Instructions to use AlterraLaniakea/jepa-guided-diffusion with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Cosmos
How to use AlterraLaniakea/jepa-guided-diffusion with Cosmos:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Clean repo: config.yaml + best_model/ only
Browse files- model.pt +0 -3
- model_config.yaml +0 -52
- query_tokenizer/tokenizer.json +0 -3
- query_tokenizer/tokenizer_config.json +0 -14
model.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:f338a9e6fa9e40a9ab9a010b691c1be064e29d5ff9f1efac08f0997bf88d1831
|
| 3 |
-
size 2107382015
|
|
|
|
|
|
|
|
|
|
|
|
model_config.yaml
DELETED
|
@@ -1,52 +0,0 @@
|
|
| 1 |
-
stride: 3
|
| 2 |
-
num_frames: 16
|
| 3 |
-
vision_config:
|
| 4 |
-
name: vjepa2_1_vit_large_384
|
| 5 |
-
load_from: facebookresearch/vjepa2
|
| 6 |
-
source: github
|
| 7 |
-
video_mode: true
|
| 8 |
-
use_sdpa: true
|
| 9 |
-
use_rope: true
|
| 10 |
-
use_activation_checkpointing: true
|
| 11 |
-
dtype: bfloat16
|
| 12 |
-
patch_size: 16
|
| 13 |
-
tubelet_size: 2
|
| 14 |
-
crop_size: 384
|
| 15 |
-
unload: true
|
| 16 |
-
freeze: true
|
| 17 |
-
predictor:
|
| 18 |
-
language:
|
| 19 |
-
model: meta-llama/Llama-3.2-1B
|
| 20 |
-
num_layers: 4
|
| 21 |
-
max_query_length: 512
|
| 22 |
-
unload: false
|
| 23 |
-
freeze: false
|
| 24 |
-
compressor:
|
| 25 |
-
type: qformer
|
| 26 |
-
num_query_tokens: 512
|
| 27 |
-
num_layers: 4
|
| 28 |
-
num_heads: 8
|
| 29 |
-
mlp_ratio: 4.0
|
| 30 |
-
dropout: 0.2
|
| 31 |
-
cross_attention_frequency: 1
|
| 32 |
-
lr_multiplier: 2.0
|
| 33 |
-
target:
|
| 34 |
-
cosmos_dir: null
|
| 35 |
-
reason1_ckpt: nvidia/Cosmos-Reason1-7B
|
| 36 |
-
crossattn_proj_path: data/wts_cache/crossattn_proj.pt
|
| 37 |
-
embedding_concat_strategy: full_concat
|
| 38 |
-
n_layers_per_group: 5
|
| 39 |
-
lr_multiplier: 0.05
|
| 40 |
-
unload: true
|
| 41 |
-
freeze: true
|
| 42 |
-
decoder:
|
| 43 |
-
model: null
|
| 44 |
-
unload: true
|
| 45 |
-
freeze: false
|
| 46 |
-
runtime:
|
| 47 |
-
trust_remote_code: true
|
| 48 |
-
hf_token: true
|
| 49 |
-
torch_dtype: auto
|
| 50 |
-
tiny: false
|
| 51 |
-
embedding_dim: 1024
|
| 52 |
-
temperature: 0.1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
query_tokenizer/tokenizer.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
|
| 3 |
-
size 17209920
|
|
|
|
|
|
|
|
|
|
|
|
query_tokenizer/tokenizer_config.json
DELETED
|
@@ -1,14 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"backend": "tokenizers",
|
| 3 |
-
"bos_token": "<|begin_of_text|>",
|
| 4 |
-
"clean_up_tokenization_spaces": true,
|
| 5 |
-
"eos_token": "<|end_of_text|>",
|
| 6 |
-
"is_local": true,
|
| 7 |
-
"model_input_names": [
|
| 8 |
-
"input_ids",
|
| 9 |
-
"attention_mask"
|
| 10 |
-
],
|
| 11 |
-
"model_max_length": 131072,
|
| 12 |
-
"pad_token": "<|end_of_text|>",
|
| 13 |
-
"tokenizer_class": "TokenizersBackend"
|
| 14 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|