45154f52
Pytorch port of SAC ( #4219 )
5 年前
fd98cddd
reformating experiment_torch.py
5 年前
05a11c96
Develop add fire exp framework ( #4213 )
* Experiment branch for comparing torch
* Updates and merging ervin changes
* improvements on experiment_torch.py
* Better printing of results
* preliminary gpu experiment
* Testing gpu
* Prepare to see a lot of commits, because I like my IDE and I am testing on a server and I am using git to sync the two
* Prepare to see a lot of commits, because I like my IDE and I am testing on a server and I am using git to sync the two
* _
* _
* _
* _
* _
* _
* _
* _
* Attempt at gpu on tf. Does not work
* _
* _
* _
* _
* _
* _
* _
* _
* _
* _
* _
* Fixing learn.py
5 年前
cde8bd29
Convert List[np.ndarray] to np.ndarray before using torch.as_tensor ( #4183 )
Big speedup in visual obs
5 年前
0476c599
Remove print statement
5 年前
68169434
Fix discrete actions and GridWorld
5 年前
0d80d87a
Fix for discrete actions ( #4181 )
5 年前
5d33aca7
Remove double setting
5 年前
b6dfb4ac
Fix ResNet
5 年前
6408fd4e
Fix bug in pdf function
5 年前
e14eb72b
Fix some issues with pdf
5 年前
46874cc7
ONNX exporting
5 年前
9724c9ac
Merge master
5 年前
28e095e0
Merge remote-tracking branch 'origin/master' into develop-add-fire
5 年前
039f545a
Small performance improvement during inference
5 年前
c02e75d6
Time action sample function
5 年前
3eef9d78
Optimize np -> tensor operations
5 年前
b7be7f04
Fix bug in probs calculation
5 年前
2b3a6347
Merge remote-tracking branch 'origin/master' into develop-add-fire
5 年前
be7e55e1
Use LSTM and fix a few merge errors
5 年前
89ad3020
Merge remote-tracking branch 'origin/master' into develop-add-fire
# Conflicts:
# ml-agents/mlagents/trainers/policy/tf_policy.py
5 年前
9835d26c
Prepare model for onnx export
5 年前
ca887743
Support tf and pytorch alongside one another
5 年前
1736559f
Combine actor and critic classes. Initial export.
5 年前
596cc103
Remove unused arg
5 年前
29223931
Fix for memories
5 年前
82688e5c
GRU in-progress and dynamic cnns
5 年前
212e2d1d
Merge remote-tracking branch 'origin/master' into develop-add-fire
5 年前
5f936990
Visual observations now train as well
5 年前
a5b5b109
Mulkti-discrete now working
5 年前
a11a79e4
Continuous and discrete now train
5 年前
4a50444f
Support discrete actions as well
5 年前
e6c9cbc6
Merge remote-tracking branch 'origin/develop-add-fire-debug' into develop-add-fire
5 年前
21a8de45
Fix normalization
5 年前
61d671d8
Add conditional sigma for distribution
5 年前
8c6f4696
Fix a couple additional bugs
5 年前
3c82bf59
Training runs, but doesn’t actually work
5 年前
947f0d32
Slightly closer to running model
5 年前
b997f214
Share more code between tf and torch policies
5 年前
2e51260a
Resolving a few bugs
5 年前
7c3bd376
Refactoring policy and optimizer
5 年前
6879bae4
Initial optimizer port
5 年前
e166d018
Dynamically construct actor and critic
5 年前
3769d943
Merge remote-tracking branch 'origin/master' into develop-add-fire
5 年前
c577ce26
Merge remote-tracking branch 'origin/master' into develop-add-fire
5 年前
dc50162d
Add ResNet and distributions
5 年前
3327c215
Begin porting work
5 年前
8b5587cc
Remove obsolete methods from Agent class ( #3770 )
* Removed the obsolete methods from the Agent class
* Documentation changes
* [skip ci] Update com.unity.ml-agents/CHANGELOG.md
Co-Authored-By: Chris Elion <chris.elion@unity3d.com>
* [skip ci] Update docs/Migrating.md
Co-Authored-By: Chris Elion <chris.elion@unity3d.com>
Co-authored-by: Chris Elion <chris.elion@unity3d.com>
5 年前
dd6aa7e2
Agent.Heuristic takes an float[] ( #3765 )
5 年前
1843279a
[barracuda] Bump version to 0.6.3. ( #3766 )
5 年前
GitHub
vincentpierre
Ervin Teng
Arthur Juliani