Skip to content

error on running dateset codes #18

@wsndy1010

Description

@wsndy1010

@yuxng said he updated the code for latest version of Sophus and Eigen. i tried it once and didnt work. with git pull origin in $ROOT/DA-RNN and complied successfully. for the error: cannot find -ltensorflow_framework. librensorflow, librensorflow_framework are tensorflow dynamic lib which be installed addionally. check http://platanios.org/tensorflow_scala/installation.html#installation-0-dependencies-1 and http://www.rubydoc.info/github/somaticio/tensorflow.rb. dont forget copy the libs to path: /usr/lib/

first run training code come out errors about python version problems. my default version of python is 3.5 and the codes is python2, so i change my default python version and reinstalled tensorflow. then comes some strange errors:
when i try
./experiments/scripts/shapenet_scene_multi_rgbd.sh 0
error : terminate called after throwing an instance of 'std::bad_alloc' what(): std::bad_alloc
i traced it back, the command line: from networks.factory import get_network, is the source for this error. i dont kown how to fix it.
when i try
sudo ./experiments/scripts/shapenet_scene_multi_rgbd_test.sh 0
error: ImportError: libkfusion.so: cannot open shared object file: No such file or directory. but libkfusion.so exits in /kinect_fusion/ build
when i try
./experiments/scripts/shapenet_scene_multi_rgbd_test.sh 0
error: ImportError: ../lib/kinect_fusion/kfusion.so: undefined symbol: _ZN2df12KinectFusion10save_modelESs . kfusion.so exits also in path /kinect_fusion.

wondering is there anyone running all codes successfully without encountering these errors or how u fix with it ?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions