_img shape size: %d, ", input2->shape().size());
for (int i = 0; i < input2->shape().size(); ++i) {
MNN_PRINT("%d --> dim:%d, ", i, input2->shape()[i]);
}
MNN_PRINT("N: %d, C:%d, H:%d, W:%d, ",input2->batch(), input2->channel(), input2->height(), input2->width());
MNN_PRINT("type:%d (0: NHWC, 1:NCHW, 2:NC4HW4)", input2->getDimensionType());
MNN_PRINT("\n\n");
auto input3 = net->getSessionInput(session, "audio_feature");
MNN_PRINT("audio_feature shape size: %d, ", input3->shape().size());
for (int i = 0; i < input3->shape().size(); ++i) {
MNN_PRINT("%d --> dim:%d, ", i, input3->shape()[i]);
}
MNN_PRINT("N: %d, C:%d, H:%d, W:%d, ",input3->batch(), input3->channel(), input3->height(), input3->width());
MNN_PRINT("type:%d (0:NHWC, 1:NCHW, 2:NC4HW4)", input3->getDimensionType());
MNN_PRINT("\n\n");`
hw.cpufamily: 3660830781 , size = 4
The device support i8sdot:1, support fp16:1, support i8mm: 0
source_img shape size: 4, 0 --> dim:1, 1 --> dim:3, 2 --> dim:416, 3 --> dim:320, N: 1, C:3, H:416, W:320, type:1 (0: NHWC, 1:NCHW, 2:NC4HW4)
ref_img shape size: 4, 0 --> dim:1, 1 --> dim:15, 2 --> dim:416, 3 --> dim:320, N: 1, C:15, H:416, W:320, type:1 (0: NHWC, 1:NCHW, 2:NC4HW4)
audio_feature shape size: 3, 0 --> dim:1, 1 --> dim:29, 2 --> dim:5, N: 1, C:29, H:5, W:1, type:1 (0:NHWC, 1:NCHW, 2:NC4HW4)
from mnn.
getDimensionType 由于历史原因,会把 NC4HW4 当成 NCHW 传回。
另外建议模型转换时加上 --keepInputFormat 参数,这样根据原始模型统一用 NCHW (Onnx) 或者 NHWC (Tensorflow)即可
from mnn.
Related Issues (20)
- 编译示例工程代码报错
- 编译测试工具报错阻塞 HOT 3
- 问一下, 在armv8.2 cpu 手机上, 是否已支持fp16 进行卷积的推理阿? HOT 1
- c++ 链接MNN后,编译报错,无法解析的外部符号 _std_init_once_link_alternate_names_and_abort和_std_min_element_4 HOT 6
- mnncompress量化模型以后大小没有变化 HOT 3
- MNN.cv has the rotation issue in loading some photos from the phone HOT 1
- 求教 expr 中_Fill 的正确用法 HOT 3
- 鸿蒙系统编译失败 HOT 1
- Facing issue with version incapitability between protobuffer library and MNN library while performing mnnquant quantization process using python API of mnn HOT 4
- 多输入模型转换时添加--optimizePrefer 2选项 并基于此使用MNNPythonOfflineQuant离线量化,在arm cpu后端推理输出异常,与x86 cpu推理不对齐 HOT 1
- 请问MNN如何实现使用同一模型做并行推理? HOT 5
- Invalidate buffer to create MNN Module HOT 5
- MNNConvert onnx转mnn 错误,nn.ConvTranspose1d forward 中 pad 会出问题 HOT 7
- How to get SD mnn model used in transformers/diffussion/diffussion_demo
- Android 编译DEMO 出现 Target "MNN_CL_WRAP" of type OBJECT_LIBRARY may not be linked into another target
- 问问MNN的兼容方式 HOT 1
- ios平台 没有getTensor方法 HOT 2
- linux x86_64 平台, 多线程情况下, mnn 和推理速度不如 onnx,有可能是什么原因? HOT 4
- 手机端vulkan特性 image_write_without_format这个特性影响大吗?
- mnn 模型,Android 平台 benchmark 测试,推理使用 openCL, vulkan 都比 CPU 慢,可能是什么原因?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mnn.