mbrl-lib
mbrl-lib copied to clipboard
facebookresearch
Library for Model Based RL
TODO for this WIP PR: - [x] New PID based / linear feedback agent(s) - [ ] Make PID accept vector inputs - [x] Training example - [ ] Migrate...
# 🚀 Feature Request When training non delta-state models, the outputs of dynamics models can take large values (way outside a unit Gaussian). In the past I have tried using...
## Types of changes - [X] Docs change / refactoring / dependency upgrade - [X] Bug fix (non-breaking change which fixes an issue) - [X] New feature (non-breaking change which...
Dreamer
## Types of changes Still a WIP, but I've managed to add most of Dreamer. The main thing left is computing the loss in planning/dreamer_agent/DreamerAgent::train(). - [X] Docs change /...
# 🚀 Feature Request Change `BasicEnsemble` to optionally use `functorch.vmap`. ## Motivation and Pitch **Is your feature request related to a problem? Please describe.** `BasicEnsemble` lets the user provide arbitrary...
## Types of changes - [ ] Docs change / refactoring / dependency upgrade - [x] Bug fix (non-breaking change which fixes an issue) - [ ] New feature (non-breaking...
## Observed mbpo.py - line 199 in the for loop that is incrementing the steps of the current epoch, the steps_epoch iteration variable is not being reset after we observe...
# 🚀 Feature Request Add support for Dreamer and Dreamerv2 ## Motivation Dreamer and Dreamerv2 are important and current SOTA in mbrl. Few standard implementations are available for both Dreamer...
# 🚀 Feature Request Add a batch dimension to all operations in `ModelEnv.evaluate_action_sequences` so that it can find rewards for several trajectories in parallel, each starting at a different observation....