Instructions to use chomeed/insertion_prior_new with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use chomeed/insertion_prior_new with LeRobot:
- Notebooks
- Google Colab
- Kaggle
File size: 934 Bytes
6b3d022 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | {
"type": "insertion_prior",
"n_obs_steps": 1,
"input_features": {},
"output_features": {},
"device": "cuda",
"use_amp": false,
"use_peft": false,
"push_to_hub": true,
"repo_id": "chomeed/insertion_prior_new",
"private": null,
"tags": null,
"license": null,
"pretrained_path": null,
"d_model": 512,
"vis_emb_dim": 768,
"text_emb_dim": 384,
"n_layers": 4,
"n_heads": 8,
"dropout": 0.1,
"num_cameras": 2,
"max_seq_len": 200,
"optimizer_lr": 0.0001,
"optimizer_betas": [
0.9,
0.95
],
"optimizer_eps": 1e-08,
"optimizer_weight_decay": 1e-10,
"optimizer_grad_clip_norm": 10.0,
"scheduler_warmup_steps": 1000,
"scheduler_decay_steps": 10000,
"scheduler_decay_lr": 2.5e-06,
"scheduler_peak_lr": 5e-05,
"normalization_mapping": {
"VISUAL": "IDENTITY",
"LANGUAGE": "IDENTITY"
}
} |