-
Platform (like ubuntu 16.04/win10): Ubuntu 16.04
Python version: Python 3.5.5
Source framework with version (like Tensorflow 1.4.1 with GPU): Tensorflow 1.10.1 with GPU
Destination framework …
-
## Issue description
I am baffled by this actually (probably because I don't understand cmake very well). After I upgraded to cuda-9.2, wiped out the whole build directory, and reran the build flow, …
-
I cannot find any resource about running inference with float16. There is resnet.py for training with float16, but no information is provided for inference with float16.
Any help?
Thank you
-
I have installed CUDA and CUDNN, but today I run MNIST for a test, then it show me that "
I0726 15:18:46.212692 6426 operator.cc:169] Engine CUDNN is not available for operator Conv.
I0726 15:18:46…
-
I want to know how to generate init_net.pb and predict_net.pb?
-
Changing the route of pre-trained model to
https://dl.fbaipublicfiles.com/detectron//ImageNetPretrained/MSRA/R-50.pkl
raises the following assertion error:
File "/home/ubuntu/deploy/detectron…
-
I implemented a naive layer normalization with "manual" LSTM and it run extremely slow, 10 times slower than using the cudnn `RNN` interface (no layer normalization feature).
Could we implement a L…
-
I have two questions:
When I read this:
// >>> conv1 = brew.conv(model, data, 'conv1', dim_in=1, dim_out=20, kernel=5)
model.AddConvOps("data", "conv1", 1, 20, 1, 0, 5, test);
What is dim_out?…
-
- PyTorch or Caffe2: PyTorch
- Build command you used:
cmake -DUSE_NATIVE_ARCH=ON -DUSE_FFMPEG=ON -DUSE_ATEN=ON -D USE_TENSORRT=ON -D USE_OPENMP=ON -DUSE_MKL=ON -DUSE_IDEEP=ON ..
- OS: …
-
This fails:
cmake -DBUILD_PYTHON=OFF -DUSE_NNPACK=OFF ..
with an error message like this:
```
CMake Error at cmake/Codegen.cmake:163 (message):
Failed to get generated_cpp list
Call St…