---
tags:
- deep-reinforcement-learning
- reinforcement-learning
- decision-transformer
- gym-continous-control

---
# Decision Transformer model trained on expert trajectories sampled from the Gym Walker2d environment
This is a trained [Decision Transformer](https://arxiv.org/abs/2106.01345) model trained on expert trajectories sampled from the Gym Walker2d environment

<video src="https://huggingface.co/edbeeching/decision-transformer-gym-walker2d-expert/resolve/main/walker2d-expert.mp4" controls autoplay loop></video>

The following normlization coeficients are required to use this model:

mean = [ 1.2384834e+00,  1.9578537e-01, -1.0475016e-01, -1.8579608e-01,  2.3003316e-01,  2.2800924e-02, -3.7383768e-01, 3.3779100e-01,  3.9250960e+00, -4.7428459e-03,  2.5267061e-02, -3.9287535e-03, -1.7367510e-02, -4.8212224e-01, 3.5432147e-04, -3.7124525e-03,  2.6285544e-03] 

std = [0.06664903, 0.16980624, 0.17309439, 0.21843709, 0.74599105, 0.02410989, 0.3729872,  0.6226182,  0.9708009,  0.72936815, 1.504065,   2.495893, 3.511518,   5.3656907, 0.79503316, 4.317483,   6.1784487]
See our [Blog Post](https://colab.research.google.com/drive/1K3UuajwoPY1MzRKNkONNRS3gS5DxZ-qF?usp=sharing),  [Colab notebook](https://colab.research.google.com/drive/1K3UuajwoPY1MzRKNkONNRS3gS5DxZ-qF?usp=sharing) or [Example Script](https://github.com/huggingface/transformers/tree/main/examples/research_projects/decision_transformer) for usage.