hardmaru/estool issues and pull requests

#40 - Class OpenES, function ask. self.mu never gets updated.

Issue - State: closed - Opened by bharathbabuyp 12 months ago - 1 comment

#39 - Bump certifi from 2020.4.5.2 to 2022.12.7

Pull Request - State: open - Opened by dependabot[bot] almost 2 years ago
Labels: dependencies

#38 - Bump numpy from 1.18.5 to 1.22.0

Pull Request - State: open - Opened by dependabot[bot] over 2 years ago
Labels: dependencies

#37 - The meaning of ”rms_stdev“ function?

Issue - State: open - Opened by mantle2048 over 2 years ago

#36 - Bump numpy from 1.18.5 to 1.21.0

Pull Request - State: closed - Opened by dependabot[bot] over 2 years ago - 1 comment
Labels: dependencies

#35 - Fixed sign of `weight_decay` in `CMAES` solver

Pull Request - State: closed - Opened by bhartl over 2 years ago

#34 - What is the best way to step through the code ?

Issue - State: open - Opened by pzdkn about 3 years ago

#33 - Bump urllib3 from 1.25.9 to 1.26.5

Pull Request - State: closed - Opened by dependabot[bot] over 3 years ago
Labels: dependencies

#32 - matplotlib isn't in requirements.txt

Issue - State: open - Opened by danbri over 3 years ago - 1 comment

#31 - Bad score when training "KukaBulletEnv-v0"

Issue - State: open - Opened by HaoyuanXu about 4 years ago

#30 - Question for CMA-ES test part

Issue - State: open - Opened by aryanna384 about 4 years ago

#29 - upgrade pybullet to 2.0.0, 1.6.3 was not working

Pull Request - State: closed - Opened by maraoz over 4 years ago

#28 - Add requirements.txt

Pull Request - State: closed - Opened by maraoz over 4 years ago - 1 comment

#27 - PEPG and NES

Issue - State: closed - Opened by Wook133 over 4 years ago - 2 comments

#26 - Natural gradients for deep layers

Issue - State: open - Opened by parvkpr over 4 years ago - 1 comment

#25 - In case you still update this repo: added CartPoleContinuousBulletEnv-v0 (bullet_cartpole)

Pull Request - State: closed - Opened by erwincoumans almost 5 years ago - 1 comment

#24 - in PEPG module change_mu = np.dot(rT, epsilon) where left side have dimention pop_size and right have dimention batch_size(half of pop_size)

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

#23 - class OpenES, function ask: np.array mu come from update and compute step Adam optimizer with shape(popsize, numparams) and it was error when try to reshape mu.reshape(1, self.num_params)

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

#22 - newest gym is at 0.10.x

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

#21 - raise NotImplementedError

Issue - State: closed - Opened by keithmgould almost 6 years ago - 4 comments

#20 - The pre-trained Kuka model did not work

Issue - State: closed - Opened by lynnezixuan almost 6 years ago - 2 comments

#19 - Bug in train.py

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

#18 - No colour

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

#17 - Can't run pre-trained biped models

Issue - State: closed - Opened by ntduong almost 6 years ago - 2 comments

#16 - Fix methods name in custom_envs to be able to use the latest OpenAI Gym

Pull Request - State: closed - Opened by mmilk1231 about 6 years ago - 2 comments

#15 - [Weight decay] Should weight decay of model parameters instead ?

Issue - State: closed - Opened by zuoxingdong over 6 years ago - 3 comments

#14 - [Update es.py] No concatenation if we forget the best or first iteration

Pull Request - State: closed - Opened by zuoxingdong over 6 years ago - 1 comment

#13 - fix reward_list (average instead of total of all run)

Pull Request - State: closed - Opened by trougnouf over 6 years ago - 3 comments

#12 - Do the parameters need to be in [a,b]

Issue - State: closed - Opened by slremy over 6 years ago - 2 comments

#11 - bipedhard doesn't converge

Issue - State: closed - Opened by radsimu over 6 years ago - 14 comments

#10 - initial attempt at a setup.py for easy pip installs

Pull Request - State: closed - Opened by slremy over 6 years ago - 3 comments

#9 - How to train "KukaBulletEnv-v0" ?

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

#8 - bug in train.py

Issue - State: closed - Opened by ruoqizzz over 6 years ago - 16 comments

#7 - License

Issue - State: closed - Opened by frthjf over 6 years ago - 1 comment

#6 - PEPG questions

Issue - State: closed - Opened by cammckenzie over 6 years ago - 4 comments

#5 - negative function_values in CMAES?

Issue - State: closed - Opened by borgwang over 6 years ago - 2 comments

#4 - Solution range?

Issue - State: closed - Opened by MaxwellRebo almost 7 years ago - 2 comments

#3 - Training time for BipedalWalkerHardcore-v2

Issue - State: closed - Opened by dgriff777 almost 7 years ago - 11 comments

#2 - Run on AWS

Issue - State: closed - Opened by ErnstTmp almost 7 years ago - 4 comments

#1 - CMA-ES with O(N)

Issue - State: closed - Opened by ErnstTmp almost 7 years ago - 2 comments

GitHub / hardmaru/estool issues and pull requests