Tmrl

Latest version: v0.6.6

Safety actively analyzes 638379 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 2 of 5

0.6.0

This release introduces support for non-real-time environments in the TMRL library, and support for Linux in the TrackMania pipeline.

Version `0.6.0` is backward-incompatible and requires a clean installation.

Major changes

- TrackMania example pipeline (see `config.json`):
- Support for Linux
- Support for saving replays automatically
- Support for reward shaping
- TMRL library:
- Support for non-real-time environments and Trainer/Worker synchronization
- Generic training pipeline (in particular, introduced a generic `Memory` class for lazy developers, compatible with random sampling in 1-step TD learning)

Minor changes
- TrackMania example pipeline (see `config.json`):
- More fine-tuning options for SAC
- Set the default Adam Betas to the RL-compatible setting described by [Mahmood et al. 2023](https://openreview.net/forum?id=m9Jfdz4ymO) in an attempt to avoid policy collapse
- The default hyperparameters changed for full vision-based training instead of LIDAR training

0.5.3

This update introduces compatibility fixes with latest versions of libraries such as `pandas` and `torch`.

Linux users: patience, `tmrl 0.6.0` is on its way... 😎

0.5.2

Minor release

Release 0.5.2 fixes a bug in the SAC implementation provided in version 0.5.1, which had been inadvertently pushed from a development branch.

_Note: this version introduces a new way of recording replays in TrackMania, but this is not officially supported yet and won't work without the corresponding OpenPlanet script. If you want to use this feature before the next version is out, please contact us._

0.5.1

---

This release complies with `rtgym>=0.9`, which in turns complies with the `gymnasium` signature of the `reset` function.

In case you are using custom `rtgym` interfaces in `tmrl`, you will want to update your `reset` implementations. This is straightforward, you can just replace:

python
def reset(self):

with:
python
def reset(self, seed=None, options=None):

0.5.0

---

Release `0.5.0` upgrades the `tmrl` framework to the the maintained [gymnasium](https://gymnasium.farama.org) framework, instead of the old and abandoned `gym` framework.

This release follows the changes made in [rtgym](https://github.com/yannbouteiller/rtgym/releases/tag/v0.8).

Version `0.5.0` does not introduce any breaking change, but you still need to delete your checkpoints from previous versions in `TmrlData/checkpoints` (or you can perform a clean install by deleting the `TmrlData` folder prior to pip-installing version `0.5.0`), since older versions of the framework use to import `gym` where it now imports `gymnasium`.

0.4.2

---

This update improves TrackMania support (thanks LaurensNeinders !)

**Perform a clean install by deleting the `TmrlData` folder AND `Plugin_GrabData_0_1.as` from `OpenPlanetNext\Scripts`**

Release 0.4.2 ships with an optimized version of the OpenPlanet script for the default TrackMania environments. In particular, it should alleviate a common issue where the environment randomly times-out for a roughly constant amount of time even on high-end PCs.

Page 2 of 5

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.