eirli
DI-drive
eirli | DI-drive | |
---|---|---|
1 | 2 | |
37 | 521 | |
- | -16.1% | |
2.8 | 0.0 | |
about 1 year ago | over 1 year ago | |
Python | Python | |
- | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
eirli
-
Image Dataset from D4RL for Mujoco Tasks
I am not aware of any for DM control tasks. Might worth looking at this if you’re interested in other environments.
DI-drive
- Try simple interfaces and customized driving policy and casezoo set on DI-drive!
-
Is reinforcement learning being used for the development of self-driving cars?
Some attempts on driving simulators have achieved good results(eg. DI-drive, DI-drive is an open-source application platform under OpenDILab. DI-drive applies different simulator/datasets/cases in Decision Intelligence Training & Testing for Autonomous Driving Policy). The basic idea mainly includes initializing with imitation learning, and then using reinforcement learning to obtain results that surpass expert data after reaching a certain performance. Some use the perceptual Label to train the backbone of the network, then freeze the backbone, and use reinforcement learning to specifically train the affordance method from perceptual embedding to action output. Others use a multi-model fusion approach, in which the model trained by reinforcement learning is used together with other methods to obtain the driving output. However, the emulator-based method is mainly end-to-end, and its security is difficult to guarantee, and it is difficult to apply to real vehicle scenarios.
What are some alternatives?
calibrated-backprojection-network - PyTorch Implementation of Unsupervised Depth Completion with Calibrated Backprojection Layers (ORAL, ICCV 2021)
mini-AlphaStar - (JAIR'2022) A mini-scale reproduction code of the AlphaStar program. Note: the original AlphaStar is the AI proposed by DeepMind to play StarCraft II. JAIR = Journal of Artificial Intelligence Research.
dino - PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO
imitation - Clean PyTorch implementations of imitation and reward learning algorithms
PaddleHelix - Bio-Computing Platform Featuring Large-Scale Representation Learning and Multi-Task Deep Learning “螺旋桨”生物计算工具集
tianshou - An elegant PyTorch deep reinforcement learning library.
pytorch-metric-learning - The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.
neat - [ICCV'21] NEAT: Neural Attention Fields for End-to-End Autonomous Driving
f-IRL - Inverse Reinforcement Learning via State Marginal Matching, CoRL 2020
pytorch_sac_ae - PyTorch implementation of Soft Actor-Critic + Autoencoder(SAC+AE)