Skip to content

Actions: Ipsedo/ArenAI

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
222 workflow runs
222 workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
more step for cosine annealing ArenAI CI #218: Commit 92c24ea pushed by Ipsedo
3m 43s develop
develop
3m 43s
fix to device ArenAI CI #217: Commit 6841c00 pushed by Ipsedo
5m 20s develop
develop
5m 20s
* remove target entropy based on current mu (PPO) ArenAI CI #216: Commit 531d1c2 pushed by Ipsedo
4m 5s develop
develop
4m 5s
Merge pull request #33 from Ipsedo/develop ArenAI CI #215: Commit 2e67b66 pushed by Ipsedo
14m 46s master
master
14m 46s
Fix training ArenAI CI #214: Pull request #33 synchronize by Ipsedo
14m 55s develop
develop
14m 55s
fix CI glm::mat4 instead of {1.f} ArenAI CI #213: Commit 5f645da pushed by Ipsedo
12m 16s develop
develop
12m 16s
Fix training ArenAI CI #212: Pull request #33 opened by Ipsedo
10m 11s develop
develop
10m 11s
add trained model that works ArenAI CI #211: Commit 5ed7139 pushed by Ipsedo
8m 1s develop
develop
8m 1s
* sample or not for torch agent ArenAI CI #210: Commit 6631566 pushed by Ipsedo
2m 33s develop
develop
2m 33s
fix pid lagrangian ArenAI CI #209: Commit eaec694 pushed by Ipsedo
7m 38s develop
develop
7m 38s
PID Lagrangian for entropy adjustment in PPO : fix discrete ArenAI CI #208: Commit 93c271a pushed by Ipsedo
8m 19s develop
develop
8m 19s
* train ppo actor entropy & action only if kl not exceeded ArenAI CI #207: Commit f39b3b2 pushed by Ipsedo
6m 33s develop
develop
6m 33s
add trained model ArenAI CI #206: Commit 8a6e078 pushed by Ipsedo
2m 32s develop
develop
2m 32s
initial sigma for SAC ArenAI CI #205: Commit 8fdb9ce pushed by Ipsedo
8m 12s develop
develop
8m 12s
remove comment ArenAI CI #204: Commit 12f3fe0 pushed by Ipsedo
2m 50s develop
develop
2m 50s
re-add alpha parameters for PPO => per continuous action alpha ArenAI CI #203: Commit 5f41fc1 pushed by Ipsedo
1m 31s develop
develop
1m 31s
* fix set train to false (LayerNorm & GroupNorm) ArenAI CI #202: Commit c747112 pushed by Ipsedo
8m 30s develop
develop
8m 30s
Big fix : damages on shell contact ArenAI CI #201: Commit 1764289 pushed by Ipsedo
2m 32s develop
develop
2m 32s
lower continuous entropy coef and lower miss distance scale ArenAI CI #200: Commit 39da94f pushed by Ipsedo
6m 58s develop
develop
6m 58s
fix windows build ArenAI CI #199: Commit 060c899 pushed by Ipsedo
6m 25s develop
develop
6m 25s
fix kl in ppo ArenAI CI #198: Commit 816ea67 pushed by Ipsedo
7m 49s develop
develop
7m 49s
initialization style ArenAI CI #197: Commit 3cb600c pushed by Ipsedo
7m 55s develop
develop
7m 55s
refactor ambiguity between params (initial/target sigma & fire proba)… ArenAI CI #196: Commit 1d71c95 pushed by Ipsedo
6m 33s develop
develop
6m 33s
PPO : continue training on other minibatch when kl exceed ArenAI CI #195: Commit 38acf00 pushed by Ipsedo
6m 49s develop
develop
6m 49s
no reward norm for SAC ArenAI CI #194: Commit 8ec924e pushed by Ipsedo
2m 49s develop
develop
2m 49s