docwza / sumolights Goto Github PK
View Code? Open in Web Editor NEWSUMO adaptive traffic signal control - DQN, DDPG, Webster's, Max-pressure, Self-Organizing Traffic Lights
License: GNU General Public License v3.0
SUMO adaptive traffic signal control - DQN, DDPG, Webster's, Max-pressure, Self-Organizing Traffic Lights
License: GNU General Public License v3.0
Hello,
I would like to know whether the green_phase/yellow_phase duration in the fixed time traffic light was same as the action repeat duration in the RL based approaches.
Thank you.
This is not an issue. First of all, wonderful work. Second, as a Civil student, I use python but just want to know how I can learn to write code well structured like yours (e.g., define folders, define classes). Could you provide some resources for me to learn to code professionally?
After i ran the steps above, then Visualize results with: python graph_results.py, there are four methods, lack of the dqn method, and the "metrics" folder also lack of dqn folder.
Hi Wade,
After running the file, it comes this issue, No module named 'tensorflow.initializers',,
Is there some py file missing?
Hello, could you tell me how to display on the Sumo-gui.I don't know how to modify these code to realize it.
I'm looking forward to your reply.
Hi Dr. Genders,
thanks for the great framework!
I could not understand the code "node_id == '-13968'" in networkdata.py. Where does this number come from? Why do you handle it with some extra code?
Best,
Hello,
I am reading your code and papers. And I have a question. In your distprocs.py, you define a method and a class. In the function, you first use the Barrier. I have search the Barrier. Are you use the threading or Barrier class?
Best wishes!
Hello, Wade Genders phd, I have read your paper and run you code. And I have a question. Can you answer it?
I have learned the ddpg algorithm suits the continuous action space. But in the traffic signal control field, the action was define as (0, 1, 2 ,3 means phase selection) or (0, 1 means change phase or keep it).
In your paper, the proposed action space for the adaptive traffic signal controller is the duration of the next green phase in seconds. I guess you define the actor network outputs a decimal and then round it. Right?
So, could teach me the ddpg algorithm actions setting?
Dear Wade, there are some questions about the program.
After I ran' pythongraph _ results.py', I found that the error of DQN' distribution of travel times' in box graph was very large (the abnormal value of DQN reached more than 2,000, while SOTL was only about 1,500). I haven't adjusted any parameters. After several training sessions, it's all the same. Excuse me, what is the reason for this?
In addition, again, no parameters have been modified. After I run' python graph _ results.py', the values of' Queue' are all smaller than the source program (for example, the peak value is only 1500, while the source program exceeds 8000), while the values of' Delay' are consistent with the results of the source program. Excuse me, what is the reason for this?There is no modification program, but the running result is different from the source program.
Hi, I am pleasure to read your paper and interested in your program project. I configure the running environment on Ubuntu18.04 and try to run the experiments. My experiment environment is as follows:
Ubuntu18.04, Anaconda3-5.3.1-Linux-x86_64.sh, python3.7.5, tensorflow1.14.0, matplotlib3.1.1, sumo-1.12.0.
When I run "./train_dqn.sh", there is an error as follow:
I check the traci docs and don not find the connect() function. I add an configuration in .bashrc as "export LIBSUMO_AS_TRACI=1" and "source ~/.bashrc". But this error still exists.
I do not the reason of the problem. Could you help me?
start running main...
networks/double.net.xml
SUCCESSFULLY GENERATED NET DATA
Retrying in 1 seconds
Loading configuration... done.
Traceback (most recent call last):
File "run.py", line 17, in <module>
main()
File "run.py", line 10, in main
distprocs = DistProcs(args, args.tsc, args.mode)
File "/home/zlw/PycharmProjects/sumolights/src/distprocs.py", line 71, in __init__
netdata = sim.update_netdata()
File "/home/zlw/PycharmProjects/sumolights/src/sumosim.py", line 120, in update_netdata
tl_junc = self.get_traffic_lights()
File "/home/zlw/PycharmProjects/sumolights/src/sumosim.py", line 94, in get_traffic_lights
green_phases = [ p.state for p in logic.getPhases()
File "/home/zlw/PycharmProjects/sumolights/src/sumosim.py", line 95, in <listcomp>
if 'y' not in p.state
AttributeError: 'Phase' object has no attribute 'state'
Error: tcpip::Socket::recvAndCheck @ recv: peer shutdown
Quitting (on error).
my sumo version is 1.2.0
Hello,docwza. I have some real traffic data here. I want to use the rou.xml file of sumo as the vehicle input. Is this feasible. Can I complete my requirements by modifying file vehiclegen
hello! I have read your program, but when I read the Neuralagent and RLagent. I did not find them. So, could you tell me the RLagent directory?
Thank you!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 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. 📊📈🎉
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 ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.