JzHuai0108
commented
2 years ago Sorry to bother you with the queer problem which I solved it an hour later. The root cause is PyQt5 is not installed in the virtual environment though it is installed in the system. The story is very complicated and I narrate it here for myself and the curious.
At the start, I installed all the python3 requirements on the system (ubuntu 18.04, python3.6.9 by default) following the readme,
sudo apt install python3-dev python3-pip python3-pyqt5.qtopengl
sudo pip3 install -r requirements.txtThen, when I ran visualize.py of the semantic-kitti-api
Traceback (most recent call last):
File "visualize_mos.py", line 9, in <module>
from auxiliary.laserscanvis import LaserScanVis
File "/media/jhuai/docker/lidarslam/semantic-kitti-api/auxiliary/laserscanvis.py", line 5, in <module>
from vispy.scene import visuals, SceneCanvas
File "/usr/local/lib/python3.6/dist-packages/vispy/scene/__init__.py", line 33, in <module>
from .visuals import * # noqa
File "/usr/local/lib/python3.6/dist-packages/vispy/scene/visuals.py", line 18, in <module>
from .. import visuals
File "/usr/local/lib/python3.6/dist-packages/vispy/visuals/__init__.py", line 44, in <module>
from .volume import VolumeVisual # noqa
File "/usr/local/lib/python3.6/dist-packages/vispy/visuals/volume.py", line 37
from __future__ import annotations
^
SyntaxError: future feature annotations is not definedAccording to here, I had to upgrade my python to 3.7 at least. However, my entire ros system packages depend on python3.6.9. So I wanted to work in a virtual environment for python 3.7. But I need python3.7 installed on the system in order to make the virtual env. Fortunately, the ubuntu package archive has libpython3.7 and other related packages. So I installed them system wide by sudo apt-get install.
sudo apt-get install libpython3.7 libpython3.7-dev python3.7-venvNow I could make a virtual environment based on python 3.7,
python3.7 -m venv python3.7-envand then installed the requirements packages.
source python3.7-env/bin/activate
pip3 install numpy
sudo pip3 install -r requirements.txtEverything looked fine. But the visualization windows did not appear. Some hint showed up when running
pip3 install -r requirements.txtin the virtual environment, warning that "No matching distribution found for PyQt5-Qt5>=5.15.0". Since I was sure that PyQt5-Qt5 was installed systemwide, I thought the cause for failing to create the windows in the virtual env should be that the system wide PyQt5 was not found in the virtual env. So I uninstalled the system wide PyQt5, since it was useless.
sudo apt-get remove python3-pyqt5.qtopenglThen I focused on installing PyQt5 in the virtual env. It turned out that the pip needed a upgrade as suggested on stackoverflow.
pip install --upgrade pipWith that, PyQt5 was successfully installed in the virtual env, among others, by calling
pip3 install -r requirements.txt
# pip3 install PyQt5-Qt5 # alternativelyThe current python packages in the virtual env look like
(python3.7-env) jhuai@ThinkPad-P53:/lidarslam/semantic-kitti-api$ pip list
Package Version
----------------------------- ----------
actionlib 1.12.0
angles 1.9.11
base_local_planner 1.16.7
bondpy 1.8.3
camera_calibration 1.14.0
camera_calibration_parsers 1.11.13
catkin 0.7.20
controller_manager 0.16.0
controller_manager_msgs 0.16.0
cv_bridge 1.13.0
cycler 0.11.0
Cython 0.29.32
diagnostic_analysis 1.9.3
diagnostic_common_diagnostics 1.9.3
diagnostic_updater 1.9.3
dynamic_reconfigure 1.6.0
fonttools 4.38.0
freetype-py 2.3.0
gazebo_plugins 2.8.6
gazebo_ros 2.8.6
gencpp 0.6.2
geneus 2.2.6
genlisp 0.4.16
genmsg 0.5.12
gennodejs 2.0.1
genpy 0.6.9
glfw 2.5.5
hsluv 5.0.3
image_geometry 1.13.0
imgui 1.4.1
interactive_markers 1.11.4
joint_state_publisher 1.12.14
kdl_parser_py 1.13.1
kiwisolver 1.4.4
laser_geometry 1.6.4
matplotlib 3.5.3
message_filters 1.14.3
numpy 1.21.6
nvidia-cublas-cu11 11.10.3.66
nvidia-cuda-nvrtc-cu11 11.7.99
nvidia-cuda-runtime-cu11 11.7.99
nvidia-cudnn-cu11 8.5.0.96
packaging 21.3
Pillow 9.3.0
pip 22.3.1
pkg_resources 0.0.0
PyOpenGL 3.1.6
pyparsing 3.0.9
PyQt5 5.15.7
PyQt5-Qt5 5.15.2
PyQt5-sip 12.11.0
python-dateutil 2.8.2
python_qt_binding 0.3.6
PyYAML 6.0
qt_dotgraph 0.3.16
qt_gui 0.3.16
qt_gui_cpp 0.3.16
qt_gui_py_common 0.3.16
resource_retriever 1.12.5
rosbag 1.14.3
rosboost_cfg 1.14.7
rosclean 1.14.7
roscreate 1.14.7
rosgraph 1.14.3
roslaunch 1.14.3
roslib 1.14.7
roslint 0.11.2
roslz4 1.14.3
rosmake 1.14.7
rosmaster 1.14.3
rosmsg 1.14.3
rosnode 1.14.3
rosparam 1.14.3
rospy 1.14.3
rosserial_python 0.8.0
rosservice 1.14.3
rostest 1.14.3
rostopic 1.14.3
rosunit 1.14.7
roswtf 1.14.3
rqt_action 0.4.9
rqt_bag 0.4.12
rqt_bag_plugins 0.4.12
rqt_console 0.4.9
rqt_dep 0.4.9
rqt_graph 0.4.11
rqt_gui 0.5.0
rqt_gui_py 0.5.0
rqt_image_view 0.4.14
rqt_launch 0.4.8
rqt_logger_level 0.4.8
rqt_moveit 0.5.7
rqt_msg 0.4.8
rqt_nav_view 0.5.7
rqt_plot 0.4.9
rqt_pose_view 0.5.8
rqt_publisher 0.4.8
rqt_py_common 0.5.0
rqt_py_console 0.4.8
rqt_reconfigure 0.5.1
rqt_robot_dashboard 0.5.7
rqt_robot_monitor 0.5.9
rqt_robot_steering 0.5.10
rqt_runtime_monitor 0.5.7
rqt_rviz 0.6.0
rqt_service_caller 0.4.8
rqt_shell 0.4.9
rqt_srv 0.4.8
rqt_tf_tree 0.6.0
rqt_top 0.4.8
rqt_topic 0.4.11
rqt_web 0.4.8
rtabmap-ros 0.20.18
rviz 1.13.7
sensor_msgs 1.12.7
setuptools 39.0.1
six 1.16.0
smach 2.0.1
smach_ros 2.0.1
smclib 1.8.3
tf 1.12.0
tf_conversions 1.12.0
tf2_geometry_msgs 0.6.5
tf2_kdl 0.6.5
tf2_py 0.6.5
tf2_ros 0.6.5
topic_tools 1.14.3
torch 1.13.0
turtlebot3_teleop 1.2.5
typing_extensions 4.4.0
urdfdom_py 0.4.2
vispy 0.12.0
wheel 0.38.4
xacro 1.13.5
After all these tinkering, the visualization windows finally came up.
YEOMJINSEOP
Hi, thank you for the great tool. I have installed the tool according to the readme, but the visualization window does not appear without fatal errors. I browsed the issues, the closest to mine is issue23, but it is caused by using a remote desktop. Can you please hint on how to solve this issue?
Detailed descriptions are as follows. I am working in a python 3.7 virtual environment on Ubuntu 18.04.
The terminal outputs:
All the required packages have been installed, as show below.