Denys88/rl_games issues and pull requests

#319 - Question regarding custom network

Issue - State: closed - Opened by SourabhPrasad about 1 month ago - 5 comments

#318 - Could you please give a figure to explain the relationship between model-model_builder-network-network_builder?

Issue - State: open - Opened by KJaebye about 2 months ago

#317 - #316 Fixed lstm demo

Pull Request - State: closed - Opened by Denys88 2 months ago

#316 - [Bug] Need to pass old RNN states while running onnx inference

Issue - State: closed - Opened by VineetTambe 3 months ago - 2 comments

#315 - WIP: Update

Pull Request - State: open - Opened by ViktorM 3 months ago

#314 - Added a few features to make rl_games more comparable to the brax with jax ppo

Pull Request - State: closed - Opened by Denys88 3 months ago

#313 - fixing get_mean_std_with_masks

Pull Request - State: closed - Opened by ZhengyiLuo 3 months ago

#312 - WIP: Myosuite and maniskill support

Pull Request - State: open - Opened by ViktorM 4 months ago

#311 - Brax fixes

Pull Request - State: open - Opened by amrzv 4 months ago - 1 comment

#310 - [WIP] Added MoE layer

Pull Request - State: open - Opened by Denys88 4 months ago

#309 - [WIP] Mixture of Experts MLP

Pull Request - State: closed - Opened by Denys88 4 months ago

#308 - [WIP ]Added data generation for offiline training

Pull Request - State: open - Opened by Denys88 4 months ago

#307 - [WIP] selfplay poker test

Pull Request - State: open - Opened by Denys88 5 months ago

#306 - Added myo suite support

Pull Request - State: closed - Opened by Denys88 5 months ago

#305 - Documation about class implemetations, customization of training/playing loops and models and networks

Pull Request - State: closed - Opened by ErcBunny 5 months ago - 2 comments

#304 - WIP: Torch upgrade. Aux loss. Vision backbone networks.

Pull Request - State: open - Opened by ViktorM 5 months ago

#303 - Added aux loss

Pull Request - State: closed - Opened by Denys88 5 months ago

#302 - Migrate brax v2 in notebook brax_training

Issue - State: open - Opened by gonzalesMK 6 months ago - 3 comments

#302 - Migrate brax v2 in notebook brax_training

Issue - State: open - Opened by gonzalesMK 6 months ago - 1 comment

#301 - How to use custom network?

Issue - State: open - Opened by ErcBunny 6 months ago - 8 comments

#301 - How to use custom network?

Issue - State: open - Opened by ErcBunny 6 months ago - 8 comments

#300 - How to load the policy model in the real robot ？

Issue - State: closed - Opened by heartInsert 7 months ago - 1 comment

#300 - How to load the policy model in the real robot ？

Issue - State: open - Opened by heartInsert 7 months ago - 1 comment

#299 - Clamped Actions Between [-1,1]

Issue - State: closed - Opened by rogerpm3 7 months ago

#298 - Update for tacsl release: CNN tower processing, critic weights loading and freezing.

Pull Request - State: closed - Opened by iakinola23 7 months ago

#297 - Ensure the consistency of central_value_net with the same initial parameters before training starts in a multi-GPU setting.

Pull Request - State: closed - Opened by annan-tang 8 months ago - 4 comments

#296 - Potential Issues with Multi-GPU/Node Training with Central Network Weights Initialization

Issue - State: closed - Opened by annan-tang 8 months ago

#295 - Increased time resolution for more precision performance tracking.

Pull Request - State: closed - Opened by ViktorM 8 months ago

#295 - Increased time resolution for more precision performance tracking.

Pull Request - State: closed - Opened by ViktorM 8 months ago

#294 - [Q] Benchmark of PPO / SAC on continuous tasks like envs in IsaacGym

Issue - State: closed - Opened by Rowing0914 8 months ago - 3 comments

#294 - [Q] Benchmark of PPO / SAC on continuous tasks like envs in IsaacGym

Issue - State: closed - Opened by Rowing0914 8 months ago - 3 comments

#293 - where is train.py?

Issue - State: closed - Opened by Rowing0914 8 months ago - 2 comments

#292 - New release?

Issue - State: open - Opened by llinauer 8 months ago - 1 comment

#291 - Fix SAC with input normalization

Pull Request - State: closed - Opened by llinauer 8 months ago - 1 comment

#290 - Fix SACPlayer deterministic action

Pull Request - State: closed - Opened by llinauer 8 months ago

#289 - SACPlayer deterministic mixup

Issue - State: closed - Opened by llinauer 8 months ago - 2 comments

#288 - Remove duplicate lines from a2c_common.py

Pull Request - State: closed - Opened by paLeziart 8 months ago - 2 comments

#287 - Fixed minibatch_size_per_env bug

Pull Request - State: closed - Opened by ViktorM 8 months ago

#286 - minibatch_size_per_env bug ?

Issue - State: open - Opened by Yougo-robotics 9 months ago - 1 comment

#286 - minibatch_size_per_env bug ?

Issue - State: closed - Opened by Yougo-robotics 9 months ago - 2 comments

#285 - How does the a2c network rescale the range of the action to [-1,1]

Issue - State: closed - Opened by des-zhong 9 months ago - 3 comments

#284 - fixed bug with multi-gpu training for continuous a2c in a2c_common.py

Pull Request - State: closed - Opened by tttonyalpha 10 months ago - 4 comments

#284 - fixed bug with multi-gpu training for continuous a2c in a2c_common.py

Pull Request - State: open - Opened by tttonyalpha 10 months ago - 3 comments

#283 - Difference between network and model

Issue - State: open - Opened by kaijieshi7 10 months ago - 1 comment

#282 - Discrete and Continuous

Issue - State: closed - Opened by mxllc 11 months ago - 2 comments

#281 - Normalization implementation is actually computing standardization

Issue - State: open - Opened by anishhdiwan 11 months ago - 2 comments

#280 - missing python 3.11 support

Issue - State: open - Opened by igozali 11 months ago - 3 comments

#279 - question about central value

Issue - State: open - Opened by JoseBarreiros-TRI 11 months ago - 1 comment

#278 - Which version has this '.model_builder.model_factory.register_builder'

Issue - State: open - Opened by dccdcc7 11 months ago - 4 comments

#278 - Which version has this '.model_builder.model_factory.register_builder'

Issue - State: open - Opened by dccdcc7 11 months ago

#277 - question about init_dones tensor in a2c_common

Issue - State: open - Opened by WSYANGSX 11 months ago - 1 comment

#277 - question about init_dones tensor in a2c_common

Issue - State: open - Opened by WSYANGSX 11 months ago

#276 - Adding categorical cross_entropy loss to yaml config.

Issue - State: open - Opened by Eirikalb 11 months ago - 1 comment

#276 - Adding categorical cross_entropy loss to yaml config.

Issue - State: open - Opened by Eirikalb 11 months ago - 1 comment

#275 - print_statistics() output of ContinuousA2CBase() might be wrong due to frame update implementation?

Issue - State: open - Opened by yutaizhou 12 months ago

#274 - How to understand the swap_and_flattern01() operation

Issue - State: closed - Opened by BadWomanY about 1 year ago - 1 comment

#273 - Add docstrings

Pull Request - State: closed - Opened by anishhdiwan about 1 year ago

#272 - Ad/documentation

Pull Request - State: closed - Opened by anishhdiwan about 1 year ago - 6 comments

#271 - Requesting docs for rl_games

Issue - State: open - Opened by anishhdiwan about 1 year ago - 1 comment

#270 - How to Correctly Integrate LSTM or GRU into the SAC Algorithm

Issue - State: open - Opened by namjiwon1023 about 1 year ago - 1 comment

#269 - How to use Isaac trained customized models (different number of observations and outputs)

Issue - State: open - Opened by DannyChen1994 about 1 year ago - 1 comment

#268 - How to specify the GPU that loads the model weights when applying to IsaacGym

Issue - State: open - Opened by Charlie0257 about 1 year ago - 2 comments

#267 - Added gymnasium and gym 0.26 support

Pull Request - State: open - Opened by Denys88 about 1 year ago

#266 - How can I export a trained neural network?

Issue - State: closed - Opened by jheyeo222 about 1 year ago - 10 comments

#265 - Fixed multi discrete with masks

Pull Request - State: closed - Opened by Denys88 about 1 year ago

#264 - MultiDiscrete action_mask cannot run

Issue - State: closed - Opened by xialuanshi over 1 year ago - 3 comments

#263 - AttributeError: 'dict' object has no attribute 'clone' in SAC algorithm

Issue - State: closed - Opened by scxzhchacker over 1 year ago - 2 comments

#262 - Getting error while testing

Issue - State: closed - Opened by qureshinomaan over 1 year ago - 4 comments

#261 - Is there any plans to move to gymnasium?

Issue - State: open - Opened by Mayankm96 over 1 year ago - 2 comments

#260 - 2023 10 16 rnn concat output

Pull Request - State: closed - Opened by tylerlum over 1 year ago

#259 - fix sac obs error

Pull Request - State: closed - Opened by kellyguo11 over 1 year ago

#258 - cleanup initial commit

Pull Request - State: closed - Opened by Denys88 over 1 year ago

#257 - Bounds / regularisation loss and action clipping

Issue - State: closed - Opened by johannespitz over 1 year ago - 2 comments

#256 - Infer trained neural network manually

Issue - State: closed - Opened by BaranAugustus over 1 year ago - 1 comment

#255 - Is there any description of the structure of this package?

Issue - State: open - Opened by 3505473356 over 1 year ago - 3 comments

#254 - Updates

Pull Request - State: open - Opened by ViktorM over 1 year ago

#253 - Ray dependency and seq_length improvements

Pull Request - State: closed - Opened by ViktorM over 1 year ago

#252 - When to use Model or Network?

Issue - State: closed - Opened by HGGshiwo over 1 year ago - 1 comment

#251 - Added get/set_param functions to RL algos to support fast pbt

Pull Request - State: closed - Opened by ViktorM over 1 year ago

#250 - advantage in a2c_common

Issue - State: closed - Opened by mxllc over 1 year ago - 1 comment

#249 - Possibilities to use pointcloud as input for the SAC

Issue - State: closed - Opened by robinvetsch over 1 year ago - 1 comment

#248 - Possibility of including Multi Agent Reinforcement Learning

Issue - State: closed - Opened by berttggg over 1 year ago - 1 comment

#247 - loading checkpoint doesn't work for resuming training and test

Issue - State: open - Opened by qiuyuchen14 over 1 year ago - 10 comments

#246 - add option for evaluating checkpoint

Pull Request - State: closed - Opened by kellyguo11 over 1 year ago - 1 comment

#245 - Is Regularizer option working ?

Issue - State: closed - Opened by arminvburren over 1 year ago - 1 comment

#244 - Update README.md

Pull Request - State: closed - Opened by ViktorM over 1 year ago

#243 - Externally Load Checkpoint

Issue - State: closed - Opened by davesarmoury over 1 year ago - 10 comments

#242 - Question about value_bootstrap

Issue - State: closed - Opened by johannespitz over 1 year ago - 4 comments

#241 - Bug in network_buider

Issue - State: open - Opened by xuanyaoming over 1 year ago - 5 comments

#240 - Multi-node training support

Pull Request - State: closed - Opened by ViktorM over 1 year ago

#239 - How to implement a new RL algorithm?

Issue - State: closed - Opened by xuanyaoming over 1 year ago - 4 comments

#238 - save_frequency works now with SAC as well

Pull Request - State: closed - Opened by ViktorM almost 2 years ago

#237 - Checkpoint when using SAC

Issue - State: closed - Opened by Baerenstark2 almost 2 years ago - 1 comment

#236 - added ppo-ue

Pull Request - State: open - Opened by Denys88 almost 2 years ago

#235 - small updates for dm

Pull Request - State: closed - Opened by Denys88 almost 2 years ago

#234 - MAPPO with isaac gym

Issue - State: closed - Opened by Asad-Shahid almost 2 years ago - 2 comments

#233 - Draft: VM/torch2

Pull Request - State: open - Opened by ViktorM almost 2 years ago

#232 - self.minibatch_size

Issue - State: open - Opened by 1tac11 almost 2 years ago - 9 comments

#231 - DM Control and Bug Fixes

Pull Request - State: closed - Opened by Denys88 almost 2 years ago

#230 - Information about documentations

Issue - State: open - Opened by 403forbiddennn almost 2 years ago - 1 comment

GitHub / Denys88/rl_games issues and pull requests