code-for-paper's People
Forkers
wwxfromtju bmpcc6k chloechsu sweetice srinivr tangzk zmonoid zhoubin-me phoenixera tianqi-777 xiaogengyaokeyan veilcooper lars12llt hejichao2020 curieuxjy bellz867 atlasgooo2 sanelyx jucaleb4 ltcsoar umfundii martain-liucode-for-paper's Issues
Which version of mujoco used?
Can you update the version of the libraries in the requirements.txt? Thank you!
License
Hi,
I would like to use the code for my research. Is it possible, that a license e.g. MIT is added to the repository?
Thanks.
Code for value function clipping
In the paper (sec 2) and description of value function clipping, the PPO-like objective is the minimum of clipped and unclipped. However, the code applies the maximum of them, does it have different effects from the paper said?
code-for-paper/src/policy_gradients/steps.py
Lines 89 to 96 in 094994f
Square after max for clipped value_gae_loss, any comment?
The original implementation first squares both the clipped and unclipped losses then takes the maximum. Any comment as to why this is handled differently here?
code-for-paper/src/policy_gradients/steps.py
Line 100 in 094994f
Issues with the RewardFilter
Hi, thanks for the great work. I have three questions if you don't mind.
- In
,
the comments suggest it usesIncorrect reward normalization
. I was wondering if you could elaborate. Does that mean we should avoid usingRewardFilter
because of the incorrect normalization and try to useZfilter
instead for the reward normalization?
Another concern I have is with the reset()
call of the RewardFilter
. It seems that in your customized envs,
def reset(self):
# Reset the state, and the running total reward
start_state = self.env.reset()
self.total_true_reward = 0.0
self.counter = 0.0
self.state_filter.reset()
return self.state_filter(start_state, reset=True)
-
It seems the
reward_filter
will never reset. However, thereward_filter
always multiply the existing returns bygamma
. Could this be a bug? -
The
reward_filter
is already using thegamma
as part of its inputs, but do you still calculate the advantage using thegamma
again or is this somehow omitted?
Thanks.
The result of PPO-M
When I run the PPO-M with the default parms in MuJoCo.json,the average mean_reward of the last train step over 10 agents is much smaller than the result in paper.In Walker2d-v2,my result of PPO-M is about 600-900.And in Hopper-v2,the result is about 300. In Humanoid-2,it always fail to run using the default ppo_lr_adam=1e-4
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
๐ 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. ๐๐๐
-
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 โค๏ธ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.