Comments (2)
-
Hello, if your main goal is to use the policy in
collect_mode
for episode collection, keep in mind that the collected data will usually be utilized for learning and training. -
Consequently, value transformations are usually required, as indicated by: "
pred_values = self.inverse_scalar_transform_handle(pred_values).detach().cpu().numpy()
". Hence, a common recommended approach is(b)
: initialize anInverseScalarTransform
object in the_init_collect()
method, as depicted below:
self.inverse_scalar_transform_handle = InverseScalarTransform(
self._cfg.model.support_scale, self._cfg.device, self._cfg.model.categorical_distribution
)
-
If you want to delve deeper into the principles behind value transformations, you could refer to the corresponding academic paper. This paper offers a comprehensive theoretical basis and case studies, providing significant support in understanding and applying value transformation techniques.
-
Of cource, If the collected data is not used for learning, you can consider alternative solution
(c)
.
from lightzero.
Hi @puyuan1996 !
Thank you very much for the detailed information!
I will follow with (b).
Best regards,
David.
from lightzero.
Related Issues (20)
- how to well model a grid env when it changes frequently? HOT 9
- Minigrid environment HOT 1
- Great!!! HOT 1
- Running a trained agent HOT 2
- Hyperparameter of Muzero and reproducibility of the results HOT 2
- Render and plot results: is there a code snippet? HOT 1
- Bad performance on long run on MsPacman and SpaceInvaders HOT 10
- Instillation Failure on Mac (Spyder) HOT 3
- Could you provide detail code example to customize the env and algo that can run successfully in a entire file? HOT 7
- Clipping reward in Atari while using invertible transform for reward and value target HOT 2
- Issues related to the operating environment HOT 4
- Clarification on the relationship between num_unroll_steps and infer_context_length in UniZero HOT 9
- Need Help with Storing and Using Best Checkpoint with ONNX Model in LightZero HOT 1
- Unexpected performance drop after resuming UniZero training HOT 3
- Implementation of Self-Play Training for Real-Time Environments HOT 2
- Flexible Input Image Size Support for UniZero: Implementation Timeline and Contribution Opportunity HOT 5
- Inquiry about Commented Out Loss Calculations in UniZero Implementation HOT 2
- Unexpected Negative Values for Perceptual Loss in UniZero Implementation HOT 3
- When will Chess be supported for AlphaZero/MuZero? HOT 7
- Discussion: Requesting assistance and guidance with implementation of RL algorithms and models in the context of Tetris HOT 3
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 lightzero.