Open jpinela opened 2 years ago
Welcome to Apache MXNet (incubating)! We are on a mission to democratize AI, and we are glad that you are contributing to it by opening this issue. Please make sure to include all the relevant context, and one of the @apache/mxnet-committers will be here shortly. If you are interested in contributing to our project, let us know! Also, be sure to check out our guide on contributing to MXNet and our development guides wiki.
Hello Team,
I have tried for the last 2 days to install and get the mxnet tutorial running
https://cv.gluon.ai/build/examples_tracking/demo_smot.html#sphx-glr-download-build-examples-tracking-demo-smot-py
, with several combinations of packages and package versions, but can't seem to get any progress done. The problem is very vague. I downgraded to python 3.7.9 because 3.8 was just not helping.the result of the diagnosis script is
nvcc is
I tried this sequence: pip3 install torch pip install opencv-python pip install mxnet-cu102==1.7.0 -f https://dist.mxnet.io/python/cu102 pip3 install gluoncv
then install the nvidia cuda 10.2 drivers installed VS2022
but still getting this error with the GPU
the only way to be able to import mxnet is to use the cpu library
pip install --upgrade mxnet-mkl
besides that, I have tried several versions of all the components, but haven't been able to get the tutorial test command
python demo.py MOT17-13-FRCNN-raw.webm --network-name ssd_512_resnet50_v1_coco --use-pretrained --custom-classes person car --detect-thresh 0.7 --use-motion
to run, because it keeps wanting the gpu (saying it was not enabled, running on the cpu library).I have looked through several issues here, but didn't come to any conclusion.
Could you please point to a group of versions and components that should be able to run the tutorial on windows 10? from what I could see, running on linux would be much easier.
Thanks Team