Sectar

KO | EN

gitlite — search

#python #react #typescript #golang #go #machine-learning #rust #linux #nodejs #java #cli #openai #mcp #open-source

(★ 96)

Self-Consistent Trajectory Autoencoder: Hierarchical Reinforcement Learning with Trajectory Embeddings

Sectar 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# SeCTAr
Code for [Self-Consistent Trajectory Autoencoder: Hierarchical Reinforcement Learning with Trajectory Embeddings](https://sites.google.com/view/sectar/)

## Running
Experiment scripts are in `exps/`, environments are located under `traj2vec/envs`, main algorithms are in `traj2vec/algos/vaepdentropy.py` and `traj2vec/algos/vae_bc.py`

## Installation
* Download
  * [sectar](https://github.com/wyndwarrior/Sectar)
  * [rllab](https://github.com/rll/rllab)
  * [doodad](https://github.com/justinjfu/doodad)
  * [baselines](https://github.com/openai/baselines)
* Add these repos to your python path and follow instructions for setting them up.
* Install Mujoco [instructions here](https://github.com/openai/mujoco-py)
* Modify `traj2vec/launchers/config.py` to point to the appropiate paths.

* Create the conda env sectar with
```
conda env create -f environment.yml
```

## Logging
The log dir for the scripts is set to `data`. You can plot recorded results by giving the exp log dir to ```traj2vec/viskit/frontend.py```.

Sectar

Sectar 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Sectar

Sectar 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README