Skip to content

Finetune: IQL #46

Merged
merged 60 commits into from
Jun 13, 2023
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
60 commits
Select commit Hold shift + click to select a range
4eda2f9
add BC configs
Mar 21, 2023
9d800e3
add bc-10 configs
Mar 21, 2023
436faed
fix a typo
Mar 21, 2023
3e82b53
fix bc-10 discount 0.99 -> 1.0
Mar 25, 2023
2527f0c
add td3+bc antmaze configs
Mar 25, 2023
b0e19e7
add awac antmaze-v2 configs + fix det. torch
Mar 25, 2023
99d2ede
Merge branch 'main' into run-antmaze-v2
Mar 25, 2023
a18fbf2
unify awac log naming
Mar 25, 2023
71f1af8
fix td3+bc checkpointing
Mar 25, 2023
60d3a8e
add iql antmaze-v2 configs
Mar 25, 2023
de27d2d
add dt antmaze-v2 configs
Mar 25, 2023
20e1d8b
add sac-n antmaze-v2 configs
Mar 29, 2023
6cf34a4
add edac antmaze-v2 configs
Mar 29, 2023
d00e58a
fix cql and add antmaze-v2 configs
Mar 30, 2023
435f6d9
add bc, bc-10, awac configs for diverse datasets
Apr 3, 2023
465af55
add sac-n, edac configs for diverse datasets
Apr 4, 2023
1c4726b
add diverse antmaze configs for iql
Apr 17, 2023
7231dc6
add antmaze-diverse configs for td3+bc
Apr 17, 2023
a26aa9d
add antmaze-diverse configs for dt
Apr 17, 2023
f4222fd
update iql for adroit
Apr 27, 2023
cb4c280
update adroit config for iql
Apr 27, 2023
d0eed11
add all configs for adroit and iql
Apr 27, 2023
bc8484c
add td3+bc adroit configs
Apr 30, 2023
97d4cc3
add bc adroit configs
Apr 30, 2023
b686a9e
add bc10 adroit configs
Apr 30, 2023
1e810e3
IQL with tunning
DT6A May 20, 2023
fc6baf1
Move offline training to separate folders
DT6A May 21, 2023
e40e55b
Merge branch 'offline-to-online' into offline-to-online-iql
DT6A May 21, 2023
e3e4c9e
Move IQL to finetune folder
DT6A May 21, 2023
78a6950
Fix linter
DT6A May 21, 2023
362e605
Fix linter
DT6A May 21, 2023
426407b
Add online episodes info logging, remove default checkpointing from c…
DT6A May 22, 2023
5b1c12d
Change logging
DT6A May 22, 2023
e44afaa
Fix linter
DT6A May 22, 2023
4b6137d
Update refs in README
DT6A May 22, 2023
5591544
Trying to add __init__.py to fix linter
DT6A May 23, 2023
1ff06e9
Moved __init__.py
DT6A May 23, 2023
e7c05c1
Merge branch 'offline-to-online' into offline-to-online-iql
DT6A May 23, 2023
808aced
Merge branch 'main' into run-antmaze-v2
DT6A May 26, 2023
e2f5b9f
Move configs
DT6A May 26, 2023
3c449db
Add adroit configs for AWAC, SAC-N, EDAC
DT6A May 26, 2023
3fea9c7
Set minimum number of trajectories to 1
DT6A May 26, 2023
1c331a0
Fix AWAC deterministic
DT6A May 26, 2023
856be07
Fix EDAC pen expert config
DT6A May 26, 2023
a261167
DT Adroit configs
DT6A May 27, 2023
07a0aaa
Add regret logging to IQL
DT6A May 27, 2023
6b107c4
Merge branch 'main' into offline-to-online-iql
DT6A May 27, 2023
dfcc511
Fix linter
DT6A May 27, 2023
f0c9c29
Clip regret scores
DT6A May 27, 2023
37991c0
Fix EDAC configs
DT6A May 29, 2023
89bffb6
Change regret logging
DT6A May 29, 2023
cb84d27
Change train success
DT6A May 29, 2023
d8a0c1f
Add typings
DT6A Jun 8, 2023
8748958
Merge branch 'main' into offline-to-online-iql
DT6A Jun 9, 2023
3813db6
Fix typings in IQL files and add lr params
DT6A Jun 9, 2023
8ffd2c6
Add lr params to configs
DT6A Jun 9, 2023
cd31ff7
Merge branch 'run-antmaze-v2' into offline-to-online-iql
DT6A Jun 9, 2023
b7cb6a9
Add lr params to offline configs
DT6A Jun 9, 2023
93de0d9
Typings fix
DT6A Jun 9, 2023
28b02bc
Merge branch 'main' into offline-to-online-iql
DT6A Jun 12, 2023
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Fix EDAC configs
  • Loading branch information
DT6A committed May 29, 2023
commit 37991c0bedc252106eebd492776bf830280839e0
4 changes: 2 additions & 2 deletions configs/offline/edac/door/expert_v1.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -6,13 +6,13 @@ checkpoints_path: null
critic_learning_rate: 0.0003
deterministic_torch: false
device: cuda
env_name: "door-cloned-v1"
env_name: "door-expert-v1"
eta: 200.0
eval_episodes: 10
eval_every: 5
eval_seed: 42
gamma: 0.99
group: "edac-door-cloned-v1-multiseed-v2"
group: "edac-door-expert-v1-multiseed-v2"
hidden_dim: 256
log_every: 100
max_action: 1.0
Expand Down
4 changes: 2 additions & 2 deletions configs/offline/edac/hammer/expert_v1.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -6,13 +6,13 @@ checkpoints_path: null
critic_learning_rate: 0.0003
deterministic_torch: false
device: cuda
env_name: "hammer-cloned-v1"
env_name: "hammer-expert-v1"
eta: 200.0
eval_episodes: 10
eval_every: 5
eval_seed: 42
gamma: 0.99
group: "edac-hammer-cloned-v1-multiseed-v2"
group: "edac-hammer-expert-v1-multiseed-v2"
hidden_dim: 256
log_every: 100
max_action: 1.0
Expand Down
4 changes: 2 additions & 2 deletions configs/offline/edac/relocate/expert_v1.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -6,13 +6,13 @@ checkpoints_path: null
critic_learning_rate: 0.0003
deterministic_torch: false
device: cuda
env_name: "relocate-cloned-v1"
env_name: "relocate-expert-v1"
eta: 200.0
eval_episodes: 10
eval_every: 5
eval_seed: 42
gamma: 0.99
group: "edac-relocate-cloned-v1-multiseed-v2"
group: "edac-relocate-expert-v1-multiseed-v2"
hidden_dim: 256
log_every: 100
max_action: 1.0
Expand Down