CUMTHenry's picture
Add model card documenting layout and contents
fed1013 verified
|
Raw History Blame Contribute Delete
5.95 kB
---
license: mit
tags:
- reinforcement-learning
- offline-rl
- diffusion-models
- consistency-models
- ogbench
- compdiffuser
- pytorch
library_name: pytorch
---
# CompDiffuser checkpoints
Model parameters for the Native Chunk Consistency Models codebase,
[Henry0132/comp_flow_map](https://github.com/Henry0132/comp_flow_map).
This repository holds weights and their saved configurations only. Rollout results,
videos and evaluation output are deliberately excluded, so it stays a checkpoint
store rather than an experiment archive.
## Download
The repository root holds one directory per dataset plus `dql_results/`, while the
code expects dataset directories under `logs/`. The helper script in the code
repository restores both into the right places:
```bash
bash download_artifacts_from_hf.sh --repo CUMTHenry/compdiffuser_checkpoints
```
By hand, the equivalent is two calls:
```bash
hf download CUMTHenry/compdiffuser_checkpoints --include "dql_results/*" --local-dir .
hf download CUMTHenry/compdiffuser_checkpoints --exclude "dql_results/*" --local-dir logs
```
Set `HF_ENDPOINT=https://hf-mirror.com` for a faster download inside mainland China.
## Layout
```text
<dataset>/diffusion/<experiment>/ original CompDiffuser planner and inverse-dynamics weights
<dataset>/chunk_consistency/<experiment>/ chunk consistency student distilled from that teacher
dql_results/<cube-dataset>/ released Cube low-level controllers
```
## Contents
Steps are the saved checkpoint step in thousands. `β€”` means that role is not part
of this dataset.
| Dataset | Planner / denoiser | Inverse dynamics | Chunk consistency |
| --- | --- | --- | --- |
| `antmaze-giant-stitch-v0` | `og_antM_Gi_o2d_Cd_Stgl_PadBuf_Ft64_ts512_T512` @ 1600k | `og_antM_Gi_o29d_g2d_invdyn_h12` @ 800k | `..._ChunkConsistency_v1_restart_s200k` @ 200k, 400k, 600k |
| `antmaze-medium-stitch-v0` | `og_antM_Me_o2d_Cd_Stgl_PadBuf_Ft64_ts512_T512` @ 1600k | `og_antM_Me_o29d_g2d_invdyn_h12` @ 800k | β€” |
| `antsoccer-arena-stitch-v0` | `og_antSoc_Ar_o17d_DiTd768_PadBuf_Ft64_ts512_fs4_h160_ovlp56MditD384_T512` @ 1800k | `og_antSoc_Ar_o42d_g17d_invdyn_h120_dout02` @ 1800k | β€” |
| `antsoccer-medium-stitch-v0` | `og_antSoc_Me_o17d_DiTd768_PadBuf_Ft64_MaxOriPlan200_ts512_fs4_h160_ovlp56MditD384_T512` @ 1800k | `og_antSoc_Me_o42d_g17d_invdyn_h100_dm4_dout02` @ 1800k | β€” |
| `cube-single-play-v0` | `og_cubeSingle_o28d_DiTd1024dp12_PadBuf_Ft64_fs4_h160_ovlp56MditD512_H160_T512` @ 1600k | β€” | β€” |
| `cube-double-play-v0` | `og_cubeDouble_o37d_DiTd1024dp12_PadBuf_Ft64_fs4_h160_ovlp56MditD512_H160_T512` @ 1600k | β€” | β€” |
| `cube-triple-play-v0` | `og_cubeTriple_o46d_DiTd1024dp12_PadBuf_Ft64_fs4_h160_ovlp56MditD512_H160_T512` @ 1600k | β€” | β€” |
| `cube-quadruple-play-v0` | `og_cubeQuadruple_o55d_DiTd1024dp12_PadBuf_Ft64_fs4_h160_ovlp56MditD512_H160_T512` @ 1600k | β€” | β€” |
| `humanoidmaze-giant-stitch-v0` | `og_humM_Gi_o2d_Cd_Stgl_PadBuf_Ft64_ts1k_h336_ovlp128_ovdm5_ts512_bs192_bd128_td96_drop02_T1000` @ 1800k | `og_humM_Gi_o69d_g2d_invdyn_h80_dm5_dout02` @ 1600k | β€” |
| `humanoidmaze-medium-stitch-v0` | `og_humM_Me_o2d_Cd_Stgl_PadBuf_Ft64_ts1k_h336_ovlp128_ovdm5_ts512_bs192_bd128_td96_drop02_T1000` @ 1600k, 1800k | `og_humM_Me_o69d_g2d_invdyn_h80_dm5_dout02` @ 800k, 1600k | β€” |
| `pointmaze-giant-stitch-v0` | `og_pntM_Gi_o2d_Cd_Stgl_PadBuf_Ft64_ts512_T512` @ 1200k | β€” | β€” |
| `pointmaze-medium-stitch-v0` | `og_pntM_Me_o2d_Cd_Stgl_PadBuf_Ft64_ts512_T512` @ 1200k | β€” | β€” |
Cube low-level controllers, one directory per dataset under `dql_results/`:
| Dataset | Files |
| --- | --- |
| `cube-single-play-v0` | `actor_200.pth`, `critic_200.pth`, `critic_target_200.pth`, `variant.json` |
| `cube-double-play-v0` | same |
| `cube-triple-play-v0` | same |
| `cube-quadruple-play-v0` | same |
## File types
| File | Contents |
| --- | --- |
| `state_<step>.pt` | Checkpoint dict. Planner and inverse-dynamics weights hold `step`, `model` and `ema`. Chunk consistency weights additionally hold `optimizer`, `consistency`, `teacher`, `checkpoint_schema_version` and `artifact_type`. |
| `model_config.pkl` | Network class and constructor arguments. |
| `dfu_model.pkl` | Diffusion process class and arguments. |
| `dataset_config.pkl` | Dataset class, normalization and sequence settings. |
| `small_trainer_config.pkl`, `inv_trainer_config.pkl` | Trainer class and settings. |
| `render_config.pkl` | Renderer settings used for evaluation videos. |
| `model_config.txt` | Human-readable module summary. |
| `args.json` | Command-line arguments of the original run. |
| `train_metrics.jsonl` | Per-interval training metrics. |
The saved pickles are authoritative. Experiment directory names encode the task,
observation and goal dimensions, architecture and training settings, but the
`ts` tokens are not documented here; read `args.json` and `model_config.pkl` for
the exact values. A checkpoint filename step and its internal `step` field can
differ, because the original release named files by epoch.
## Notes
- Only `antmaze-giant-stitch-v0` has a chunk consistency student so far. Its run
was still training when these weights were uploaded, so later checkpoints may
appear in a future revision.
- `train_metrics.jsonl` is append-only and grows while training runs, so its size
can differ from the local copy.
- Planning additionally needs the datasets and the per-task assets expected by
each environment; those are not part of this repository.
## License and acknowledgements
Released under the [MIT License](https://github.com/Henry0132/comp_flow_map/blob/main/LICENSE).
Built on [RCD](https://refining-compositional-diffusion.github.io/),
[CompDiffuser](https://github.com/devinluo27/comp_diffuser_release),
[CDGS](https://github.com/UtkarshMishra04/CDGS_ogbench) and
[OGBench](https://github.com/seohongpark/ogbench). Preserve the upstream notices
and cite the corresponding works.