udacity / didi-competition Goto Github PK
View Code? Open in Web Editor NEWResources for the Udacity/Didi $100k competition
Resources for the Udacity/Didi $100k competition
all my files is correct, but problem
$ sudo ./run-bag_to_kitti.sh -i ~/Data/Didi/Didi-Release-2 -o ~/Data/Didi/output2
Running 'python /python/bag_to_kitti.py' with input dir '/home/elninowang/Data/Didi/Didi-Release-2', output dir '/home/elninowang/Data/Didi/output2', docker image 'udacity-sdc2017-reader'...
Extracting bag info /data/Data/2/1.bag
ERROR bag unindexed: /data/Data/2/1.bag. Run rosbag reindex.
Traceback (most recent call last):
File "/python/bag_to_kitti.py", line 471, in
main()
File "/python/bag_to_kitti.py", line 258, in main
bagsets = find_bagsets(indir, filter_topics=filter_topics, set_per_file=True, metadata_filename='metadata.csv')
File "/python/bag_utils.py", line 169, in find_bagsets
bag_set = BagSet(set_name, [f], filter_topics, metadata_path)
File "/python/bag_utils.py", line 81, in init
self._process_infos(filter_topics)
File "/python/bag_utils.py", line 91, in _process_infos
info = get_bag_info(f)
File "/python/bag_utils.py", line 28, in get_bag_info
if 'start' in info:
TypeError: argument of type 'NoneType' is not iterable
From https://discussions.udacity.com/t/kitti-to-bag-py-code-is-wrong/241269
Regarding bag_to_kitti.get_yaw()
A line between points with nearly equal x components is not parallel to the x-axis, so the early return is incorrect.
I use the script to transfer the data to Kitti.
When running bag_to_kitti.py to obtain the necessary groundtruth tracklets for training, this error is thrown:
1000 images, 1416 messages processed...
Writing done. 1873 images, 2607 messages processed.
Traceback (most recent call last):
File "bag_to_kitti.py", line 378, in
main()
File "bag_to_kitti.py", line 327, in main
cap_rear_gps_interp = interpolate_to_camera(camera_index_df, cap_rear_gps_df, filter_cols=gps_cols)
File "bag_to_kitti.py", line 125, in interpolate_to_camera
merged.interpolate(method='time', inplace=True, limit=100, limit_direction='both')
File "/usr/local/lib/python2.7/dist-packages/pandas/core/generic.py", line 3670, in interpolate
raise TypeError("Cannot interpolate with all NaNs.")
TypeError: Cannot interpolate with all NaNs.
This happens right after extracting all the bags in the input folder, while processing the first .bag file. It's been tested in all folders (1, 2 & 3) from Didi-Dataset 2.
Hello,
I'm writing a bachalor thesis about obstacle detection. I would like to run your code. However I cannot download calibration data. Could you share calibration.tar.gz ? Torrent doesn't work, because no seeders.
I would appreciate your help.
Thank you in advance
Regards
Robert Czwartosz
There appears to be a null byte at the end of the line in
https://github.com/udacity/didi-competition/blob/master/tracklets/python/parse_tracklet.py#L3
it's not showing up in github, but is in the file when I download and open it in sublime
in the file generate_tracklet.py, line 86:
class_id = obj.write_xml(f, class_id, tab_level)
However, the funtion write_xml has no return value,
I tried following the instructions here https://github.com/udacity/didi-competition/blob/master/docs/GettingStarted.md
As per the instructions I downloaded the Didi-Training-Release-1.tar.gz and ran rviz while playing 5mph.bag. In Rviz I could only select base_link
or radar
as the fixed frame, the instructions request selecting velodyne
as the fixed frame, which was not an option. After selecting radar
, I tried to add the velodyne point-cloud by topic. After doing this I received the error:
For frame [velodyne]: Frame [velodyne] does not exist
I get the same error if I select base_link
as the fixed frame. I'm running ROS Kinetic on Ubuntu 16.04.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.