Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / rail-berkeley/softlearning issues and pull requests

#92 - Bug in HER replay pool (and multi goal setup not finished)

Issue - State: open - Opened by jendelel over 5 years ago - 2 comments

#91 - Misc tweaks 20190601

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#90 - Simulate policy not working anymore

Issue - State: closed - Opened by kapsl over 5 years ago - 1 comment

#89 - invalidgitrepository error

Issue - State: closed - Opened by surbhi1944 over 5 years ago - 4 comments

#88 - Module 'gym' has no attribute 'register' on MacOS Mojave 10.14.4

Issue - State: closed - Opened by weijiafeng over 5 years ago - 8 comments

#87 - `run_example_debug` on Mac OS doesn't save videos to the right directory

Issue - State: open - Opened by justinvyu over 5 years ago - 1 comment

#86 - Usability enhancements for the RL pipeline

Pull Request - State: closed - Opened by hartikainen over 5 years ago - 3 comments

#85 - Refactor/remove serializable from softlearning env

Pull Request - State: closed - Opened by hartikainen over 5 years ago - 1 comment

#84 - Set the seeds for both training and eval environments

Pull Request - State: open - Opened by quanvuong over 5 years ago

#83 - Initial exploration steps differ between Mujoco envs like TD3 ?

Issue - State: closed - Opened by zuoxingdong over 5 years ago - 1 comment

#82 - Hierarchical training and reward set

Issue - State: closed - Opened by JakobSchmitt over 5 years ago - 1 comment

#81 - No module named 'softlearning.utils'

Issue - State: closed - Opened by kracon7 over 5 years ago - 5 comments

#80 - Environment seeding doesn't allow reproducibility

Issue - State: open - Opened by quanvuong over 5 years ago - 3 comments

#79 - Restore does not work at all anymore

Issue - State: open - Opened by kapsl over 5 years ago - 5 comments

#78 - Misc fixes 20190424

Pull Request - State: closed - Opened by hartikainen over 5 years ago - 1 comment

#77 - Bound std of Gaussian policy via beta-sigmoid ?

Issue - State: closed - Opened by zuoxingdong over 5 years ago - 6 comments

#76 - SAC Hyperparameters MountainCarContinuous-v0 - Env with deceptive reward

Issue - State: open - Opened by araffin over 5 years ago - 10 comments

#75 - Unable to reproduce result on HalfCheetah-v2

Issue - State: closed - Opened by quanvuong over 5 years ago - 18 comments

#74 - Replay pools should allow saving environment info

Issue - State: open - Opened by hartikainen over 5 years ago

#73 - Import error. Trying to rebuild mujoco_py.

Issue - State: open - Opened by szrlee over 5 years ago - 6 comments

#72 - [WIP] Robosuite Control

Pull Request - State: closed - Opened by Alacarter over 5 years ago - 5 comments

#71 - Fix bibtext in README

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#69 - Progress on Deepmind control suite?

Issue - State: closed - Opened by quanvuong over 5 years ago - 3 comments

#68 - Fix requirements.txt syntax

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#67 - Add missing google compute engine package

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#66 - Misc improvements

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#65 - [Enhancement] Make replay buffer memory-efficient

Issue - State: open - Opened by avisingh599 over 5 years ago

#64 - unstable training curve for default SQL

Issue - State: open - Opened by LuhuanWu over 5 years ago - 1 comment

#63 - What does it mean to set 'eval_deterministic' = True for SQL?

Issue - State: open - Opened by AlphaFrank over 5 years ago - 6 comments

#62 - Union pool

Issue - State: closed - Opened by tianheyu927 over 5 years ago - 1 comment

#61 - How to set random seed?

Issue - State: closed - Opened by quanvuong over 5 years ago - 1 comment

#60 - Stop gradient for alpha in non alpha losses

Issue - State: closed - Opened by jendelel over 5 years ago - 2 comments

#59 - Resume training

Issue - State: open - Opened by kapsl over 5 years ago - 7 comments

#58 - Fix image_shape key error

Pull Request - State: closed - Opened by vitchyr over 5 years ago - 1 comment

#57 - [wip] Deepmind control suite

Pull Request - State: closed - Opened by Alacarter over 5 years ago - 1 comment

#56 - Run on real robot

Issue - State: open - Opened by kapsl over 5 years ago - 1 comment

#55 - delete certifi from requirements.txt

Pull Request - State: closed - Opened by Alacarter over 5 years ago - 1 comment

#54 - Fix MultiGoal example

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#53 - Feature/independent evaluation env

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#52 - Resurrect SquashBijector due to numerical instability in tfb.Tanh

Pull Request - State: closed - Opened by hartikainen over 5 years ago - 1 comment

#51 - Version updates

Pull Request - State: closed - Opened by hartikainen over 5 years ago - 2 comments

#50 - Update gym version

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#49 - Environment sampling should be parallelized

Issue - State: open - Opened by hartikainen over 5 years ago - 1 comment

#48 - Fix Point2DEnv-Default-v0 entrypoint

Pull Request - State: closed - Opened by hartikainen over 5 years ago - 1 comment

#47 - MultivariateNormalDiag log_prob with target_entropy and alpha

Issue - State: closed - Opened by acohen13 over 5 years ago - 2 comments

#46 - Misc tweaks

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#45 - Fix a naming error that prevents running the ImageReach-v0 task

Pull Request - State: closed - Opened by brandontrabucco over 5 years ago - 1 comment

#44 - Feature/video saving

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#43 - Fix output redirect arguments in debug mode

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#42 - Implement SimpleSampler._process_observations

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#41 - Feature/trajectory replay pool

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#40 - Flexible replay pool field attrs variable name

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#39 - Typo in refactoring flexible replay pool field logic

Issue - State: closed - Opened by mihaic over 5 years ago - 1 comment

#38 - Misc improvements

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#37 - Using log_alpha in alpha loss

Issue - State: closed - Opened by acohen13 over 5 years ago - 2 comments

#36 - Update SAC.tf_saveables to check _alpha_optimizer attribute

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#34 - Misc tweaks and fixes

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#33 - Conda Issues installing on macOS 10.14 Mojave,

Issue - State: closed - Opened by BruceK4t1qbit over 5 years ago - 2 comments

#32 - Conda issues installing patchelf

Issue - State: closed - Opened by BruceK4t1qbit over 5 years ago - 5 comments

#31 - Error occurs while runing

Issue - State: closed - Opened by Tzenthin over 5 years ago - 2 comments

#30 - Gaussian mixture policies

Issue - State: closed - Opened by seann999 over 5 years ago - 1 comment

#29 - Fix remote sampler get/set_state

Pull Request - State: closed - Opened by ryzhu over 5 years ago - 1 comment

#28 - Add reset_noise_scale argument for custom mujoco envs

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#27 - Additional information for baseline algorithms

Issue - State: closed - Opened by wsjeon over 5 years ago - 4 comments

#26 - GPU issues

Issue - State: closed - Opened by yangwenhaosms over 5 years ago - 4 comments

#25 - Feature/custom half cheetah env

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#24 - Remove repeated line

Pull Request - State: closed - Opened by haarnoja over 5 years ago

#23 - Add support for SQL

Pull Request - State: closed - Opened by henry-zhang-bohan over 5 years ago - 25 comments

#22 - Add compatibility with Mac

Pull Request - State: closed - Opened by henry-zhang-bohan over 5 years ago - 1 comment

#21 - Target q hard initialization

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#20 - Support for SQL

Issue - State: closed - Opened by henry-zhang-bohan over 5 years ago - 5 comments

#19 - Update ray to 0.6.1

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#18 - Q-targets should be initialized with hard update.

Issue - State: closed - Opened by hartikainen over 5 years ago - 1 comment

#17 - Automating alpha

Issue - State: closed - Opened by lgvaz over 5 years ago - 2 comments

#16 - why not support pip ?

Issue - State: closed - Opened by starimpact over 5 years ago - 2 comments

#15 - Pusher low level policy for ('any',-1) is not learning

Issue - State: closed - Opened by ahq1993 over 5 years ago - 2 comments

#14 - Update ray configs to allow automatic branch switching

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#13 - Fix/action smoothing

Pull Request - State: closed - Opened by hartikainen over 5 years ago

#12 - Remove conda from pip requirements

Pull Request - State: closed - Opened by hartikainen almost 6 years ago

#11 - Error on Docker/GPU installation

Issue - State: closed - Opened by wsjeon almost 6 years ago - 1 comment

#9 - How to use gym environment ?

Issue - State: closed - Opened by starimpact almost 6 years ago - 1 comment

#8 - Parallelization

Issue - State: open - Opened by kapsl almost 6 years ago - 3 comments

#7 - Add missing simulate_policy.py

Pull Request - State: closed - Opened by hartikainen almost 6 years ago

#6 - No module named examples.development.simulate_policy

Issue - State: closed - Opened by ycps almost 6 years ago - 1 comment

#5 - Update paper links

Pull Request - State: closed - Opened by hartikainen almost 6 years ago

#4 - self._Serializable__initialize(locals()) missing. serializable package missing

Issue - State: closed - Opened by garymcintire almost 6 years ago - 5 comments

#3 - [question] Action Smoothing

Issue - State: closed - Opened by araffin almost 6 years ago - 3 comments

#2 - Code slower than before refactoring

Issue - State: open - Opened by hartikainen almost 6 years ago

#1 - Checkpointing should not store cumulative replay pool

Issue - State: closed - Opened by hartikainen almost 6 years ago