transformer-diffusion / config.json
ISdept's picture
Update config.json
92fe3a2 verified
raw
history blame
1.9 kB
{
"type": "transformer_diffusion",
"n_obs_steps": 4,
"input_features": {
"observation.state": {
"type": "STATE",
"shape": [
7
]
},
"observation.images.gripper": {
"type": "VISUAL",
"shape": [
3,
400,
640
]
},
"observation.images.front": {
"type": "VISUAL",
"shape": [
3,
400,
640
]
},
"observation.images.right": {
"type": "VISUAL",
"shape": [
3,
400,
640
]
}
},
"output_features": {
"action": {
"type": "ACTION",
"shape": [
7
]
}
},
"device": "cuda",
"use_amp": false,
"use_peft": false,
"push_to_hub": true,
"repo_id": null,
"private": null,
"tags": null,
"license": null,
"pretrained_path": null,
"horizon": 16,
"n_action_steps": 8,
"normalization_mapping": {
"VISUAL": "IDENTITY",
"STATE": "MEAN_STD",
"ACTION": "MEAN_STD"
},
"vision_backbone": "vit_b_16",
"input_image_size": [
224,
224
],
"num_cameras": 3,
"vision_freeze_layers": 0,
"state_dim": 7,
"action_dim": 7,
"d_model": 512,
"nhead": 8,
"num_encoder_layers": 4,
"num_decoder_layers": 6,
"dim_feedforward": 512,
"diffusion_step_embed_dim": 256,
"kernel_size": 3,
"n_groups": 8,
"num_inference_steps": 50,
"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
}