Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
57 commits
Select commit Hold shift + click to select a range
88e4658
fix(kl_divergence): fix wrong passed param when using kl_penalty (#129)
anHappyDog Sep 25, 2025
92b44d7
docs(example-gallery): update example gallery doc pages (#137)
QuanluZhang Sep 27, 2025
59cc43e
fix: handle num_nodes configuration mismatch with actual node number …
andylin-hao Sep 28, 2025
7875bdd
feat: simplify the maniskill reset id (#135)
zanghz21 Sep 28, 2025
fca00ce
fix(embodied): bug for single node test (#141)
XuS1994 Sep 28, 2025
35f6a4b
fix: openvla/oft dependency issues (#143)
andylin-hao Sep 29, 2025
547c8ce
fix: default gloo group due to prismatic import & dep enhancement (#145)
andylin-hao Sep 30, 2025
cae65b8
feat(async_vllm): add async vllm worker and unified sync/async interf…
anHappyDog Sep 30, 2025
1d8f5ff
chore: fix README-EN and add README-CN (#152)
secretsites Oct 9, 2025
2e6850e
fix: remove env channel timeout (#156)
andylin-hao Oct 10, 2025
89e398e
fix: optimal embodied yaml and doc update (#144)
zanghz21 Oct 10, 2025
55a1173
chore: update README (#158)
secretsites Oct 10, 2025
d611e91
feat: unified docker image and dockerfile (#147)
andylin-hao Oct 11, 2025
a475bc4
fix(embodied): fix function of ordered reset id; make gradient ckpt i…
WinstonWmj Oct 11, 2025
e77a20d
feat: the open source version of cursor online rl for tab completion …
qurakchin Oct 11, 2025
5cdfe4f
fix: maniskill render, sim stage seed, rename step-level to action-le…
zanghz21 Oct 11, 2025
69324c6
fix(embodied): urgent fix for libero_10 yaml; (#161)
WinstonWmj Oct 11, 2025
af32ed0
fix: libero not fully install in image (#160)
andylin-hao Oct 11, 2025
5e73652
ci: use new CI machines and accelerate CI process (#162)
andylin-hao Oct 12, 2025
78ddc65
feat: unfied CI script and new embodied CI tests (#163)
andylin-hao Oct 12, 2025
461e4e6
Merge pull request #1 from anHappyDog/feature/weight_convertor
anHappyDog Sep 8, 2025
5db6fab
feat(rollout_mm): add multimodal input/output for rollout backend (#2)
anHappyDog Sep 12, 2025
90a1308
feat(vlm): support VLM sglang rollout and fsdp training (#6)
guozhen1997 Sep 18, 2025
7c382c8
feat(dataset): refactor and add lazy loader process
anHappyDog Sep 19, 2025
fa0fc75
fix(vllm): fix wrong image_data param when running vlm in vllm
anHappyDog Sep 19, 2025
6a7d4bc
feat: add vqa reward function, unify math and vqa reward
guozhen1997 Sep 22, 2025
7100e6b
feat: add reward worker
guozhen1997 Sep 22, 2025
f7c2fba
fix: fix vqa reward bugs and ruff format
guozhen1997 Sep 23, 2025
0c38831
feat: rename and reorganize example config
guozhen1997 Sep 23, 2025
e6ebd60
fix: fix ruff, fix merge bugs
guozhen1997 Sep 23, 2025
dc446fc
fix: fix multi modal inputs
guozhen1997 Sep 25, 2025
a04f2d0
fix(math): fix some bugs when running math model
anHappyDog Sep 29, 2025
a7df8fc
fix(math): fix some merge_batch when item is not tensor,add support f…
anHappyDog Sep 29, 2025
14cbdf0
chore: add corresponding changes to yaml because of RewardModel and o…
anHappyDog Sep 29, 2025
1245988
fix(megatron): apply corresponding changes due to fsdp
anHappyDog Sep 30, 2025
4bf2d81
fix(reward): change math_verify_call's result from {0,1} to {-1,1}
anHappyDog Sep 30, 2025
fc77e2a
feat(ci): change corresponding ci config for refactored code
anHappyDog Sep 30, 2025
9d40cb4
chore: refactor dataset parts
anHappyDog Oct 2, 2025
ecb1ed0
fix(mm_data): unify vllm/sglang's mm_data passing
anHappyDog Oct 2, 2025
582a438
fix(rollout): fix some problems in sglang/vllm, now both are ok
anHappyDog Oct 2, 2025
fa5b861
fix(ci): add ci for vqa
anHappyDog Oct 2, 2025
74535e6
fix(ci): fix some bugs in ci
anHappyDog Oct 3, 2025
62df313
fix(fsdp): add forgotten backward and optimizer step
anHappyDog Oct 5, 2025
1ec7e54
fix(collocated): fix inference/rollout do jobs parallelly which cause…
anHappyDog Oct 5, 2025
e57d10d
fix(sync_weight): fix oom bugs
anHappyDog Oct 8, 2025
d0edcd0
fix(vlm): in torch260's image, transformers version is 4.51.1 and it'…
anHappyDog Oct 9, 2025
19f2a27
fix(fsdp): use bf16 instead of fp16 for training
anHappyDog Oct 10, 2025
d67365c
feat(ci): add fsdp ci
anHappyDog Oct 10, 2025
01f95ff
feat(fsdp): fix ci, add fsdp optimizations like overlap and gradient …
anHappyDog Oct 10, 2025
a8023c8
fix(ci): add fsdp's run_inference, fix ci
anHappyDog Oct 11, 2025
803c4c6
fix(ci): fix some errors
anHappyDog Oct 12, 2025
c1a74b0
feat(ci): fix ci
anHappyDog Oct 13, 2025
2d51313
fix(reward): remove redundant reward definitions
anHappyDog Oct 13, 2025
fbc9be7
fix(lock): set fsdp's recompute_logprobs True for lock competition sa…
anHappyDog Oct 13, 2025
2969359
chore: remove useless code, add correct dp_group param for mg
anHappyDog Oct 14, 2025
9dae32e
fix(reward): move reward worker's timer to where reward computation r…
anHappyDog Oct 14, 2025
b334af1
Rlinf-Acend
Taoxu-unihannover Nov 7, 2025
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
2 changes: 1 addition & 1 deletion .github/CODEOWNERS
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@

/tests @andylin-hao
/toolkits/auto_placement @i-Taozi
/toolkits/ckpt_convertor @Louis-J
/toolkits/ckpt_convertor @qurakchin

/.pre-commit-config.yaml @andylin-hao
/pyproject.toml @andylin-hao
Expand Down
20 changes: 18 additions & 2 deletions .github/ISSUE_TEMPLATE/bug-report.yml
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,22 @@ body:
Please provide a clear and concise description of what the bug is.
validations:
required: true
- type: textarea
attributes:
label: Log file
description: |
Please provide upload the log file or paste the full log messages when the bug happens.
value: |
You can find the log file in `logs/` folder or the $output_dir/$experiment_name folder (defined in the yaml config) if you are using our example scripts.

Log file:

If you cannot find the log, please provide the full log messages here.
```
<paste log here>
```
validations:
required: true
- type: textarea
attributes:
label: Environment
Expand All @@ -28,13 +44,13 @@ body:
pip list | grep -E "torch|sglang|vllm|ray|transformers"
nvidia-smi
```
Additionally, please provide the RLinf version, Megatron version if you are using Megatron, and docker image version if you are using our images
Additionally, please provide the RLinf version, Megatron version if you are using Megatron, and docker image tag if you are using our images
value: |
Python version:
PIP list:
RLinf version:
Megatron version:
Docker image version:
Docker image tag:
nvidia-smi:
validations:
required: true
Expand Down
58 changes: 0 additions & 58 deletions .github/workflows/auto_placement.yml

This file was deleted.

Loading