ditmeanflow_stack_multi_v1 / train_config.json
danielsanjosepro's picture
Upload policy weights, train config and readme
fb79655 verified
{
"dataset": {
"repo_id": "danielsanjosepro/stack_multi_v1",
"root": null,
"episodes": null,
"image_transforms": {
"enable": false,
"max_num_transforms": 3,
"random_order": false,
"tfs": {
"brightness": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"brightness": [
0.8,
1.2
]
}
},
"contrast": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"contrast": [
0.8,
1.2
]
}
},
"saturation": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"saturation": [
0.5,
1.5
]
}
},
"hue": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"hue": [
-0.05,
0.05
]
}
},
"sharpness": {
"weight": 1.0,
"type": "SharpnessJitter",
"kwargs": {
"sharpness": [
0.5,
1.5
]
}
}
}
},
"revision": null,
"use_imagenet_stats": true,
"video_backend": "torchcodec"
},
"env": null,
"policy": {
"type": "dit_meanflow",
"n_obs_steps": 2,
"normalization_mapping": {
"VISUAL": "MEAN_STD",
"STATE": "MIN_MAX",
"ACTION": "MEAN_STD"
},
"input_features": {
"observation.images.primary": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.images.wrist": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.state.cartesian": {
"type": "STATE",
"shape": [
6
]
},
"observation.state.gripper": {
"type": "STATE",
"shape": [
1
]
},
"observation.state.joints": {
"type": "STATE",
"shape": [
7
]
},
"observation.state.sensors_ft_sensor": {
"type": "STATE",
"shape": [
6
]
},
"observation.state.target": {
"type": "STATE",
"shape": [
6
]
},
"observation.state": {
"type": "STATE",
"shape": [
26
]
}
},
"output_features": {
"action": {
"type": "ACTION",
"shape": [
7
]
}
},
"device": "cuda",
"use_amp": false,
"push_to_hub": true,
"repo_id": "danielsanjosepro/ditmeanflow_stack_multi_v1",
"private": null,
"tags": null,
"license": null,
"horizon": 32,
"n_action_steps": 16,
"use_proprioceptive": true,
"drop_n_last_frames": 7,
"vision_backbone": "resnet18",
"crop_shape": null,
"crop_is_random": false,
"pretrained_backbone_weights": null,
"use_group_norm": true,
"spatial_softmax_num_keypoints": 32,
"use_separate_rgb_encoder_per_camera": true,
"frequency_embedding_dim": 256,
"hidden_dim": 512,
"num_blocks": 6,
"num_heads": 16,
"dropout": 0.1,
"dim_feedforward": 4096,
"activation": "gelu",
"training_noise_sampling": "uniform",
"clip_sample": true,
"clip_sample_range": 1.0,
"num_inference_steps": 100,
"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,
"use_meanflow": true,
"do_multi_step_sampling": false,
"inference_timesteps": 100,
"time_distribution": "uniform",
"log_norm_mu": -0.4,
"log_norm_sigma": 1.0,
"use_autograd_functional_jvp": false,
"use_adaptative_loss": true,
"flow_ratio": 0.5,
"cfg_prob": 0.1,
"cfg_omega": 2.0,
"cfg_w": 2.0,
"cfg_kappa": 0.0,
"optimizer_grad_clip_norm": 0.5
},
"output_dir": "outputs/train/2025-10-18/17-29-03_ditflow_mean_stack_multi_v1",
"job_name": "ditflow_mean_stack_multi_v1",
"resume": false,
"seed": 1000,
"num_workers": 4,
"batch_size": 64,
"steps": 50000,
"eval_freq": 1000,
"log_freq": 100,
"save_checkpoint": true,
"save_freq": 5000,
"use_policy_training_preset": true,
"optimizer": {
"type": "adam",
"lr": 0.0001,
"weight_decay": 1e-06,
"grad_clip_norm": 10.0,
"betas": [
0.95,
0.999
],
"eps": 1e-08
},
"scheduler": {
"type": "diffuser",
"num_warmup_steps": 500,
"name": "cosine"
},
"eval": {
"n_episodes": 50,
"batch_size": 50,
"use_async_envs": false,
"max_episodes_rendered": 4
},
"wandb": {
"enable": true,
"disable_artifact": false,
"project": "master_thesis",
"entity": null,
"notes": null,
"run_id": "wwdvjntf",
"mode": null
},
"eval_with_dataset": "danielsanjosepro/stack_multi_v1"
}