Rui Chen

When I 'ldd dirt/librasterise.so`, I got ``` linux-vdso.so.1 (0x00007ffdfb1fd000) libOpenGL.so.0 => /usr/lib/x86_64-linux-gnu/libOpenGL.so.0 (0x00007f519f318000) libEGL.so.1 => /usr/lib/x86_64-linux-gnu/libEGL.so.1 (0x00007f519f104000) libtensorflow_framework.so.1 => not found librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x00007f519eefc000) libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f519ecdd000) libdl.so.2...

Result of `nvidia-smi -q` ``` ==============NVSMI LOG============== Timestamp : Sat Jan 23 12:47:12 2021 Driver Version : 450.102.04 CUDA Version : 11.0 Attached GPUs : 1 GPU 00000000:01:00.0 Product Name...

The result of `ls -l /usr/lib*/*/*GL*` ``` -rwxr-xr-x 1 root root 12424 Jan 23 11:11 /usr/lib/nvidia/libGLX_installcheck.so.0 -rwxr-xr-x 1 root root 240464 Dec 2 02:30 /usr/lib/slack/libEGL.so -rwxr-xr-x 1 root root 6814152...

The result of `cmake ../csrc`: ``` -- The CXX compiler identification is GNU 7.5.0 -- The CUDA compiler identification is NVIDIA 11.0.221 -- Check for working CXX compiler: /usr/bin/c++ --...

The result of `python tests/square_text.py` ``` WARNING:tensorflow:From tests/square_test.py:43: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead. 2021-01-25 10:24:37.611545: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1 2021-01-25 10:24:37.647538: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983]...

I installed the driver using the nvidia runfile. And there is no `libEGL_nvidia*` in my system. I'll try `apt install` later. Thanks again for your quick reply.

I think that has little effect on the evaluation result, because the evaluation only considers points above the table.

One set represents one point cloud of one object. Each object is evaluated separately.

I've not seen this problem before. Apply clip to `d_coordiantes` may be a good solution.

I think it's related to the version of PyTorch. Try downgrade to pytorch1.0.1 and compile again.