Comments (8)
mw66
commented on May 18, 2024
BTW, it's on CPU since no GPU available.
from pytorch-a2c-ppo-acktr-gail.
ikostrikov
commented on May 18, 2024
Oh. I think I know what's wrong. Monitor didn't flush the results.
As a temporary solution you can wrap visdom_plot into try ... catch IOError: pass
I should avoid using logs for plotting. I will try to find a way to get numbers directly from the monitor.
from pytorch-a2c-ppo-acktr-gail.
ikostrikov
commented on May 18, 2024
Fixed in b8d4c2c
from pytorch-a2c-ppo-acktr-gail.
SerialIterator
commented on May 18, 2024
I am having this problem as well. I see it has been fixed and my main.py file is current. This is the error that matches the above although my string is much shorter.
visualize.py", line 65, in load_data
tmp = [t_time, int(tmp[1]), float(tmp[0])]
ValueError: could not convert string to float: '\x00-20.0'
I have never used visdom before but I run it using a seperate terminal window and see it open in browser. I see a new socket opened but nothing changes in browser. I am running Pong with PPO as suggested in the readme file and it trains without --vis but I can't figure out how to see anything using visdom even when I directly run the visualize.py file (which is missing num_steps arg in the visdom_plot call at the bottom) where I get the same ValueError.
from pytorch-a2c-ppo-acktr-gail.
IsaacKam
commented on May 18, 2024
I have the same issue as SerialIterator above, never manages to load the data , nothing pops up on visdom. Can't seem to find the monitor.csv files ( or where data is loaded into them). Thanks in advance for any advice!
from pytorch-a2c-ppo-acktr-gail.
ikostrikov
commented on May 18, 2024
I'm planning to deprecate visdom since there are better ways to visualize the data now.
from pytorch-a2c-ppo-acktr-gail.
IsaacKam
commented on May 18, 2024
Hi Ilya,
Thanks for this! I have found a one line efficient solution to the problem discussed above which could be beneficial to add to the repo. If you replace approx line 63 tmp = line.split(‘,’) with tmp = line.replace('\x00','').split(',’). It gets rid of the Hex codes, allowing the data to be loaded correctly. Hope it helps,
Isaac
On 24 Feb 2019, at 16:34, Ilya Kostrikov <[email protected]<mailto:[email protected]>> wrote:
I'm planning to deprecate visdom since there are better ways to visualize the data now.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub<#18 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AbmPYUSsWWTXYC1hO2HptFX64uUG0LiEks5vQr8rgaJpZM4QSvws>.
from pytorch-a2c-ppo-acktr-gail.
ikostrikov
commented on May 18, 2024
Great! Could you create a pull request with the bug fix?
from pytorch-a2c-ppo-acktr-gail.
Related Issues (20)
- does mask introduce bias in the gail implementation ?
- observation reset before insert
- Why acktr algorithm cannot be used in Mujoco settings?
- Can I train in my own game
- Can not run enjoy.py
- Stale hidden states
- Possible bug on the sign of policy log prob. in Fisher computation
- CNN Architecture
- Operations that have no effect
- No softmax before categorical loss?
- object has no attribute 'steps' in acktr
- why PPO needs to store action_log_probs instead of using stop_gradient for better efficiency? HOT 1
- [Question]Can I use Recurrent_policy for GAIL at this implementation?
- question about the recurrent HOT 1
- Oops! wrong repo :-D HOT 1
- setup.py and requirements.py have same dependencies except for h5py
- Where are the experts data for GAIL get from?
- Why is episode_rewards negative when running ant_v3 with PPO?
- Why didn't run to generate log?
- Updates: Support the latest Atari environment and state entropy maximization-based exploration.
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pytorch-a2c-ppo-acktr-gail.