Posts in category ‘Releases’.

Pytorch OpenCL backend - simplified

Thursday, October 27, 2022, by artyom ; Posted in: Releases; 2 comments

Now installation of opencl backend for pytorch is really simple.

  1. Install nighly version of pytorch for CPU in virtual environment
  2. Clone dlrpim_backend repository and checkout true_out_of_tree_support branch
  3. Update submodules
  4. Run few commands inside repo

     mkdir build
     cd build
     cmake -DCMAKE_PREFIX_PATH=$VIRTUAL_ENV/lib/python3.8/site-packages/torch/share/cmake/Torch ..
     cd ..
  5. Run mnist training:

     python --device=ocl:0

That's it.

Keep updated

Inference of ONNX Models using DLPrimitives

Sunday, January 16, 2022, by artyom ; Posted in: Releases; 0 comments

I worked on integration of inference of ONNX models using DLPrimitives. It isn't a simple task since ONNX operator set is very reach and many things can be implemented in different ways.

After many revisions and improvements I managed to validate multiple imagenet pretrained networks from pytorch, mxnet and few based on TensorFlow (see about issues with TF later)

How do you create a dlprimitives network using ONNX Model?

// load and parse ONNX Model
dp::ONNXModel model;

// create network
dp::Context ctx(device_id);
dp::Net net(ctx);

// load parameters

And you are ready to go.

I validated following networks and frameworks:

  • Pytorch, op-sets 9, 11, 13, nets alexnet, vgg16, resnet18, resnext50_32x4d, wide_resnet50_2, efficientnet_b0, efficientnet_b4, regnet_y_400mf, squeezenet1_0, mobilenet_v2, densenet121
  • MXNet: vgg11_bn, alexnet, mobilenetv2_0.25, mobilenet0.25, densenet121, resnet18_v1, squeezenet1.0
  • Tensorflow: op-sets 9, and 11 limited initial support, channel first format: resnet50, densenet121

Some networks on pytorch don't pass due to lack of some of the operators. The situation with TensorFlow is more complicated and only few networks worked ok.


When I stated validated pretrained keras networks I discovered very surprising thing. TensorFlow uses asymmetrical padding in some cases, since in TF/Keras you don't explicitly provide padding but rather give some vague definition of "same" or "valid" for the padding, in some cases padding may differ on start and end of the image.

Interestingly, cuDNN does not even provide asymmetrical padding option for convolutions. Looking into the code TF does padding manually is such case (that is actually huge waste of memory and memory bandwidth)

So implementing these convolutions will require implementing of new simple padding layer just to make sure we can use dlprimitives for inference of TF models.

To be continued...

Documentation is Online

Thursday, September 16, 2021, by artyom ; Posted in: Releases; 0 comments

I published recent documentation online:

next page