KunhangLI commited on
Commit
d6b709a
·
verified ·
1 Parent(s): b78fdff

Add FD2 + TR and FD2 + TR + DMD checkpoints

Browse files
README.md CHANGED
@@ -2,7 +2,7 @@
2
  license: other
3
  license_name: babel-amass-non-commercial
4
  license_link: https://babel.is.tue.mpg.de/license.html
5
- pretty_name: Triangular Resampling data release (BABEL 263D)
6
  tags:
7
  - motion-generation
8
  - text-to-motion
@@ -12,16 +12,17 @@ tags:
12
  - diffusion
13
  ---
14
 
15
- # Triangular Resampling: data release
16
 
17
- Data for **Triangular Resampling for Long-Horizon Motion Generation**
18
  ([arXiv:2609.34697](https://arxiv.org/abs/2609.34697)).
19
  Code: [github.com/KunhangL/TriangularResampling](https://github.com/KunhangL/TriangularResampling).
20
 
21
  This repository holds the BABEL data and the FloodDiffusion 2 (FD2) training metadata used
22
  in the paper. These are the exact files of the paper's runs. HumanML3D is not included; it
23
  comes with the [FloodDiffusion release](https://huggingface.co/ShandaAI/FloodDiffusionDownloads),
24
- whose `HumanML3D.zip` is identical to the HumanML3D copy used in the paper.
 
25
 
26
  | File | Extract into | Contents | SHA-256 |
27
  | --- | --- | --- | --- |
@@ -49,6 +50,15 @@ for name in ("BABEL.zip", "manifests.zip"):
49
  zipfile.ZipFile(path).extractall("raw_data")
50
  ```
51
 
 
 
 
 
 
 
 
 
 
52
  ## `BABEL.zip`
53
 
54
  ```
@@ -90,6 +100,31 @@ manifests/
90
  `problems` field lists two HumanML3D test features with non-finite values (`007975`,
91
  `M007975`), which FD2 validation skips.
92
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
93
  ## License
94
 
95
  These files are derived from [BABEL](https://babel.is.tue.mpg.de/) (Punnakkal et al.,
@@ -103,6 +138,9 @@ The data is for **non-commercial scientific research only**. By using it you agr
103
  [AMASS license](https://amass.is.tue.mpg.de/license.html). If you need the raw motion data or a commercial license, register
104
  with and contact the dataset owners.
105
 
 
 
 
106
  ## Citation
107
 
108
  ```bibtex
 
2
  license: other
3
  license_name: babel-amass-non-commercial
4
  license_link: https://babel.is.tue.mpg.de/license.html
5
+ pretty_name: Triangular Resampling data (BABEL 263D) and checkpoints
6
  tags:
7
  - motion-generation
8
  - text-to-motion
 
12
  - diffusion
13
  ---
14
 
15
+ # Triangular Resampling: data and checkpoints
16
 
17
+ Data and checkpoints for **Triangular Resampling for Long-Horizon Motion Generation**
18
  ([arXiv:2609.34697](https://arxiv.org/abs/2609.34697)).
19
  Code: [github.com/KunhangL/TriangularResampling](https://github.com/KunhangL/TriangularResampling).
20
 
21
  This repository holds the BABEL data and the FloodDiffusion 2 (FD2) training metadata used
22
  in the paper. These are the exact files of the paper's runs. HumanML3D is not included; it
23
  comes with the [FloodDiffusion release](https://huggingface.co/ShandaAI/FloodDiffusionDownloads),
24
+ whose `HumanML3D.zip` is identical to the HumanML3D copy used in the paper. The repository
25
+ also holds FD2 post-training checkpoints, described under [`checkpoints/`](#checkpoints).
26
 
27
  | File | Extract into | Contents | SHA-256 |
28
  | --- | --- | --- | --- |
 
50
  zipfile.ZipFile(path).extractall("raw_data")
51
  ```
52
 
53
+ The checkpoints, into `checkpoints/` of the code repository:
54
+
55
+ ```bash
56
+ hf download KunhangLI/TriangularResampling \
57
+ checkpoints/fd2_tr_step200.ckpt \
58
+ checkpoints/fd2_tr_dmd_step400.ckpt \
59
+ --local-dir .
60
+ ```
61
+
62
  ## `BABEL.zip`
63
 
64
  ```
 
100
  `problems` field lists two HumanML3D test features with non-finite values (`007975`,
101
  `M007975`), which FD2 validation skips.
102
 
103
+ ## `checkpoints/`
104
+
105
+ FD2 post-training checkpoints for HumanML3D + BABEL (263D features at 20 fps). Both start
106
+ from the official FD2 `humanml3d_babel_fk_200k` checkpoint
107
+ ([AlayaLab/FloodDiffusion2](https://huggingface.co/AlayaLab/FloodDiffusion2)).
108
+
109
+ | File | Model | Config in the code repository | SHA-256 |
110
+ | --- | --- | --- | --- |
111
+ | `checkpoints/fd2_tr_step200.ckpt` | FD2 + TR, 200 post-training steps | `methods/fd2/configs/humanml_babel_tr.yaml` | `4aa7fc36cea05b6e1ac72e78bcd225a75ae6835a172e294537414545e546f104` |
112
+ | `checkpoints/fd2_tr_dmd_step400.ckpt` | FD2 + TR + DMD, 400 post-training steps | `methods/fd2/configs/humanml_babel_tr_dmd.yaml` | `bd505473cd8377d6c86e27c98fc573d737bfed969254bfa1708ad185a7214d16` |
113
+
114
+ - **Format.** Inference exports of about 686 MB each, with three entries: `state_dict`
115
+ (bfloat16), `ema_state` (float32 EMA weights, the ones generation uses) and `global_step`.
116
+ They hold no optimizer state, so training cannot be resumed from them.
117
+ - **Use.** Pass the file and its config to `methods/fd2/training/generate.py`:
118
+
119
+ ```bash
120
+ python -m methods.fd2.training.generate --config methods/fd2/configs/humanml_babel_tr.yaml \
121
+ --checkpoint checkpoints/fd2_tr_step200.ckpt \
122
+ --manifest common/evaluation/manifests/humanml3d_prompts_256_uniform_seed1234.jsonl \
123
+ --output outputs/eval/fd2_tr_fixed/generated --method fd2_tr --seed 42
124
+ ```
125
+
126
+ The code repository's README lists the 120-second evaluation scores of both checkpoints.
127
+
128
  ## License
129
 
130
  These files are derived from [BABEL](https://babel.is.tue.mpg.de/) (Punnakkal et al.,
 
138
  [AMASS license](https://amass.is.tue.mpg.de/license.html). If you need the raw motion data or a commercial license, register
139
  with and contact the dataset owners.
140
 
141
+ The checkpoints are trained on HumanML3D and BABEL, starting from the official FD2 checkpoint.
142
+ They are released under the same terms: non-commercial scientific research only.
143
+
144
  ## Citation
145
 
146
  ```bibtex
SHA256SUMS CHANGED
@@ -1,2 +1,4 @@
1
  681fdd86ebc70d24bec18b1b83e04f0070a8ae8fc3cd3a33ee7f1cd7ba2dfdcf BABEL.zip
2
  71b2b95ff2c072500562aa8ebe2b42d03dbbd4102b2d07f7e9c83e566d88cc32 manifests.zip
 
 
 
1
  681fdd86ebc70d24bec18b1b83e04f0070a8ae8fc3cd3a33ee7f1cd7ba2dfdcf BABEL.zip
2
  71b2b95ff2c072500562aa8ebe2b42d03dbbd4102b2d07f7e9c83e566d88cc32 manifests.zip
3
+ 4aa7fc36cea05b6e1ac72e78bcd225a75ae6835a172e294537414545e546f104 checkpoints/fd2_tr_step200.ckpt
4
+ bd505473cd8377d6c86e27c98fc573d737bfed969254bfa1708ad185a7214d16 checkpoints/fd2_tr_dmd_step400.ckpt
checkpoints/fd2_tr_dmd_step400.ckpt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd505473cd8377d6c86e27c98fc573d737bfed969254bfa1708ad185a7214d16
3
+ size 685790155
checkpoints/fd2_tr_step200.ckpt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4aa7fc36cea05b6e1ac72e78bcd225a75ae6835a172e294537414545e546f104
3
+ size 685775867