Robotics
LeRobot
Safetensors
diffusion
oct_diffusion_model / config.json
SPeramanu's picture
Upload policy weights, train config and readme
994af0b verified
{
"type": "diffusion",
"n_obs_steps": 2,
"input_features": {
"observation.pointcloud": {
"type": "STATE",
"shape": [
2048,
3
]
},
"observation.state": {
"type": "STATE",
"shape": [
18
]
}
},
"output_features": {
"action": {
"type": "ACTION",
"shape": [
11
]
}
},
"device": "cuda",
"use_amp": false,
"use_peft": false,
"push_to_hub": true,
"repo_id": "SPeramanu/oct_diffusion_model",
"private": null,
"tags": null,
"license": null,
"pretrained_path": "/mnt/job/output/checkpoints/last/pretrained_model",
"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,
"use_vit_encoder": false,
"vit_embed_dim": 512,
"vit_depth": 6,
"vit_num_heads": 8,
"vit_dropout": 0.1,
"use_point_cloud_encoder": false,
"pc_embed_dim": 256,
"pc_lift_dim": 256,
"pc_depth": 6,
"pc_num_heads": 8,
"pc_dropout": 0.1,
"use_voxelnext_encoder": true,
"voxelnext_voxel_size": 0.01,
"voxelnext_point_cloud_range": [
-1.0,
-1.0,
-1.0,
1.0,
1.0,
1.0
],
"voxelnext_embed_dim": 256,
"down_dims": [
256,
512,
1024
],
"kernel_size": 5,
"n_groups": 8,
"diffusion_step_embed_dim": 128,
"use_film_scale_modulation": true,
"noise_scheduler_type": "DDPM",
"num_train_timesteps": 100,
"beta_schedule": "squaredcos_cap_v2",
"beta_start": 0.0001,
"beta_end": 0.02,
"prediction_type": "epsilon",
"clip_sample": true,
"clip_sample_range": 1.0,
"num_inference_steps": 8,
"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
}