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

pipeline_tag: reinforcement-learning

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

The following normlization coeficients are required to use this model:

mean = [-0.06845774,  0.01641455, -0.18354906, -0.27624607, -0.34061527, -0.09339716, -0.21321271, -0.08774239,  5.1730075,  -0.04275195, -0.03610836,  0.14053793, 0.06049833,  0.09550975,  0.067391,    0.00562739,  0.01338279] 

std = [0.07472999,  0.30234998,  0.3020731, 0.34417078, 0.17619242,  0.5072056,  0.25670078,  0.32948127,  1.2574149,   0.7600542,   1.9800916,   6.5653625,  7.4663677,   4.472223, 10.566964, 5.6719327,   7.498259]

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.