menoh
menoh copied to clipboard
Menoh: fast DNN inference library with multiple programming language support
Currently mkldnn backend reorders fixed parameters (e.g. weights of Conv) at runtime. It should be done once at build time.
Because `menoh_model_run()` can take long time, it is desirable to have a function for canceling the computation from other threads. In Ruby binding, I'm trying to release the GVL (giant...
I'd like to know memory consumption and computation time for each layer. It would be helpful for tuning networks.
Currently `attribute_completion_and_shape_inference.hpp` is generated at coding time but it should be gnerated at building time.
I tried to run [super resolution example of MXNet](https://github.com/onnx/tutorials/blob/master/tutorials/OnnxMxnetImport.ipynb). The model contains unsupported operator `Transpose` so I decided to specify the variable just before the `Transpose` (i.e. variable "25") as...
I noticed that current AWS Lambda did not support libstdc++.so with GLIBCXX_3.4.20 or greater, which means all dependencies must be compiled with gcc < v5.0. (C++14 is not supported) It...
Hi I was trying Menoh vgg16 example. Does Menoh utilizes SIMD IS (like sse4, avx2 ) to speed up the inference ? If not is there an option to utilize...
The building time for each commit is time consuming (it takes [20~40 mins](https://travis-ci.org/pfnet-research/menoh/builds/429850052) on Travis CI and Appveyor). It (potentially) hinders faster iterative development.
As introduced in #105, building Menoh on Travis CI will be [dockerized](https://docs.docker.com/engine/examples/) (containerized) for Linux-based platforms. It is also useful for organizing development environment and running the tests on developer's...
I ran the following program with the attached `ggnn.onnx` in [ggnn.zip](https://github.com/pfnet-research/menoh/files/2352221/ggnn.zip). ```c #include #include #define ERROR_CHECK(statement) \ { \ menoh_error_code ec = statement; \ if(ec) { \ printf("%s\n", menoh_get_last_error_message()); \...