{ "type": "flow_matching", "n_obs_steps": 2, "input_features": { "observation.images.image": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.images.image2": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.state": { "type": "STATE", "shape": [ 8 ] } }, "output_features": { "action": { "type": "ACTION", "shape": [ 7 ] } }, "device": "cuda", "use_amp": false, "use_peft": false, "push_to_hub": true, "repo_id": "fm-refactor-test", "private": null, "tags": null, "license": null, "pretrained_path": null, "horizon": 16, "n_action_steps": 8, "normalization_mapping": { "VISUAL": "MEAN_STD", "STATE": "MEAN_STD", "ACTION": "MEAN_STD" }, "drop_n_last_frames": 7, "vision_backbone": "resnet18", "vision_encoder": "spatial_softmax", "resize_shape": null, "crop_ratio": 0.8, "crop_shape": [ 204, 204 ], "crop_is_random": true, "pretrained_backbone_weights": null, "use_group_norm": true, "spatial_softmax_num_keypoints": 32, "use_separate_rgb_encoder_per_camera": false, "backbone": "transformer", "down_dims": [ 256, 512, 1024 ], "kernel_size": 5, "n_groups": 8, "diffusion_step_embed_dim": 256, "use_film_scale_modulation": true, "transformer_d_model": 256, "transformer_n_heads": 4, "transformer_n_layers": 4, "transformer_dropout": 0.1, "sigma": 0.0, "num_inference_steps": 10, "ode_solver": "euler", "delta_actions": false, "use_optimal_transport": false, "clip_sample": false, "clip_sample_range": 1.0, "compile_model": false, "compile_mode": "reduce-overhead", "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 }