Robotics
Transformers
Safetensors
English
3d-detection
vision-language-action
pose-estimation
grounding
Instructions to use hetolin/PoseVLA-stage1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use hetolin/PoseVLA-stage1 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("hetolin/PoseVLA-stage1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 1,185 Bytes
16d0df3 | 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 40 41 42 43 44 45 46 47 | {
"type": "pi0_ours",
"n_obs_steps": 1,
"normalization_mapping": {
"VISUAL": "IDENTITY",
"STATE": "IDENTITY",
"ACTION": "IDENTITY"
},
"input_features": {},
"output_features": {},
"chunk_size": 50,
"n_action_steps": 50,
"max_state_dim": 32,
"max_action_dim": 32,
"resize_imgs_with_padding": [
224,
224
],
"empty_cameras": 0,
"tokenizer_max_length": 512,
"tokenizer_model_path": "google/paligemma-3b-pt-224",
"proj_width": 1024,
"num_steps": 10,
"use_cache": true,
"attention_implementation": "eager",
"pi05": false,
"is_knowledge_insulation": false,
"add_extra_token": true,
"add_image_token": true,
"add_prior": true,
"freeze_vision_encoder": false,
"train_expert_only": false,
"train_state_proj": true,
"optimizer_lr": 5e-05,
"optimizer_betas": [
0.9,
0.95
],
"optimizer_eps": 1e-08,
"optimizer_weight_decay": 1e-10,
"scheduler_warmup_steps": 16000,
"scheduler_decay_steps": 1280000,
"scheduler_decay_lr": 5e-06,
"vis_attn": true,
"skip_init_weights": false,
"device": "cpu"
} |