nalyda/picking_garbage01

1{2    "type": "act",3    "n_obs_steps": 1,4    "input_features": {5        "observation.state": {6            "type": "STATE",7            "shape": [8                69            ]10        },11        "observation.images.top": {12            "type": "VISUAL",13            "shape": [14                3,15                480,16                64017            ]18        },19        "observation.images.side": {20            "type": "VISUAL",21            "shape": [22                3,23                480,24                64025            ]26        }27    },28    "output_features": {29        "action": {30            "type": "ACTION",31            "shape": [32                633            ]34        }35    },36    "device": "cuda",37    "use_amp": false,38    "push_to_hub": true,39    "repo_id": "nalyda/picking_garbage01",40    "private": null,41    "tags": null,42    "license": null,43    "pretrained_path": "outputs/train/picking_01/checkpoints/last/pretrained_model",44    "chunk_size": 100,45    "n_action_steps": 100,46    "normalization_mapping": {47        "VISUAL": "MEAN_STD",48        "STATE": "MEAN_STD",49        "ACTION": "MEAN_STD"50    },51    "vision_backbone": "resnet18",52    "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",53    "replace_final_stride_with_dilation": 0,54    "pre_norm": false,55    "dim_model": 512,56    "n_heads": 8,57    "dim_feedforward": 3200,58    "feedforward_activation": "relu",59    "n_encoder_layers": 4,60    "n_decoder_layers": 1,61    "use_vae": true,62    "latent_dim": 32,63    "n_vae_encoder_layers": 4,64    "temporal_ensemble_coeff": null,65    "dropout": 0.1,66    "kl_weight": 10.0,67    "optimizer_lr": 1e-05,68    "optimizer_weight_decay": 0.0001,69    "optimizer_lr_backbone": 1e-0570}