Light

mdeib / berkeley-deep-rl-pytorch-solutions Goto Github PK

View Code? Open in Web Editor NEW

117.0 2.0 24.0 22.48 MB

Pytorch solutions for UC Berkeley's cs285 assignments

Python 100.00%

cs285 reinforcement-learning pytorch-rl

berkeley-deep-rl-pytorch-solutions's Introduction

UC Berkeley Deep RL Pytorch Solutions

Pytorch solutions for UC Berkeley's CS285 Deep RL course. If you wish to complete the assignments yourself, a pytorch version of the official starter code has also been made.

While these solutions have produced reasonable results be aware that there may still be small bugs in the code and/or the solutions.

berkeley-deep-rl-pytorch-solutions's People

Contributors

Stargazers

Watchers

berkeley-deep-rl-pytorch-solutions's Issues

Average the loss?

berkeley-deep-RL-pytorch-solutions/hw2/cs285/policies/MLP_policy.py

Line 115 in 47da611

loss = torch.sum((-logprob_pi * torch.Tensor(adv_n).to(self.device)))

Shall the loss be averaged by N? I apologize if I am wrong. Do not have much experience with RL. Thanks.

It seems to be another loop instead of "mod"

berkeley-deep-RL-pytorch-solutions/hw3/cs285/critics/bootstrapped_continuous_critic.py

Line 55 in 7f2a5e1

if update % self.num_grad_steps_per_target_update == 0:

Question about hw1 Q1.2 BC Agent Analysis

As the picture shown here, I'm wondering how tensorboard can draw different event data with user horizontal axis...
horizontal axis by different train step inside the tensorboard
since you need to run different loops and it's different log files...

Excuse me, Some trouble with training time.

hello, I met some problems in hw2 training HalfCheetah with batch size 30000, It takes me just a few seconds to collect data in hw3 within same batch size, but In hw2, it will take several minutes, which is really confusing, would you please tell me what may be wrong with it ? Thank you !

It seems to be '+=' instead of '+'

berkeley-deep-RL-pytorch-solutions/hw2/cs285/agents/pg_agent.py

Line 129 in edbbc6d

indice = len(rew_list)

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.