apatsekin

if you are building in Nvidia docker container without actual GPU, you can use something like this: ``` CUDA_VERSION=$(/usr/local/cuda/bin/nvcc --version | sed -n 's/^.*release \([0-9]\+\.[0-9]\+\).*$/\1/p') if [[ ${CUDA_VERSION} == 9.0*...

It works in a [0.0,255.0] range but instead of random channel shift i.e. r+20, g-10, b+3, it just applies random brightness: r+5, g+5,b+5 (shifting all channels by the same value).

if you are using docker, try to run it with ` --ipc=host` flag

Got exactly the same exception: > mxnet.base.MXNetError: [08:59:39] src/operator/nn/./../tensor/../elemwise_op_common.h:135: Check failed: assign(&dattr, vec.at(i)): Incompatible attr in node at 0-th output: expected [1,64,1,1], got [64] While just running rpn_test.py on trained...

Because during training tridentnet creates two separate data providers for landscape and portrait images. During testing everything is taken from one iterator, since batch size = 1 it doesn't matter....

For me it was: ``` brew install cmake brew install protobuf brew link --overwrite protobuf pip3 install onnx ``` Works for 1.12.0. ``` $ python3 --version Python 3.10.6 $ which...

> `data = scatter(data, [device])[0]` -> `data = scatter(data, ['mps'])[0]` should work but be sure include a block to check if the model and data are both in `mps`. >...

As I figured, PyPi's wheel has all 3rd party libs (including openblas) stuffed in, they are unpacked to `opencv_contrib_python_headless.libs` during `pip install`. I'm not getting the same result when following...