Image-to-Video
Cosmos
world-model
autonomous-driving
video-prediction
thermal
lidar
adverse-weather
lora
Instructions to use SORA-Lab/TriCosmos-2.5-predict with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Cosmos
How to use SORA-Lab/TriCosmos-2.5-predict 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
Download config/experiment.json from SORA-Lab/TriCosmos-2.5-predict: direct link, hf CLI and curl.
- Browser
- Download file 1.98 kB
-
https://huggingface.co/SORA-Lab/TriCosmos-2.5-predict/resolve/main/config/experiment.json
- Command line
-
hf download hf://SORA-Lab/TriCosmos-2.5-predict/config/experiment.json
-
curl -L -o experiment.json https://huggingface.co/SORA-Lab/TriCosmos-2.5-predict/resolve/main/config/experiment.json
1.98 kB
| { | |
| "experiment": "dsert_lora", | |
| "job_name": "tri_lora", | |
| "model_opts": [ | |
| "model.config.use_lora=true", | |
| "model.config.lora_rank=16", | |
| "model.config.lora_alpha=32", | |
| "model.config.lora_target_modules='q_proj,k_proj,v_proj,output_proj,mlp.layer1,mlp.layer2'", | |
| "model.config.init_lora_weights=true", | |
| "model.config.min_num_conditional_frames=2", | |
| "model.config.max_num_conditional_frames=2", | |
| "model.config.conditional_frames_probs=null", | |
| "model.config.conditional_frame_timestep=0.1" | |
| ], | |
| "train_opts": [ | |
| "job.project='dsert_wm'", | |
| "job.group='lora'", | |
| "job.name='tri_lora'", | |
| "dataloader_train.dataset.plan_file='<repo_root>/dataset/dsert_work/runs/tri_lora/plan_train.json'", | |
| "dataloader_train.sampler.plan_file='<repo_root>/dataset/dsert_work/runs/tri_lora/plan_train.json'", | |
| "dataloader_train.sampler.seed=0", | |
| "dataloader_train.batch_size=1", | |
| "dataloader_val.dataset.plan_file='<repo_root>/dataset/dsert_work/runs/tri_lora/plan_val.json'", | |
| "trainer.max_iter=500", | |
| "trainer.grad_accum_iter=1", | |
| "trainer.seed=0", | |
| "trainer.run_validation=true", | |
| "trainer.validation_iter=100", | |
| "trainer.max_val_iter=null", | |
| "trainer.callbacks.dsert_csv.out_file='<repo_root>/dataset/dsert_work/runs/tri_lora/train_log.csv'", | |
| "checkpoint.save_iter=100", | |
| "optimizer.lr=0.0001", | |
| "optimizer.weight_decay=0.001", | |
| "scheduler.warm_up_steps=[50]", | |
| "scheduler.cycle_lengths=[500]", | |
| "scheduler.f_max=[1.0]", | |
| "scheduler.f_min=[0.1]", | |
| "model.config.text_encoder_config.compute_online=false", | |
| "dataloader_train.dataset.text_cache_dir='<repo_root>/dataset/dsert_work/runs/tri_lora/text_embeddings'", | |
| "dataloader_val.dataset.text_cache_dir='<repo_root>/dataset/dsert_work/runs/tri_lora/text_embeddings'" | |
| ], | |
| "checkpoint_uuid": "81edfebe-bd6a-4039-8c1d-737df1a790bf", | |
| "pinned_commit": "a2c298b0a3df3778b973fe65e9e58877b292d8a7", | |
| "imaginaire_root": "<repo_root>/dataset/dsert_work/runs/tri_lora/imaginaire", | |
| "train_config": "configs/train_tri_lora.yaml" | |
| } |