Ruo-Ping Dong
|
7196d44b
|
remove GroupMaxStep
|
3 年前 |
Ervin Teng
|
da6a55a0
|
Revert "Use attention tests from master"
This reverts commit 78e052be8f36381bb6857817ff0f505716be83b9.
|
3 年前 |
|
Ervin Teng
|
c7054d76
|
Use attention tests from master
|
3 年前 |
|
Ruo-Ping Dong
|
d1107648
|
fix tests
|
3 年前 |
|
Ruo-Ping Dong
|
471a2e82
|
fix tests
|
3 年前 |
|
Ruo-Ping Dong
|
f3899bd2
|
more team -> group
|
3 年前 |
|
Ruo-Ping Dong
|
7271556f
|
change name TeamManager to MultiAgentGroup
|
3 年前 |
|
Ruo-Ping Dong
|
6dfd640e
|
remove OnTeamEpisodeBegin
|
3 年前 |
|
Ruo-Ping Dong
|
3e560168
|
unregister on disabled
|
3 年前 |
|
Ruo-Ping Dong
|
8ba38267
|
typo
|
3 年前 |
|
Ruo-Ping Dong
|
33f34772
|
add teamreward to decision step
|
3 年前 |
|
Ruo-Ping Dong
|
55265294
|
fix unregister agents
|
3 年前 |
|
Ruo-Ping Dong
|
bdc857f7
|
put team reward in decision steps
|
3 年前 |
|
Ruo-Ping Dong
|
63cf2d12
|
use delegate to avoid agent-manager cyclic reference
|
3 年前 |
|
Ruo-Ping Dong
|
83344b9c
|
address comments
|
3 年前 |
|
Ruo-Ping Dong
|
0b60877a
|
change method name to GetRegisteredAgents
|
3 年前 |
GitHub
|
790f2d32
|
fix setTeamReward
Co-authored-by: Vincent-Pierre BERGES <vincentpierre@unity3d.com>
|
3 年前 |
|
Ruo-Ping Dong
|
c22ed805
|
use 0 as default manager id
|
3 年前 |
|
Ruo-Ping Dong
|
a94acec6
|
put team reward in decision steps
|
3 年前 |
|
Ruo-Ping Dong
|
0dc83860
|
move manager
|
3 年前 |
|
Ruo-Ping Dong
|
0ed78a36
|
remove manager from academy when dispose
|
3 年前 |
|
Ruo-Ping Dong
|
f5ee561f
|
check agent by agent.enabled
|
3 年前 |
Chris Elion
|
dbf1c946
|
WIP
|
3 年前 |
|
Ruo-Ping Dong
|
b99c6f8b
|
add maxstep to teammanager and hook to academy
|
3 年前 |
|
Ruo-Ping Dong
|
38621840
|
set team reward
|
3 年前 |
|
Ruo-Ping Dong
|
f547f201
|
add team reward field to agent and proto
|
3 年前 |
|
Chris Elion
|
9d70220e
|
Merge remote-tracking branch 'origin/master' into MLA-1734-demo-provider
|
3 年前 |
|
Chris Elion
|
4c257df7
|
changelog
|
3 年前 |
|
Ruo-Ping Dong
|
aad7d342
|
add base team manager
|
3 年前 |
|
Chris Elion
|
4388aa9c
|
barracuda@1.3.0
|
3 年前 |
vincentpierre
|
186dafa8
|
Adding a new line to Readme.md so the next line is not part of the table
|
3 年前 |
|
Chris Elion
|
d6b9bf25
|
WIP
|
3 年前 |
|
GitHub
|
ddc3192c
|
Make ActionSpec constructor public (#4896)
|
3 年前 |
|
GitHub
|
0e2eaf8a
|
run lockbot more often, close after 30 days (#4894)
|
3 年前 |
|
GitHub
|
9c2eb951
|
fix broken download link (#4895)
|
3 年前 |
|
GitHub
|
d8835857
|
[MLA-1540] Training Analytics (#4780)
|
3 年前 |
|
GitHub
|
b6ce74a9
|
Add verified package info to table (#4889)
|
3 年前 |
|
GitHub
|
457ed0b8
|
Set torch device from commandline (#4888)
|
3 年前 |
|
GitHub
|
c556335e
|
Add enum for sensor implementations, send in analytics (#4871)
|
3 年前 |
|
GitHub
|
bc92b27f
|
remove unnecessary allocs in SideChannelMananger (#4886)
|
3 年前 |
|
GitHub
|
e83b35dd
|
[MLA-1724] Reduce use of IEnumerable during inference (#4887)
* improve allocations in inference
* Add IList overload for VectorSensor.AddObservation
* [skip ci] changelog
* [skip ci] migrating
|
3 年前 |
|
GitHub
|
cbb36a43
|
replace old bot with gh action (#4885)
|
3 年前 |
|
GitHub
|
2c09525e
|
Lock issues after 30 days. (#4884)
|
3 年前 |
|
GitHub
|
2af86534
|
[MLA-1712] Make UnityEnvironment fail fast if the env crashes (#4880)
|
3 年前 |
|
GitHub
|
d7f549f9
|
Run pytest on GPU (#4865)
* make tests device-friendly
* mark all tests in test_simple_rl
|
3 年前 |
|
GitHub
|
db4436e9
|
Merge pull request #4872 from Unity-Technologies/fix-numti-env-delayed-spawn
[Bug Fix] Fix crash if spawn is delayed in multi-env
|
3 年前 |
|
GitHub
|
bd4bc66b
|
Merge branch 'master' into fix-numti-env-delayed-spawn
|
3 年前 |
|
vincentpierre
|
52e4069f
|
fixing formatting
|
3 年前 |
|
vincentpierre
|
ff826bd2
|
added a test
|
3 年前 |
|
GitHub
|
9bb905da
|
use Debug.AssertFormat to avoid string formatting in DEBUG (#4879)
|
3 年前 |
