koba-jon / pytorch_cpp Goto Github PK

Hi,
I'm trying to build this project according to the given instructions, but the make fails because it says it needs -std=c++11 support. Where do I add this flag? I do not have g++-8 (I'm not able to install it, so I have to work with g++-5).
Thank you so much!

I checked your examples but I still cannot figure out how to build TCN using Pytorch C++.
Can you give me guidance?
or It would be great if the TCN example is added.
Thank you

Hi.
Thanks for you codes.

I need to know how heatmap make from anomaly result image and origin image.

Thanks.

Do you have a plan to implement Libtorch version YOLOV4 or YOLOV5 or YOLOX can be trained?

Hello. I tried to run test.sh of Dimensionality_Reduction/AE2d on my PC which doesn't have GPU, but failed with the below error message.

terminate called after throwing an instance of 'c10::Error'
  what():  No CUDA GPUs are available

It's because test() in AE2d/test.cpp calls torch::cuda::synchronize() regardless of device type. I think it should be called only on GPU environment. For example:

        if (!device.is_cpu()) torch::cuda::synchronize();

Hi!
I want to be able to run pix2pix in SGX, but SGX only supports statically linked programs. So I was wondering if you think it would be possible to build pix2pix statically? I would need the *.a files instead of the *.so files for each of the dependencies, but where do I make the changes in the cmake files for this? Thank you so much for your help!

Hello,
No can find the RBM code, you not use it ad in original paper?

Hey, you have showed such an amazing work to train NNs in C++.
I would like to know what are the reasons for which you started training models in C++ instead of python? Because once the model definition has been written in pytorch in python and data pipeline has been setup, all the computation needs to be done will be done on GPU. So there wont be much drastic performance gains when migrating from python to c++. Please tell me some of your thoughts on this.

Hello,
I saw your post on https://qiita.com/koba-jon/items/2b15865f5b4c0c9fbbf7

I am trying to read and write a tensor to and from a PNG using libpng without using OpenCV.
See this: prabhuomkar/pytorch-cpp#72

Do you think its possible?

Thanks