Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Update README with Public Dataset V3 #29

Open
wants to merge 53 commits into
base: rllib
Choose a base branch
from
Open

Update README with Public Dataset V3 #29

wants to merge 53 commits into from

Conversation

lzardy
Copy link

@lzardy lzardy commented Aug 26, 2022

Credit: altf4 for public dataset v3.

vladfi1 added 30 commits May 4, 2022 20:00
@vladfi1
slippi db
c15dc6a
@vladfi1
Rename regime to env.
e0c307a
@vladfi1
Move secrets into json file.
53bc9c5
@vladfi1
Decompress with ray.
cd2784f
@vladfi1
Upload dbox files to s3.
b724cd1
@vladfi1
In-memory dbox upload.
3c73239
@vladfi1
Upload drive files to s3.
78750fd
@vladfi1
Fix drive_to_s3.
272a33c
@vladfi1
Fix flask app.
2d91c8f
@vladfi1
Decompress raw uploads with ray.
42fe063
@vladfi1
Preprocessing of slippi files.
b691e07
@vladfi1
Add invulnerable field.
c99d30f
@vladfi1
Move train filter into preprocessing.
5f0cb58
@vladfi1
Update peppi version in dockerfile.
5481a01
@vladfi1
Update cluster yamls.
6b6049e
@vladfi1
Generate parquet files for machine learning.
499732b
@vladfi1
Add lots of notebooks with cleaned outputs.
5115c2c
@vladfi1
Create datasets as tar archives with pd.DataFrame metadata.
e7018d3
@vladfi1
Document new dataset format.
dc589f4
@vladfi1
Train using new data format.
99f0bee
@vladfi1
WIP: refactor so that eval works
6022519
@vladfi1
Get training working again.
a1a94ac
@vladfi1
Python 3.8 compatibility.
ccb1e9d
@vladfi1
Minor bug in eval.
58fab46
@vladfi1
Fix parameter restoring.
4c6b96e
@vladfi1
Don't overlap trajectories, it breaks initial state.
defc3ac
@vladfi1
Headless (exi+ffw) option for Dolphin.
4bb2f5b
@vladfi1
Update run_dolphin script.
8a02964
@vladfi1
Use metadata to filter replays.
db4213c
@vladfi1
Minimal rllib training.
de5d198
vladfi1 and others added 23 commits June 5, 2022 11:34
@vladfi1
wip
3e91952
@vladfi1
[minor] type signature
d724248
@vladfi1
Checkpointing and restoring.
60acad0
@vladfi1
Rllib eval script.
9822580
@vladfi1
notebooks
232369f
@vladfi1
Be robust to slippi EnetDisconnect.
0159c52
@vladfi1
Be robust to slippi EnetDisconnect.
45798cb
@vladfi1
Validate player ports.
7ee0edf
@vladfi1
Profile multiple dolphins on a single cpu core.
f04a39a
@vladfi1
Fix deps.
67bcfea
@vladfi1
More options for rllib/tune.
48210da
@vladfi1
Update inference profiling script.
c8d0b75
@vladfi1
Profile on multiple cpu cores.
d445daf
@vladfi1
Non-ray dolphin profiling.
036b8b4
@vladfi1
Option to use ray for profiling dolphin_one_cpu.
5b64cbc
@vladfi1
Re-enable single-core profiling.
3df493f
@vladfi1
Wandb logging.
d007522
@vladfi1
Run rllib's ppo algorithm with wandb on aws.
4be13b9
@vladfi1
Support a2c, impala and ppo.
1436990
@vladfi1
WIP: use custom model from slippi_ai.
4da15ee
@lzardy
Update README with Public Dataset V3
cfb803b 
Credit: altf4 for public dataset v3.
@lzardy
Update README.md
5d646c2
@lzardy
Update README.md
21a97ee
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@lzardy @vladfi1