| { | |
| "type": "x2v_diffusion", | |
| "n_obs_steps": 2, | |
| "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, | |
| "push_to_hub": true, | |
| "repo_id": "Russo0405/test", | |
| "private": null, | |
| "tags": null, | |
| "license": null, | |
| "pretrained_path": null, | |
| "horizon": 16, | |
| "n_action_steps": 8, | |
| "normalization_mapping": { | |
| "VISUAL": "MEAN_STD", | |
| "STATE": "MIN_MAX", | |
| "ACTION": "MIN_MAX" | |
| }, | |
| "drop_n_last_frames": 7, | |
| "vision_backbone": "resnet18", | |
| "crop_shape": [ | |
| 84, | |
| 84 | |
| ], | |
| "crop_is_random": true, | |
| "pretrained_backbone_weights": null, | |
| "use_group_norm": true, | |
| "spatial_softmax_num_keypoints": 32, | |
| "use_separate_rgb_encoder_per_camera": false, | |
| "down_dims": [ | |
| 256, | |
| 512, | |
| 1024 | |
| ], | |
| "kernel_size": 5, | |
| "n_groups": 8, | |
| "diffusion_step_embed_dim": 128, | |
| "use_film_scale_modulation": true, | |
| "training_noise_sampling": "uniform", | |
| "num_inference_steps": null, | |
| "clip_sample": false, | |
| "clip_sample_range": 1.0, | |
| "frequency_embedding_dim": 256, | |
| "learnable_time_freq": false, | |
| "do_mask_loss_for_padding": false, | |
| "optimizer_lr": 0.0001, | |
| "optimizer_betas": [ | |
| 0.95, | |
| 0.999 | |
| ], | |
| "optimizer_eps": 1e-08, | |
| "optimizer_weight_decay": 1e-06, | |
| "scheduler_name": "cosine", | |
| "scheduler_warmup_steps": 500 | |
| } |