caffe
caffe copied to clipboard
intel
This fork of BVLC/Caffe is dedicated to improving performance of this deep learning framework when running on CPU, in particular Intel® Xeon processors.
I am trying to build intel caffe. I use g++ version 5.5.0. I have error at src/caffe/layer_factory.cpp:91:8: error: variable ‘use_dilation’ set but not used [-Werror=unused-but-set-variable] bool use_dilation = false; ^...
for now, streamHolder is a global singleton. And all the primitives share one mkldnn stream. It has 2 problems: 1. not thread-safe. Even when a net is created and used...
Hi! Firstly, very thankful for your contribution. And right now, I am doing some experiments about gabor filter. I am fresh in this filed. I want to know why you...
I cant understand how did you find params for quantization such as scale_in scale_out scale_params. Is any automatic tool for this?
Hi, I have been trying to use Intel Caffe for a project that serves a deep model to multiple users in parallel. We have multiple copies of the model loaded...
Dear Whom It May Concern, When trying to export a model trained using Intel Caffe, we get a compatibility error due to the different implementation of BN layer in IntelCaffe....
Hi, is there any example/tutorial on how to generate a quantized net model? I'm very new to intel caffe. I'm attracted because this is the only framework which has MKL-DNN...
This patch adds 3d support to all the layers needed to run 3dUnet in inference. For others trying to run 3dUnet on caffe, you will run into two issues. 1....
WiAU_v2.0.prototxt and solver.prototxt is a great deep cnn_resnet model for csi data, and Transform.py can transform the csi data to lmdb file, which can be easily trained by intel-caffe
