Tensorflow error when running demo.py file with pretrained model.

[fathyshalaby]

I got this error when I tried running the scripts in a docker image (https://hub.docker.com/r/nvidia/cuda/tags?page=1&name=ubuntu): with the same parameters mentioned in the readme file.

tensorflow.python.framework.errors_impl.InternalError: 2 root error(s) found. (0) Internal: Blas xGEMM launch failed : a.shape=[1,256,24], b.shape=[1,24,256], m=256, n=256, k=24 [[node generator/generator/feature_extraction_coarse/layer1/knn_point/MatMul (defined at /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/framework/ops.py:1748) ]] [[generator/generator/coarse_coordinate_regressor/fc_layer2/BiasAdd/_147]] (1) Internal: Blas xGEMM launch failed : a.shape=[1,256,24], b.shape=[1,24,256], m=256, n=256, k=24 [[node generator/generator/feature_extraction_coarse/layer1/knn_point/MatMul (defined at /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/framework/ops.py:1748) ]]

I understand that this may be due to an error in the tf_ops/compile_ops.sh file, but just wanted to know if anybody else got this problem and how could I fix it.

[ooklwq]

I met the same problem. Make sure you have no other processes using the GPU running. Run nvidia-smi to check this.

[anshen666]

I met the same problem. Make sure you have no other processes using the GPU running. Run nvidia-smi to check this.

请问方便留个联系方式吗？这个项目一直没跑通，请问一定要在ubuntu环境下才可以跑通吗？