Instructions to use mpreda/pusht_vqbet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use mpreda/pusht_vqbet with LeRobot:
- Notebooks
- Google Colab
- Kaggle
| { | |
| "dataset": { | |
| "repo_id": "lerobot/pusht", | |
| "root": null, | |
| "episodes": null, | |
| "image_transforms": { | |
| "enable": false, | |
| "max_num_transforms": 3, | |
| "random_order": false, | |
| "tfs": { | |
| "brightness": { | |
| "weight": 1.0, | |
| "type": "ColorJitter", | |
| "kwargs": { | |
| "brightness": [ | |
| 0.8, | |
| 1.2 | |
| ] | |
| } | |
| }, | |
| "contrast": { | |
| "weight": 1.0, | |
| "type": "ColorJitter", | |
| "kwargs": { | |
| "contrast": [ | |
| 0.8, | |
| 1.2 | |
| ] | |
| } | |
| }, | |
| "saturation": { | |
| "weight": 1.0, | |
| "type": "ColorJitter", | |
| "kwargs": { | |
| "saturation": [ | |
| 0.5, | |
| 1.5 | |
| ] | |
| } | |
| }, | |
| "hue": { | |
| "weight": 1.0, | |
| "type": "ColorJitter", | |
| "kwargs": { | |
| "hue": [ | |
| -0.05, | |
| 0.05 | |
| ] | |
| } | |
| }, | |
| "sharpness": { | |
| "weight": 1.0, | |
| "type": "SharpnessJitter", | |
| "kwargs": { | |
| "sharpness": [ | |
| 0.5, | |
| 1.5 | |
| ] | |
| } | |
| }, | |
| "affine": { | |
| "weight": 1.0, | |
| "type": "RandomAffine", | |
| "kwargs": { | |
| "degrees": [ | |
| -5.0, | |
| 5.0 | |
| ], | |
| "translate": [ | |
| 0.05, | |
| 0.05 | |
| ] | |
| } | |
| } | |
| } | |
| }, | |
| "revision": null, | |
| "use_imagenet_stats": true, | |
| "video_backend": "torchcodec", | |
| "streaming": false | |
| }, | |
| "env": { | |
| "type": "pusht", | |
| "task": "PushT-v0", | |
| "fps": 10, | |
| "features": { | |
| "action": { | |
| "type": "ACTION", | |
| "shape": [ | |
| 2 | |
| ] | |
| }, | |
| "agent_pos": { | |
| "type": "STATE", | |
| "shape": [ | |
| 2 | |
| ] | |
| }, | |
| "pixels": { | |
| "type": "VISUAL", | |
| "shape": [ | |
| 384, | |
| 384, | |
| 3 | |
| ] | |
| } | |
| }, | |
| "features_map": { | |
| "action": "action", | |
| "agent_pos": "observation.state", | |
| "environment_state": "observation.environment_state", | |
| "pixels": "observation.image" | |
| }, | |
| "max_parallel_tasks": 1, | |
| "disable_env_checker": true, | |
| "episode_length": 300, | |
| "obs_type": "pixels_agent_pos", | |
| "render_mode": "rgb_array", | |
| "visualization_width": 384, | |
| "visualization_height": 384, | |
| "observation_height": 384, | |
| "observation_width": 384 | |
| }, | |
| "policy": { | |
| "type": "vqbet", | |
| "n_obs_steps": 5, | |
| "input_features": { | |
| "observation.image": { | |
| "type": "VISUAL", | |
| "shape": [ | |
| 3, | |
| 96, | |
| 96 | |
| ] | |
| }, | |
| "observation.state": { | |
| "type": "STATE", | |
| "shape": [ | |
| 2 | |
| ] | |
| } | |
| }, | |
| "output_features": { | |
| "action": { | |
| "type": "ACTION", | |
| "shape": [ | |
| 2 | |
| ] | |
| } | |
| }, | |
| "device": "cuda", | |
| "use_amp": false, | |
| "use_peft": false, | |
| "push_to_hub": true, | |
| "repo_id": "mpreda/pusht_vqbet", | |
| "private": false, | |
| "tags": null, | |
| "license": null, | |
| "pretrained_path": null, | |
| "n_action_pred_token": 3, | |
| "action_chunk_size": 5, | |
| "normalization_mapping": { | |
| "VISUAL": "IDENTITY", | |
| "STATE": "MIN_MAX", | |
| "ACTION": "MIN_MAX" | |
| }, | |
| "vision_backbone": "resnet18", | |
| "crop_shape": [ | |
| 84, | |
| 84 | |
| ], | |
| "crop_is_random": true, | |
| "pretrained_backbone_weights": null, | |
| "use_group_norm": true, | |
| "spatial_softmax_num_keypoints": 32, | |
| "n_vqvae_training_steps": 20000, | |
| "vqvae_n_embed": 16, | |
| "vqvae_embedding_dim": 256, | |
| "vqvae_enc_hidden_dim": 128, | |
| "gpt_block_size": 500, | |
| "gpt_input_dim": 512, | |
| "gpt_output_dim": 512, | |
| "gpt_n_layer": 8, | |
| "gpt_n_head": 8, | |
| "gpt_hidden_dim": 512, | |
| "dropout": 0.1, | |
| "offset_loss_weight": 10000.0, | |
| "primary_code_loss_weight": 5.0, | |
| "secondary_code_loss_weight": 0.5, | |
| "bet_softmax_temperature": 0.1, | |
| "sequentially_select": false, | |
| "optimizer_lr": 0.0001, | |
| "optimizer_betas": [ | |
| 0.95, | |
| 0.999 | |
| ], | |
| "optimizer_eps": 1e-08, | |
| "optimizer_weight_decay": 1e-06, | |
| "optimizer_vqvae_lr": 0.001, | |
| "optimizer_vqvae_weight_decay": 0.0001, | |
| "scheduler_warmup_steps": 500 | |
| }, | |
| "output_dir": "/content/drive/MyDrive/Colab Notebooks/Matteo_notebook/vqbet_pusht", | |
| "job_name": "pusht_vqbet", | |
| "resume": false, | |
| "seed": 1000, | |
| "cudnn_deterministic": false, | |
| "num_workers": 0, | |
| "batch_size": 8, | |
| "steps": 20000, | |
| "eval_freq": 4000, | |
| "log_freq": 200, | |
| "tolerance_s": 0.0001, | |
| "save_checkpoint": true, | |
| "save_freq": 4000, | |
| "use_policy_training_preset": true, | |
| "optimizer": { | |
| "type": "adam", | |
| "lr": 0.0001, | |
| "weight_decay": 1e-06, | |
| "grad_clip_norm": 10.0, | |
| "betas": [ | |
| 0.95, | |
| 0.999 | |
| ], | |
| "eps": 1e-08 | |
| }, | |
| "scheduler": { | |
| "type": "vqbet", | |
| "num_warmup_steps": 500, | |
| "num_vqvae_training_steps": 20000, | |
| "num_cycles": 0.5 | |
| }, | |
| "eval": { | |
| "n_episodes": 1, | |
| "batch_size": 1, | |
| "use_async_envs": false | |
| }, | |
| "wandb": { | |
| "enable": false, | |
| "disable_artifact": false, | |
| "project": "lerobot", | |
| "entity": null, | |
| "notes": null, | |
| "run_id": null, | |
| "mode": null, | |
| "add_tags": true | |
| }, | |
| "peft": null, | |
| "use_rabc": false, | |
| "rabc_progress_path": null, | |
| "rabc_kappa": 0.01, | |
| "rabc_epsilon": 1e-06, | |
| "rabc_head_mode": "sparse", | |
| "rename_map": {}, | |
| "checkpoint_path": null | |
| } |