Upload folder using huggingface_hub
Browse files- config.json +105 -0
- experiment_cfg/conf.yaml +310 -0
- experiment_cfg/config.yaml +320 -0
- experiment_cfg/dataset_statistics.json +316 -0
- experiment_cfg/final_model_config.json +153 -0
- experiment_cfg/final_processor_config.json +0 -0
- model-00001-of-00003.safetensors +3 -0
- model-00002-of-00003.safetensors +3 -0
- model-00003-of-00003.safetensors +3 -0
- model.safetensors.index.json +730 -0
- processor/embodiment_id.json +38 -0
- processor/processor_config.json +2980 -0
- processor/statistics.json +0 -0
- trainer_state.json +0 -0
- training_args.bin +3 -0
- wandb_config.json +1 -0
- zero_to_fp32.py +760 -0
config.json
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"action_horizon": 16,
|
| 3 |
+
"action_model_lora_alpha": 32,
|
| 4 |
+
"action_model_lora_dropout": 0.0,
|
| 5 |
+
"action_model_lora_rank": 16,
|
| 6 |
+
"action_model_lora_target_modules": [
|
| 7 |
+
"vl_qkv",
|
| 8 |
+
"vl_proj",
|
| 9 |
+
"sa_qkv",
|
| 10 |
+
"sa_proj",
|
| 11 |
+
"p_qkv",
|
| 12 |
+
"p_proj",
|
| 13 |
+
"linear1",
|
| 14 |
+
"linear2"
|
| 15 |
+
],
|
| 16 |
+
"action_model_use_lora": false,
|
| 17 |
+
"architectures": [
|
| 18 |
+
"RLDX"
|
| 19 |
+
],
|
| 20 |
+
"backbone_lora_alpha": 32,
|
| 21 |
+
"backbone_lora_dropout": 0.0,
|
| 22 |
+
"backbone_lora_num_layers": -1,
|
| 23 |
+
"backbone_lora_rank": 16,
|
| 24 |
+
"backbone_lora_target_modules": [
|
| 25 |
+
"q_proj",
|
| 26 |
+
"k_proj",
|
| 27 |
+
"v_proj",
|
| 28 |
+
"o_proj",
|
| 29 |
+
"gate_proj",
|
| 30 |
+
"up_proj",
|
| 31 |
+
"down_proj"
|
| 32 |
+
],
|
| 33 |
+
"backbone_use_lora": false,
|
| 34 |
+
"color_jitter_params": {
|
| 35 |
+
"brightness": 0.3,
|
| 36 |
+
"contrast": 0.4,
|
| 37 |
+
"hue": 0.08,
|
| 38 |
+
"saturation": 0.5
|
| 39 |
+
},
|
| 40 |
+
"conversation_image_first": false,
|
| 41 |
+
"diffusion_model_cfg": {
|
| 42 |
+
"action_model_max_seq_len": 512,
|
| 43 |
+
"attention_head_dim": 64,
|
| 44 |
+
"depth_multi_stream": 4,
|
| 45 |
+
"depth_single_stream": 8,
|
| 46 |
+
"dropout": 0.2,
|
| 47 |
+
"final_dropout": true,
|
| 48 |
+
"num_attention_heads": 24,
|
| 49 |
+
"output_dim": 1024,
|
| 50 |
+
"physics_dim": 0,
|
| 51 |
+
"positional_embeddings": "rope_sa_only",
|
| 52 |
+
"pre_norm": "layer_norm",
|
| 53 |
+
"qk_norm": "rms_norm",
|
| 54 |
+
"rope_theta": 10000.0,
|
| 55 |
+
"sa_dim": 1536,
|
| 56 |
+
"temb_type": "input_token",
|
| 57 |
+
"use_physics": false,
|
| 58 |
+
"use_swiglu": true,
|
| 59 |
+
"vl_dim": 4096
|
| 60 |
+
},
|
| 61 |
+
"dtype": "bfloat16",
|
| 62 |
+
"freeze_cog_tokens": false,
|
| 63 |
+
"general_embodiment_train_ratio": 0,
|
| 64 |
+
"image_max_area": 65536,
|
| 65 |
+
"image_resize_m": 32,
|
| 66 |
+
"memory_cfg": {
|
| 67 |
+
"hidden_size": 4096,
|
| 68 |
+
"intermediate_size": 16384,
|
| 69 |
+
"max_position_embeddings": 32,
|
| 70 |
+
"num_attention_heads": 16,
|
| 71 |
+
"num_hidden_layers": 2,
|
| 72 |
+
"num_key_value_heads": 16,
|
| 73 |
+
"rms_norm_eps": 1e-05,
|
| 74 |
+
"use_causal_attn": true,
|
| 75 |
+
"use_rope": true
|
| 76 |
+
},
|
| 77 |
+
"model_name": "/data/taeyoung/models/RLDX-1-VLM",
|
| 78 |
+
"model_revision": null,
|
| 79 |
+
"model_type": "RLDX-1",
|
| 80 |
+
"n_cog_tokens": 64,
|
| 81 |
+
"physics_delta_indices": null,
|
| 82 |
+
"physics_dims": [],
|
| 83 |
+
"physics_keys": [],
|
| 84 |
+
"physics_use_flow_matching": true,
|
| 85 |
+
"random_crop_fraction": null,
|
| 86 |
+
"random_rotation_angle": null,
|
| 87 |
+
"rtc_inference_delay": 0,
|
| 88 |
+
"rtc_inference_exec_horizon": 0,
|
| 89 |
+
"rtc_inference_mode": "none",
|
| 90 |
+
"rtc_jacobian_steps_only": 3,
|
| 91 |
+
"rtc_training_max_delay": 0,
|
| 92 |
+
"state_dropout_prob": 0.0,
|
| 93 |
+
"transformers_version": "4.57.0",
|
| 94 |
+
"tune_diffusion_model": true,
|
| 95 |
+
"tune_llm": false,
|
| 96 |
+
"tune_projector": true,
|
| 97 |
+
"tune_top_llm_layers": 4,
|
| 98 |
+
"tune_visual": false,
|
| 99 |
+
"use_memory": false,
|
| 100 |
+
"use_motion": false,
|
| 101 |
+
"use_physics": false,
|
| 102 |
+
"use_video": true,
|
| 103 |
+
"video_length": 1,
|
| 104 |
+
"video_stride": 2
|
| 105 |
+
}
|
experiment_cfg/conf.yaml
ADDED
|
@@ -0,0 +1,310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
data:
|
| 2 |
+
datasets:
|
| 3 |
+
- dataset_paths:
|
| 4 |
+
- /data/taeyoung/data/shared_dataset/robocasa_mg_gr00t_300
|
| 5 |
+
embodiment_tag: general_embodiment
|
| 6 |
+
mix_ratio: 1.0
|
| 7 |
+
dataset_type: physical_embodiment
|
| 8 |
+
val_dataset_path: null
|
| 9 |
+
modality_configs:
|
| 10 |
+
general_embodiment:
|
| 11 |
+
video:
|
| 12 |
+
delta_indices:
|
| 13 |
+
- 0
|
| 14 |
+
modality_keys:
|
| 15 |
+
- left_view
|
| 16 |
+
- right_view
|
| 17 |
+
- wrist_view
|
| 18 |
+
sin_cos_embedding_keys: null
|
| 19 |
+
mean_std_embedding_keys: null
|
| 20 |
+
action_configs: null
|
| 21 |
+
state:
|
| 22 |
+
delta_indices:
|
| 23 |
+
- 0
|
| 24 |
+
modality_keys:
|
| 25 |
+
- end_effector_position_relative
|
| 26 |
+
- end_effector_rotation_relative
|
| 27 |
+
- gripper_qpos
|
| 28 |
+
- base_position
|
| 29 |
+
- base_rotation
|
| 30 |
+
sin_cos_embedding_keys: null
|
| 31 |
+
mean_std_embedding_keys: null
|
| 32 |
+
action_configs: null
|
| 33 |
+
action:
|
| 34 |
+
delta_indices:
|
| 35 |
+
- 0
|
| 36 |
+
- 1
|
| 37 |
+
- 2
|
| 38 |
+
- 3
|
| 39 |
+
- 4
|
| 40 |
+
- 5
|
| 41 |
+
- 6
|
| 42 |
+
- 7
|
| 43 |
+
- 8
|
| 44 |
+
- 9
|
| 45 |
+
- 10
|
| 46 |
+
- 11
|
| 47 |
+
- 12
|
| 48 |
+
- 13
|
| 49 |
+
- 14
|
| 50 |
+
- 15
|
| 51 |
+
modality_keys:
|
| 52 |
+
- end_effector_position
|
| 53 |
+
- end_effector_rotation
|
| 54 |
+
- gripper_close
|
| 55 |
+
- base_motion
|
| 56 |
+
- control_mode
|
| 57 |
+
sin_cos_embedding_keys: null
|
| 58 |
+
mean_std_embedding_keys: null
|
| 59 |
+
action_configs:
|
| 60 |
+
- rep: DELTA
|
| 61 |
+
type: EEF
|
| 62 |
+
format: DEFAULT
|
| 63 |
+
state_key: null
|
| 64 |
+
- rep: DELTA
|
| 65 |
+
type: EEF
|
| 66 |
+
format: DEFAULT
|
| 67 |
+
state_key: null
|
| 68 |
+
- rep: ABSOLUTE
|
| 69 |
+
type: NON_EEF
|
| 70 |
+
format: DEFAULT
|
| 71 |
+
state_key: null
|
| 72 |
+
- rep: DELTA
|
| 73 |
+
type: NON_EEF
|
| 74 |
+
format: DEFAULT
|
| 75 |
+
state_key: null
|
| 76 |
+
- rep: ABSOLUTE
|
| 77 |
+
type: NON_EEF
|
| 78 |
+
format: DEFAULT
|
| 79 |
+
state_key: null
|
| 80 |
+
language:
|
| 81 |
+
delta_indices:
|
| 82 |
+
- 0
|
| 83 |
+
modality_keys:
|
| 84 |
+
- annotation.human.action.task_description
|
| 85 |
+
sin_cos_embedding_keys: null
|
| 86 |
+
mean_std_embedding_keys: null
|
| 87 |
+
action_configs: null
|
| 88 |
+
dataset_mode: sharded
|
| 89 |
+
download_cache: false
|
| 90 |
+
shard_size: 1024
|
| 91 |
+
episode_sampling_rate: 0.1
|
| 92 |
+
num_shards_per_epoch: 100000
|
| 93 |
+
override_pretraining_statistics: false
|
| 94 |
+
mode: single_turn
|
| 95 |
+
random_chop: 0.0
|
| 96 |
+
mock_dataset_mode: false
|
| 97 |
+
shuffle: true
|
| 98 |
+
seed: 42
|
| 99 |
+
multiprocessing_context: fork
|
| 100 |
+
allow_padding: true
|
| 101 |
+
subsample_ratio: 1.0
|
| 102 |
+
video_backend: torchcodec
|
| 103 |
+
load_config_path: null
|
| 104 |
+
model:
|
| 105 |
+
model_type: RLDX-1
|
| 106 |
+
model_dtype: bfloat16
|
| 107 |
+
model_name: /data/taeyoung/models/RLDX-1-VLM
|
| 108 |
+
backbone_model_type: vtc_qwen3_vl
|
| 109 |
+
model_revision: null
|
| 110 |
+
tune_top_llm_layers: 4
|
| 111 |
+
backbone_embedding_dim: 4096
|
| 112 |
+
tune_llm: false
|
| 113 |
+
tune_visual: false
|
| 114 |
+
select_layer: 18
|
| 115 |
+
reproject_vision: false
|
| 116 |
+
use_flash_attention: true
|
| 117 |
+
load_bf16: true
|
| 118 |
+
backbone_trainable_params_fp32: true
|
| 119 |
+
freeze_cog_tokens: false
|
| 120 |
+
image_max_area: 65536
|
| 121 |
+
image_resize_m: 32
|
| 122 |
+
random_crop_fraction: null
|
| 123 |
+
random_rotation_angle: null
|
| 124 |
+
color_jitter_params:
|
| 125 |
+
brightness: 0.3
|
| 126 |
+
contrast: 0.4
|
| 127 |
+
saturation: 0.5
|
| 128 |
+
hue: 0.08
|
| 129 |
+
formalize_language: true
|
| 130 |
+
apply_sincos_state_encoding: false
|
| 131 |
+
use_relative_action: true
|
| 132 |
+
use_percentiles: true
|
| 133 |
+
conversation_image_first: false
|
| 134 |
+
max_state_dim: 64
|
| 135 |
+
max_action_dim: 64
|
| 136 |
+
action_horizon: 16
|
| 137 |
+
hidden_size: 1024
|
| 138 |
+
input_embedding_dim: 1536
|
| 139 |
+
general_embodiment_train_ratio: 0.0
|
| 140 |
+
add_pos_embed: true
|
| 141 |
+
attn_dropout: 0.2
|
| 142 |
+
use_vlln: false
|
| 143 |
+
max_seq_len: 1024
|
| 144 |
+
n_cog_tokens: 64
|
| 145 |
+
diffusion_model_cfg:
|
| 146 |
+
action_model_max_seq_len: 512
|
| 147 |
+
attention_head_dim: 64
|
| 148 |
+
depth_multi_stream: 4
|
| 149 |
+
depth_single_stream: 8
|
| 150 |
+
dropout: 0.2
|
| 151 |
+
final_dropout: true
|
| 152 |
+
num_attention_heads: 24
|
| 153 |
+
output_dim: 1024
|
| 154 |
+
positional_embeddings: rope_sa_only
|
| 155 |
+
pre_norm: layer_norm
|
| 156 |
+
qk_norm: rms_norm
|
| 157 |
+
rope_theta: 10000.0
|
| 158 |
+
temb_type: input_token
|
| 159 |
+
use_swiglu: true
|
| 160 |
+
num_inference_timesteps: 4
|
| 161 |
+
noise_beta_alpha: 1.5
|
| 162 |
+
noise_beta_beta: 1.0
|
| 163 |
+
noise_s: 0.999
|
| 164 |
+
num_timestep_buckets: 1000
|
| 165 |
+
tune_projector: true
|
| 166 |
+
tune_diffusion_model: true
|
| 167 |
+
tune_vlln: true
|
| 168 |
+
action_model_use_lora: false
|
| 169 |
+
action_model_lora_rank: 16
|
| 170 |
+
action_model_lora_alpha: 32
|
| 171 |
+
action_model_lora_dropout: 0.0
|
| 172 |
+
action_model_lora_target_modules:
|
| 173 |
+
- vl_qkv
|
| 174 |
+
- vl_proj
|
| 175 |
+
- sa_qkv
|
| 176 |
+
- sa_proj
|
| 177 |
+
- p_qkv
|
| 178 |
+
- p_proj
|
| 179 |
+
- linear1
|
| 180 |
+
- linear2
|
| 181 |
+
backbone_use_lora: false
|
| 182 |
+
backbone_lora_rank: 16
|
| 183 |
+
backbone_lora_alpha: 32
|
| 184 |
+
backbone_lora_dropout: 0.0
|
| 185 |
+
backbone_lora_num_layers: -1
|
| 186 |
+
backbone_lora_target_modules:
|
| 187 |
+
- q_proj
|
| 188 |
+
- k_proj
|
| 189 |
+
- v_proj
|
| 190 |
+
- o_proj
|
| 191 |
+
- gate_proj
|
| 192 |
+
- up_proj
|
| 193 |
+
- down_proj
|
| 194 |
+
state_dropout_prob: 0.0
|
| 195 |
+
state_additive_noise_scale: 0.0
|
| 196 |
+
rtc_training_max_delay: 0
|
| 197 |
+
rtc_inference_mode: none
|
| 198 |
+
rtc_inference_delay: 0
|
| 199 |
+
rtc_inference_exec_horizon: 0
|
| 200 |
+
rtc_jacobian_steps_only: 3
|
| 201 |
+
use_memory: false
|
| 202 |
+
memory_length: 4
|
| 203 |
+
memory_n_cog_tokens: null
|
| 204 |
+
concat_memory: false
|
| 205 |
+
memory_dropout_prob: 0.0
|
| 206 |
+
memory_stride: 16
|
| 207 |
+
memory_cfg:
|
| 208 |
+
hidden_size: 4096
|
| 209 |
+
intermediate_size: 16384
|
| 210 |
+
max_position_embeddings: 32
|
| 211 |
+
num_attention_heads: 16
|
| 212 |
+
num_hidden_layers: 2
|
| 213 |
+
num_key_value_heads: 16
|
| 214 |
+
rms_norm_eps: 1.0e-05
|
| 215 |
+
use_causal_attn: true
|
| 216 |
+
use_rope: true
|
| 217 |
+
use_motion: false
|
| 218 |
+
motion_insert_layer: 9
|
| 219 |
+
motion_d_hid: 512
|
| 220 |
+
motion_window:
|
| 221 |
+
- 5
|
| 222 |
+
- 9
|
| 223 |
+
- 9
|
| 224 |
+
motion_ext_chnls:
|
| 225 |
+
- 256
|
| 226 |
+
motion_int_chnls:
|
| 227 |
+
- 256
|
| 228 |
+
- 256
|
| 229 |
+
- 512
|
| 230 |
+
motion_corr_func: cosine
|
| 231 |
+
motion_n_encoders: 1
|
| 232 |
+
motion_use_layerscale: false
|
| 233 |
+
motion_layerscale_init: 1.0e-05
|
| 234 |
+
motion_use_layernorm: false
|
| 235 |
+
motion_use_syncbn: false
|
| 236 |
+
motion_injection_point: vision_encoder
|
| 237 |
+
motion_pool_type: avg
|
| 238 |
+
motion_drop: true
|
| 239 |
+
motion_gradient_check: false
|
| 240 |
+
motion_int_mode: lite
|
| 241 |
+
use_video: true
|
| 242 |
+
video_length: 1
|
| 243 |
+
video_stride: 2
|
| 244 |
+
use_physics: false
|
| 245 |
+
physics_keys: []
|
| 246 |
+
physics_dims: []
|
| 247 |
+
physics_loss_weight: 0.1
|
| 248 |
+
allow_missing_physics: false
|
| 249 |
+
physics_delta_indices: null
|
| 250 |
+
physics_use_flow_matching: true
|
| 251 |
+
physics_dropout_prob: 0.0
|
| 252 |
+
max_num_embodiments: 36
|
| 253 |
+
training:
|
| 254 |
+
output_dir: /data/taeyoung/workspace/RLDX-1/ckpt/rldx1/finetuned/rldx1_image_ft_robocasa300_bsz1024_60k
|
| 255 |
+
experiment_name: rldx1_image_ft_robocasa300_bsz1024_60k
|
| 256 |
+
max_steps: 60000
|
| 257 |
+
global_batch_size: 1024
|
| 258 |
+
batch_size: null
|
| 259 |
+
gradient_accumulation_steps: 1
|
| 260 |
+
learning_rate: 0.0001
|
| 261 |
+
lr_scheduler_type: cosine
|
| 262 |
+
weight_decay: 1.0e-05
|
| 263 |
+
warmup_ratio: 0.05
|
| 264 |
+
warmup_steps: 0
|
| 265 |
+
max_grad_norm: 1.0
|
| 266 |
+
new_param_warmup_steps: 0
|
| 267 |
+
optim: adamw_torch_fused
|
| 268 |
+
start_from_checkpoint: /data/taeyoung/workspace/RLDX-1/ckpt/rldx1/pretrained/rldx1_image_pt_mixv5_bsz8192_100k/rldx1_image_pt_mixv5_bsz8192_100k/checkpoint-100000
|
| 269 |
+
tf32: true
|
| 270 |
+
fp16: false
|
| 271 |
+
bf16: true
|
| 272 |
+
eval_bf16: true
|
| 273 |
+
logging_steps: 10
|
| 274 |
+
save_steps: 1000
|
| 275 |
+
save_total_limit: 10
|
| 276 |
+
save_vl_model: false
|
| 277 |
+
upload_checkpoints: false
|
| 278 |
+
upload_every: 1000
|
| 279 |
+
upload_last_n_checkpoints: 5
|
| 280 |
+
max_concurrent_uploads: 2
|
| 281 |
+
eval_strategy: 'no'
|
| 282 |
+
eval_steps: 500
|
| 283 |
+
eval_set_split_ratio: 0.1
|
| 284 |
+
eval_batch_size: 2
|
| 285 |
+
save_best_eval_metric_name: ''
|
| 286 |
+
save_best_eval_metric_greater_is_better: true
|
| 287 |
+
deepspeed_stage: 2
|
| 288 |
+
gradient_checkpointing: false
|
| 289 |
+
transformers_trust_remote_code: true
|
| 290 |
+
transformers_local_files_only: false
|
| 291 |
+
transformers_cache_dir: null
|
| 292 |
+
transformers_access_token: null
|
| 293 |
+
use_ddp: false
|
| 294 |
+
ddp_bucket_cap_mb: 100
|
| 295 |
+
num_gpus: 16
|
| 296 |
+
dataloader_num_workers: 12
|
| 297 |
+
remove_unused_columns: false
|
| 298 |
+
use_wandb: true
|
| 299 |
+
wandb_project: rldx-finetune
|
| 300 |
+
enable_profiling: false
|
| 301 |
+
max_retries: 3
|
| 302 |
+
assert_loss_less_than: null
|
| 303 |
+
add_rl_callback: false
|
| 304 |
+
enable_open_loop_eval: false
|
| 305 |
+
open_loop_eval_traj_ids:
|
| 306 |
+
- 0
|
| 307 |
+
open_loop_eval_steps_per_traj: 100
|
| 308 |
+
open_loop_eval_plot_indices: null
|
| 309 |
+
max_steps: 60000
|
| 310 |
+
save_steps: 1000
|
experiment_cfg/config.yaml
ADDED
|
@@ -0,0 +1,320 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
!!python/object:rldx.configs.base_config.Config
|
| 2 |
+
data: !!python/object:rldx.configs.data.data_config.DataConfig
|
| 3 |
+
allow_padding: true
|
| 4 |
+
dataset_mode: sharded
|
| 5 |
+
datasets:
|
| 6 |
+
- !!python/object:rldx.configs.data.data_config.SingleDatasetConfig
|
| 7 |
+
dataset_paths:
|
| 8 |
+
- /data/taeyoung/data/shared_dataset/robocasa_mg_gr00t_300
|
| 9 |
+
dataset_type: physical_embodiment
|
| 10 |
+
embodiment_tag: general_embodiment
|
| 11 |
+
mix_ratio: 1.0
|
| 12 |
+
val_dataset_path: null
|
| 13 |
+
download_cache: false
|
| 14 |
+
episode_sampling_rate: 0.1
|
| 15 |
+
mock_dataset_mode: false
|
| 16 |
+
modality_configs:
|
| 17 |
+
general_embodiment:
|
| 18 |
+
action: !!python/object:rldx.data.types.ModalityConfig
|
| 19 |
+
action_configs:
|
| 20 |
+
- !!python/object:rldx.data.types.ActionConfig
|
| 21 |
+
format: &id001 !!python/object/apply:rldx.data.types.ActionFormat
|
| 22 |
+
- default
|
| 23 |
+
rep: &id002 !!python/object/apply:rldx.data.types.ActionRepresentation
|
| 24 |
+
- delta
|
| 25 |
+
state_key: null
|
| 26 |
+
type: &id003 !!python/object/apply:rldx.data.types.ActionType
|
| 27 |
+
- eef
|
| 28 |
+
- !!python/object:rldx.data.types.ActionConfig
|
| 29 |
+
format: *id001
|
| 30 |
+
rep: *id002
|
| 31 |
+
state_key: null
|
| 32 |
+
type: *id003
|
| 33 |
+
- !!python/object:rldx.data.types.ActionConfig
|
| 34 |
+
format: *id001
|
| 35 |
+
rep: &id005 !!python/object/apply:rldx.data.types.ActionRepresentation
|
| 36 |
+
- absolute
|
| 37 |
+
state_key: null
|
| 38 |
+
type: &id004 !!python/object/apply:rldx.data.types.ActionType
|
| 39 |
+
- non_eef
|
| 40 |
+
- !!python/object:rldx.data.types.ActionConfig
|
| 41 |
+
format: *id001
|
| 42 |
+
rep: *id002
|
| 43 |
+
state_key: null
|
| 44 |
+
type: *id004
|
| 45 |
+
- !!python/object:rldx.data.types.ActionConfig
|
| 46 |
+
format: *id001
|
| 47 |
+
rep: *id005
|
| 48 |
+
state_key: null
|
| 49 |
+
type: *id004
|
| 50 |
+
delta_indices:
|
| 51 |
+
- 0
|
| 52 |
+
- 1
|
| 53 |
+
- 2
|
| 54 |
+
- 3
|
| 55 |
+
- 4
|
| 56 |
+
- 5
|
| 57 |
+
- 6
|
| 58 |
+
- 7
|
| 59 |
+
- 8
|
| 60 |
+
- 9
|
| 61 |
+
- 10
|
| 62 |
+
- 11
|
| 63 |
+
- 12
|
| 64 |
+
- 13
|
| 65 |
+
- 14
|
| 66 |
+
- 15
|
| 67 |
+
mean_std_embedding_keys: null
|
| 68 |
+
modality_keys:
|
| 69 |
+
- end_effector_position
|
| 70 |
+
- end_effector_rotation
|
| 71 |
+
- gripper_close
|
| 72 |
+
- base_motion
|
| 73 |
+
- control_mode
|
| 74 |
+
sin_cos_embedding_keys: null
|
| 75 |
+
language: !!python/object:rldx.data.types.ModalityConfig
|
| 76 |
+
action_configs: null
|
| 77 |
+
delta_indices:
|
| 78 |
+
- 0
|
| 79 |
+
mean_std_embedding_keys: null
|
| 80 |
+
modality_keys:
|
| 81 |
+
- annotation.human.action.task_description
|
| 82 |
+
sin_cos_embedding_keys: null
|
| 83 |
+
state: !!python/object:rldx.data.types.ModalityConfig
|
| 84 |
+
action_configs: null
|
| 85 |
+
delta_indices:
|
| 86 |
+
- 0
|
| 87 |
+
mean_std_embedding_keys: null
|
| 88 |
+
modality_keys:
|
| 89 |
+
- end_effector_position_relative
|
| 90 |
+
- end_effector_rotation_relative
|
| 91 |
+
- gripper_qpos
|
| 92 |
+
- base_position
|
| 93 |
+
- base_rotation
|
| 94 |
+
sin_cos_embedding_keys: null
|
| 95 |
+
video: !!python/object:rldx.data.types.ModalityConfig
|
| 96 |
+
action_configs: null
|
| 97 |
+
delta_indices:
|
| 98 |
+
- 0
|
| 99 |
+
mean_std_embedding_keys: null
|
| 100 |
+
modality_keys:
|
| 101 |
+
- left_view
|
| 102 |
+
- right_view
|
| 103 |
+
- wrist_view
|
| 104 |
+
sin_cos_embedding_keys: null
|
| 105 |
+
mode: single_turn
|
| 106 |
+
multiprocessing_context: fork
|
| 107 |
+
num_shards_per_epoch: 100000
|
| 108 |
+
override_pretraining_statistics: false
|
| 109 |
+
random_chop: 0.0
|
| 110 |
+
seed: 42
|
| 111 |
+
shard_size: 1024
|
| 112 |
+
shuffle: true
|
| 113 |
+
subsample_ratio: 1.0
|
| 114 |
+
video_backend: torchcodec
|
| 115 |
+
load_config_path: null
|
| 116 |
+
model: !!python/object:rldx.configs.model.rldx.RLDXConfig
|
| 117 |
+
_attn_implementation_internal: null
|
| 118 |
+
_commit_hash: null
|
| 119 |
+
_name_or_path: ''
|
| 120 |
+
_output_attentions: false
|
| 121 |
+
action_horizon: 16
|
| 122 |
+
action_model_lora_alpha: 32
|
| 123 |
+
action_model_lora_dropout: 0.0
|
| 124 |
+
action_model_lora_rank: 16
|
| 125 |
+
action_model_lora_target_modules:
|
| 126 |
+
- vl_qkv
|
| 127 |
+
- vl_proj
|
| 128 |
+
- sa_qkv
|
| 129 |
+
- sa_proj
|
| 130 |
+
- p_qkv
|
| 131 |
+
- p_proj
|
| 132 |
+
- linear1
|
| 133 |
+
- linear2
|
| 134 |
+
action_model_use_lora: false
|
| 135 |
+
add_cross_attention: false
|
| 136 |
+
architectures: null
|
| 137 |
+
backbone_lora_alpha: 32
|
| 138 |
+
backbone_lora_dropout: 0.0
|
| 139 |
+
backbone_lora_num_layers: -1
|
| 140 |
+
backbone_lora_rank: 16
|
| 141 |
+
backbone_lora_target_modules:
|
| 142 |
+
- q_proj
|
| 143 |
+
- k_proj
|
| 144 |
+
- v_proj
|
| 145 |
+
- o_proj
|
| 146 |
+
- gate_proj
|
| 147 |
+
- up_proj
|
| 148 |
+
- down_proj
|
| 149 |
+
backbone_use_lora: false
|
| 150 |
+
bad_words_ids: null
|
| 151 |
+
begin_suppress_tokens: null
|
| 152 |
+
bos_token_id: null
|
| 153 |
+
chunk_size_feed_forward: 0
|
| 154 |
+
color_jitter_params:
|
| 155 |
+
brightness: 0.3
|
| 156 |
+
contrast: 0.4
|
| 157 |
+
hue: 0.08
|
| 158 |
+
saturation: 0.5
|
| 159 |
+
conversation_image_first: false
|
| 160 |
+
cross_attention_hidden_size: null
|
| 161 |
+
decoder_start_token_id: null
|
| 162 |
+
diffusion_model_cfg:
|
| 163 |
+
action_model_max_seq_len: 512
|
| 164 |
+
attention_head_dim: 64
|
| 165 |
+
depth_multi_stream: 4
|
| 166 |
+
depth_single_stream: 8
|
| 167 |
+
dropout: 0.2
|
| 168 |
+
final_dropout: true
|
| 169 |
+
num_attention_heads: 24
|
| 170 |
+
output_dim: 1024
|
| 171 |
+
positional_embeddings: rope_sa_only
|
| 172 |
+
pre_norm: layer_norm
|
| 173 |
+
qk_norm: rms_norm
|
| 174 |
+
rope_theta: 10000.0
|
| 175 |
+
temb_type: input_token
|
| 176 |
+
use_swiglu: true
|
| 177 |
+
diversity_penalty: 0.0
|
| 178 |
+
do_sample: false
|
| 179 |
+
dtype: null
|
| 180 |
+
early_stopping: false
|
| 181 |
+
encoder_no_repeat_ngram_size: 0
|
| 182 |
+
eos_token_id: null
|
| 183 |
+
exponential_decay_length_penalty: null
|
| 184 |
+
finetuning_task: null
|
| 185 |
+
forced_bos_token_id: null
|
| 186 |
+
forced_eos_token_id: null
|
| 187 |
+
freeze_cog_tokens: false
|
| 188 |
+
general_embodiment_train_ratio: 0
|
| 189 |
+
id2label:
|
| 190 |
+
0: LABEL_0
|
| 191 |
+
1: LABEL_1
|
| 192 |
+
image_max_area: 65536
|
| 193 |
+
image_resize_m: 32
|
| 194 |
+
is_decoder: false
|
| 195 |
+
is_encoder_decoder: false
|
| 196 |
+
label2id:
|
| 197 |
+
LABEL_0: 0
|
| 198 |
+
LABEL_1: 1
|
| 199 |
+
length_penalty: 1.0
|
| 200 |
+
max_length: 20
|
| 201 |
+
memory_cfg:
|
| 202 |
+
hidden_size: 4096
|
| 203 |
+
intermediate_size: 16384
|
| 204 |
+
max_position_embeddings: 32
|
| 205 |
+
num_attention_heads: 16
|
| 206 |
+
num_hidden_layers: 2
|
| 207 |
+
num_key_value_heads: 16
|
| 208 |
+
rms_norm_eps: 1.0e-05
|
| 209 |
+
use_causal_attn: true
|
| 210 |
+
use_rope: true
|
| 211 |
+
min_length: 0
|
| 212 |
+
model_name: /data/taeyoung/models/RLDX-1-VLM
|
| 213 |
+
model_revision: null
|
| 214 |
+
model_type: RLDX-1
|
| 215 |
+
n_cog_tokens: 64
|
| 216 |
+
no_repeat_ngram_size: 0
|
| 217 |
+
num_beam_groups: 1
|
| 218 |
+
num_beams: 1
|
| 219 |
+
num_return_sequences: 1
|
| 220 |
+
output_hidden_states: false
|
| 221 |
+
output_scores: false
|
| 222 |
+
pad_token_id: null
|
| 223 |
+
physics_dims: []
|
| 224 |
+
physics_keys: []
|
| 225 |
+
prefix: null
|
| 226 |
+
problem_type: null
|
| 227 |
+
pruned_heads: {}
|
| 228 |
+
random_crop_fraction: null
|
| 229 |
+
random_rotation_angle: null
|
| 230 |
+
remove_invalid_values: false
|
| 231 |
+
repetition_penalty: 1.0
|
| 232 |
+
return_dict: true
|
| 233 |
+
return_dict_in_generate: false
|
| 234 |
+
rtc_inference_delay: 0
|
| 235 |
+
rtc_inference_exec_horizon: 0
|
| 236 |
+
rtc_inference_mode: none
|
| 237 |
+
rtc_jacobian_steps_only: 3
|
| 238 |
+
rtc_training_max_delay: 0
|
| 239 |
+
sep_token_id: null
|
| 240 |
+
state_dropout_prob: 0.0
|
| 241 |
+
suppress_tokens: null
|
| 242 |
+
task_specific_params: null
|
| 243 |
+
temperature: 1.0
|
| 244 |
+
tf_legacy_loss: false
|
| 245 |
+
tie_encoder_decoder: false
|
| 246 |
+
tie_word_embeddings: true
|
| 247 |
+
tokenizer_class: null
|
| 248 |
+
top_k: 50
|
| 249 |
+
top_p: 1.0
|
| 250 |
+
torchscript: false
|
| 251 |
+
transformers_version: null
|
| 252 |
+
tune_diffusion_model: true
|
| 253 |
+
tune_llm: false
|
| 254 |
+
tune_projector: true
|
| 255 |
+
tune_top_llm_layers: 4
|
| 256 |
+
tune_visual: false
|
| 257 |
+
typical_p: 1.0
|
| 258 |
+
use_bfloat16: false
|
| 259 |
+
use_memory: false
|
| 260 |
+
use_motion: false
|
| 261 |
+
use_physics: false
|
| 262 |
+
use_video: true
|
| 263 |
+
video_length: 1
|
| 264 |
+
video_stride: 2
|
| 265 |
+
training: !!python/object:rldx.configs.training.training_config.TrainingConfig
|
| 266 |
+
add_rl_callback: false
|
| 267 |
+
assert_loss_less_than: null
|
| 268 |
+
batch_size: null
|
| 269 |
+
bf16: true
|
| 270 |
+
dataloader_num_workers: 12
|
| 271 |
+
ddp_bucket_cap_mb: 100
|
| 272 |
+
deepspeed_stage: 2
|
| 273 |
+
enable_open_loop_eval: false
|
| 274 |
+
enable_profiling: false
|
| 275 |
+
eval_batch_size: 2
|
| 276 |
+
eval_bf16: true
|
| 277 |
+
eval_set_split_ratio: 0.1
|
| 278 |
+
eval_steps: 500
|
| 279 |
+
eval_strategy: 'no'
|
| 280 |
+
experiment_name: rldx1_image_ft_robocasa300_bsz1024_60k
|
| 281 |
+
fp16: false
|
| 282 |
+
global_batch_size: 1024
|
| 283 |
+
gradient_accumulation_steps: 1
|
| 284 |
+
gradient_checkpointing: false
|
| 285 |
+
learning_rate: 0.0001
|
| 286 |
+
logging_steps: 10
|
| 287 |
+
lr_scheduler_type: cosine
|
| 288 |
+
max_concurrent_uploads: 2
|
| 289 |
+
max_grad_norm: 1.0
|
| 290 |
+
max_retries: 3
|
| 291 |
+
max_steps: 60000
|
| 292 |
+
new_param_warmup_steps: 0
|
| 293 |
+
num_gpus: 16
|
| 294 |
+
open_loop_eval_plot_indices: null
|
| 295 |
+
open_loop_eval_steps_per_traj: 100
|
| 296 |
+
open_loop_eval_traj_ids:
|
| 297 |
+
- 0
|
| 298 |
+
optim: adamw_torch_fused
|
| 299 |
+
output_dir: /data/taeyoung/workspace/RLDX-1/ckpt/rldx1/finetuned/rldx1_image_ft_robocasa300_bsz1024_60k
|
| 300 |
+
remove_unused_columns: false
|
| 301 |
+
save_best_eval_metric_greater_is_better: true
|
| 302 |
+
save_best_eval_metric_name: ''
|
| 303 |
+
save_steps: 1000
|
| 304 |
+
save_total_limit: 10
|
| 305 |
+
save_vl_model: false
|
| 306 |
+
start_from_checkpoint: /data/taeyoung/workspace/RLDX-1/ckpt/rldx1/pretrained/rldx1_image_pt_mixv5_bsz8192_100k/rldx1_image_pt_mixv5_bsz8192_100k/checkpoint-100000
|
| 307 |
+
tf32: true
|
| 308 |
+
transformers_access_token: null
|
| 309 |
+
transformers_cache_dir: null
|
| 310 |
+
transformers_local_files_only: false
|
| 311 |
+
transformers_trust_remote_code: true
|
| 312 |
+
upload_checkpoints: false
|
| 313 |
+
upload_every: 1000
|
| 314 |
+
upload_last_n_checkpoints: 5
|
| 315 |
+
use_ddp: false
|
| 316 |
+
use_wandb: true
|
| 317 |
+
wandb_project: rldx-finetune
|
| 318 |
+
warmup_ratio: 0.05
|
| 319 |
+
warmup_steps: 0
|
| 320 |
+
weight_decay: 1.0e-05
|
experiment_cfg/dataset_statistics.json
ADDED
|
@@ -0,0 +1,316 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"general_embodiment": {
|
| 3 |
+
"action": {
|
| 4 |
+
"end_effector_position": {
|
| 5 |
+
"min": [
|
| 6 |
+
-1.0,
|
| 7 |
+
-1.0,
|
| 8 |
+
-1.0
|
| 9 |
+
],
|
| 10 |
+
"max": [
|
| 11 |
+
1.0,
|
| 12 |
+
1.0,
|
| 13 |
+
1.0
|
| 14 |
+
],
|
| 15 |
+
"mean": [
|
| 16 |
+
0.029150759801268578,
|
| 17 |
+
0.007607606239616871,
|
| 18 |
+
-0.037477411329746246
|
| 19 |
+
],
|
| 20 |
+
"std": [
|
| 21 |
+
0.4175023138523102,
|
| 22 |
+
0.41343387961387634,
|
| 23 |
+
0.36864909529685974
|
| 24 |
+
],
|
| 25 |
+
"q01": [
|
| 26 |
+
-1.0,
|
| 27 |
+
-1.0,
|
| 28 |
+
-1.0
|
| 29 |
+
],
|
| 30 |
+
"q99": [
|
| 31 |
+
1.0,
|
| 32 |
+
1.0,
|
| 33 |
+
1.0
|
| 34 |
+
]
|
| 35 |
+
},
|
| 36 |
+
"end_effector_rotation": {
|
| 37 |
+
"min": [
|
| 38 |
+
-1.0,
|
| 39 |
+
-1.0,
|
| 40 |
+
-1.0
|
| 41 |
+
],
|
| 42 |
+
"max": [
|
| 43 |
+
1.0,
|
| 44 |
+
1.0,
|
| 45 |
+
1.0
|
| 46 |
+
],
|
| 47 |
+
"mean": [
|
| 48 |
+
0.007235987111926079,
|
| 49 |
+
-0.023321982473134995,
|
| 50 |
+
-0.0024119403678923845
|
| 51 |
+
],
|
| 52 |
+
"std": [
|
| 53 |
+
0.147710919380188,
|
| 54 |
+
0.15399158000946045,
|
| 55 |
+
0.13929523527622223
|
| 56 |
+
],
|
| 57 |
+
"q01": [
|
| 58 |
+
-0.42321433901786804,
|
| 59 |
+
-0.5219977021217346,
|
| 60 |
+
-0.4101182746887207
|
| 61 |
+
],
|
| 62 |
+
"q99": [
|
| 63 |
+
0.47503985762596074,
|
| 64 |
+
0.38744416117668146,
|
| 65 |
+
0.4242807936668394
|
| 66 |
+
]
|
| 67 |
+
},
|
| 68 |
+
"gripper_close": {
|
| 69 |
+
"min": [
|
| 70 |
+
0.0
|
| 71 |
+
],
|
| 72 |
+
"max": [
|
| 73 |
+
1.0
|
| 74 |
+
],
|
| 75 |
+
"mean": [
|
| 76 |
+
0.34521719813346863
|
| 77 |
+
],
|
| 78 |
+
"std": [
|
| 79 |
+
0.47949448227882385
|
| 80 |
+
],
|
| 81 |
+
"q01": [
|
| 82 |
+
0.0
|
| 83 |
+
],
|
| 84 |
+
"q99": [
|
| 85 |
+
1.0
|
| 86 |
+
]
|
| 87 |
+
},
|
| 88 |
+
"base_motion": {
|
| 89 |
+
"min": [
|
| 90 |
+
0.0,
|
| 91 |
+
0.0,
|
| 92 |
+
0.0,
|
| 93 |
+
0.0
|
| 94 |
+
],
|
| 95 |
+
"max": [
|
| 96 |
+
0.0,
|
| 97 |
+
0.0,
|
| 98 |
+
0.0,
|
| 99 |
+
0.0
|
| 100 |
+
],
|
| 101 |
+
"mean": [
|
| 102 |
+
0.0,
|
| 103 |
+
0.0,
|
| 104 |
+
0.0,
|
| 105 |
+
0.0
|
| 106 |
+
],
|
| 107 |
+
"std": [
|
| 108 |
+
0.0,
|
| 109 |
+
0.0,
|
| 110 |
+
0.0,
|
| 111 |
+
0.0
|
| 112 |
+
],
|
| 113 |
+
"q01": [
|
| 114 |
+
0.0,
|
| 115 |
+
0.0,
|
| 116 |
+
0.0,
|
| 117 |
+
0.0
|
| 118 |
+
],
|
| 119 |
+
"q99": [
|
| 120 |
+
0.0,
|
| 121 |
+
0.0,
|
| 122 |
+
0.0,
|
| 123 |
+
0.0
|
| 124 |
+
]
|
| 125 |
+
},
|
| 126 |
+
"control_mode": {
|
| 127 |
+
"min": [
|
| 128 |
+
0.0
|
| 129 |
+
],
|
| 130 |
+
"max": [
|
| 131 |
+
0.0
|
| 132 |
+
],
|
| 133 |
+
"mean": [
|
| 134 |
+
0.0
|
| 135 |
+
],
|
| 136 |
+
"std": [
|
| 137 |
+
0.0
|
| 138 |
+
],
|
| 139 |
+
"q01": [
|
| 140 |
+
0.0
|
| 141 |
+
],
|
| 142 |
+
"q99": [
|
| 143 |
+
0.0
|
| 144 |
+
]
|
| 145 |
+
}
|
| 146 |
+
},
|
| 147 |
+
"state": {
|
| 148 |
+
"end_effector_position_relative": {
|
| 149 |
+
"min": [
|
| 150 |
+
-0.46594083309173584,
|
| 151 |
+
-0.8278615474700928,
|
| 152 |
+
0.011948669329285622
|
| 153 |
+
],
|
| 154 |
+
"max": [
|
| 155 |
+
0.8734620809555054,
|
| 156 |
+
0.8441421985626221,
|
| 157 |
+
1.1348456144332886
|
| 158 |
+
],
|
| 159 |
+
"mean": [
|
| 160 |
+
0.266306072473526,
|
| 161 |
+
-0.04002884030342102,
|
| 162 |
+
0.5537278652191162
|
| 163 |
+
],
|
| 164 |
+
"std": [
|
| 165 |
+
0.19139817357063293,
|
| 166 |
+
0.2582354247570038,
|
| 167 |
+
0.2461009919643402
|
| 168 |
+
],
|
| 169 |
+
"q01": [
|
| 170 |
+
-0.251964715719223,
|
| 171 |
+
-0.6177875232696534,
|
| 172 |
+
0.07541620314121246
|
| 173 |
+
],
|
| 174 |
+
"q99": [
|
| 175 |
+
0.736240336894989,
|
| 176 |
+
0.6068667078018188,
|
| 177 |
+
0.9772176861763
|
| 178 |
+
]
|
| 179 |
+
},
|
| 180 |
+
"end_effector_rotation_relative": {
|
| 181 |
+
"min": [
|
| 182 |
+
-0.9999996423721313,
|
| 183 |
+
-0.9910850524902344,
|
| 184 |
+
-0.9571061134338379,
|
| 185 |
+
6.351022108219695e-08
|
| 186 |
+
],
|
| 187 |
+
"max": [
|
| 188 |
+
0.9999980926513672,
|
| 189 |
+
0.990567147731781,
|
| 190 |
+
0.964046835899353,
|
| 191 |
+
0.9740113019943237
|
| 192 |
+
],
|
| 193 |
+
"mean": [
|
| 194 |
+
-0.14963096380233765,
|
| 195 |
+
0.1248803362250328,
|
| 196 |
+
0.0027880165725946426,
|
| 197 |
+
0.2082664668560028
|
| 198 |
+
],
|
| 199 |
+
"std": [
|
| 200 |
+
0.7683019042015076,
|
| 201 |
+
0.36296239495277405,
|
| 202 |
+
0.37635329365730286,
|
| 203 |
+
0.2322801947593689
|
| 204 |
+
],
|
| 205 |
+
"q01": [
|
| 206 |
+
-0.9973158955574035,
|
| 207 |
+
-0.6837371301651001,
|
| 208 |
+
-0.7814695286750794,
|
| 209 |
+
0.0013687612861394884
|
| 210 |
+
],
|
| 211 |
+
"q99": [
|
| 212 |
+
0.9957762026786804,
|
| 213 |
+
0.8491807365417481,
|
| 214 |
+
0.7687161755561829,
|
| 215 |
+
0.802422981262207
|
| 216 |
+
]
|
| 217 |
+
},
|
| 218 |
+
"gripper_qpos": {
|
| 219 |
+
"min": [
|
| 220 |
+
-0.027591193094849586,
|
| 221 |
+
-0.05319468677043915
|
| 222 |
+
],
|
| 223 |
+
"max": [
|
| 224 |
+
0.07896614819765091,
|
| 225 |
+
0.10181591659784317
|
| 226 |
+
],
|
| 227 |
+
"mean": [
|
| 228 |
+
0.030854083597660065,
|
| 229 |
+
-0.03072655387222767
|
| 230 |
+
],
|
| 231 |
+
"std": [
|
| 232 |
+
0.013360796496272087,
|
| 233 |
+
0.013391095213592052
|
| 234 |
+
],
|
| 235 |
+
"q01": [
|
| 236 |
+
-4.0135159069905056e-05,
|
| 237 |
+
-0.04032127052545548
|
| 238 |
+
],
|
| 239 |
+
"q99": [
|
| 240 |
+
0.04066328033804893,
|
| 241 |
+
-0.00040660659200511887
|
| 242 |
+
]
|
| 243 |
+
},
|
| 244 |
+
"base_position": {
|
| 245 |
+
"min": [
|
| 246 |
+
0.31250813603401184,
|
| 247 |
+
-5.100343704223633,
|
| 248 |
+
0.6963803172111511
|
| 249 |
+
],
|
| 250 |
+
"max": [
|
| 251 |
+
5.601607322692871,
|
| 252 |
+
-0.4872598946094513,
|
| 253 |
+
0.7495372295379639
|
| 254 |
+
],
|
| 255 |
+
"mean": [
|
| 256 |
+
2.5315651893615723,
|
| 257 |
+
-1.5020122528076172,
|
| 258 |
+
0.7073471546173096
|
| 259 |
+
],
|
| 260 |
+
"std": [
|
| 261 |
+
1.4670116901397705,
|
| 262 |
+
0.8559004664421082,
|
| 263 |
+
0.006971539929508914
|
| 264 |
+
],
|
| 265 |
+
"q01": [
|
| 266 |
+
0.5000457167625427,
|
| 267 |
+
-4.100119590759277,
|
| 268 |
+
0.6999446749687195
|
| 269 |
+
],
|
| 270 |
+
"q99": [
|
| 271 |
+
5.5999836921691895,
|
| 272 |
+
-0.778001925945282,
|
| 273 |
+
0.7071865797042847
|
| 274 |
+
]
|
| 275 |
+
},
|
| 276 |
+
"base_rotation": {
|
| 277 |
+
"min": [
|
| 278 |
+
0.0,
|
| 279 |
+
0.0,
|
| 280 |
+
-1.0,
|
| 281 |
+
0.0
|
| 282 |
+
],
|
| 283 |
+
"max": [
|
| 284 |
+
0.0,
|
| 285 |
+
0.0,
|
| 286 |
+
1.0,
|
| 287 |
+
1.0
|
| 288 |
+
],
|
| 289 |
+
"mean": [
|
| 290 |
+
0.0,
|
| 291 |
+
0.0,
|
| 292 |
+
0.2765239179134369,
|
| 293 |
+
0.6414101719856262
|
| 294 |
+
],
|
| 295 |
+
"std": [
|
| 296 |
+
0.0,
|
| 297 |
+
0.0,
|
| 298 |
+
0.6192871332168579,
|
| 299 |
+
0.35383468866348267
|
| 300 |
+
],
|
| 301 |
+
"q01": [
|
| 302 |
+
0.0,
|
| 303 |
+
0.0,
|
| 304 |
+
-1.0,
|
| 305 |
+
5.569559434661642e-06
|
| 306 |
+
],
|
| 307 |
+
"q99": [
|
| 308 |
+
0.0,
|
| 309 |
+
0.0,
|
| 310 |
+
1.0,
|
| 311 |
+
1.0
|
| 312 |
+
]
|
| 313 |
+
}
|
| 314 |
+
}
|
| 315 |
+
}
|
| 316 |
+
}
|
experiment_cfg/final_model_config.json
ADDED
|
@@ -0,0 +1,153 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "RLDX-1",
|
| 3 |
+
"model_dtype": "bfloat16",
|
| 4 |
+
"model_name": "/data/taeyoung/models/RLDX-1-VLM",
|
| 5 |
+
"backbone_model_type": "vtc_qwen3_vl",
|
| 6 |
+
"model_revision": null,
|
| 7 |
+
"tune_top_llm_layers": 4,
|
| 8 |
+
"backbone_embedding_dim": 4096,
|
| 9 |
+
"tune_llm": false,
|
| 10 |
+
"tune_visual": false,
|
| 11 |
+
"select_layer": 18,
|
| 12 |
+
"reproject_vision": false,
|
| 13 |
+
"use_flash_attention": true,
|
| 14 |
+
"load_bf16": true,
|
| 15 |
+
"backbone_trainable_params_fp32": true,
|
| 16 |
+
"freeze_cog_tokens": false,
|
| 17 |
+
"image_max_area": 65536,
|
| 18 |
+
"image_resize_m": 32,
|
| 19 |
+
"apply_sincos_state_encoding": false,
|
| 20 |
+
"use_relative_action": true,
|
| 21 |
+
"use_percentiles": true,
|
| 22 |
+
"conversation_image_first": false,
|
| 23 |
+
"max_state_dim": 64,
|
| 24 |
+
"max_action_dim": 64,
|
| 25 |
+
"action_horizon": 16,
|
| 26 |
+
"hidden_size": 1024,
|
| 27 |
+
"input_embedding_dim": 1536,
|
| 28 |
+
"general_embodiment_train_ratio": 0,
|
| 29 |
+
"add_pos_embed": true,
|
| 30 |
+
"attn_dropout": 0.2,
|
| 31 |
+
"use_vlln": false,
|
| 32 |
+
"max_seq_len": 1024,
|
| 33 |
+
"n_cog_tokens": 64,
|
| 34 |
+
"diffusion_model_cfg": {
|
| 35 |
+
"action_model_max_seq_len": 512,
|
| 36 |
+
"attention_head_dim": 64,
|
| 37 |
+
"depth_multi_stream": 4,
|
| 38 |
+
"depth_single_stream": 8,
|
| 39 |
+
"dropout": 0.2,
|
| 40 |
+
"final_dropout": true,
|
| 41 |
+
"num_attention_heads": 24,
|
| 42 |
+
"output_dim": 1024,
|
| 43 |
+
"positional_embeddings": "rope_sa_only",
|
| 44 |
+
"pre_norm": "layer_norm",
|
| 45 |
+
"qk_norm": "rms_norm",
|
| 46 |
+
"rope_theta": 10000.0,
|
| 47 |
+
"temb_type": "input_token",
|
| 48 |
+
"use_swiglu": true,
|
| 49 |
+
"sa_dim": 1536,
|
| 50 |
+
"vl_dim": 4096,
|
| 51 |
+
"use_physics": false,
|
| 52 |
+
"physics_dim": 0
|
| 53 |
+
},
|
| 54 |
+
"num_inference_timesteps": 4,
|
| 55 |
+
"noise_beta_alpha": 1.5,
|
| 56 |
+
"noise_beta_beta": 1.0,
|
| 57 |
+
"noise_s": 0.999,
|
| 58 |
+
"num_timestep_buckets": 1000,
|
| 59 |
+
"tune_projector": true,
|
| 60 |
+
"tune_diffusion_model": true,
|
| 61 |
+
"tune_vlln": true,
|
| 62 |
+
"action_model_use_lora": false,
|
| 63 |
+
"action_model_lora_rank": 16,
|
| 64 |
+
"action_model_lora_alpha": 32,
|
| 65 |
+
"action_model_lora_dropout": 0.0,
|
| 66 |
+
"action_model_lora_target_modules": [
|
| 67 |
+
"vl_qkv",
|
| 68 |
+
"vl_proj",
|
| 69 |
+
"sa_qkv",
|
| 70 |
+
"sa_proj",
|
| 71 |
+
"p_qkv",
|
| 72 |
+
"p_proj",
|
| 73 |
+
"linear1",
|
| 74 |
+
"linear2"
|
| 75 |
+
],
|
| 76 |
+
"backbone_use_lora": false,
|
| 77 |
+
"backbone_lora_rank": 16,
|
| 78 |
+
"backbone_lora_alpha": 32,
|
| 79 |
+
"backbone_lora_dropout": 0.0,
|
| 80 |
+
"backbone_lora_num_layers": -1,
|
| 81 |
+
"backbone_lora_target_modules": [
|
| 82 |
+
"q_proj",
|
| 83 |
+
"k_proj",
|
| 84 |
+
"v_proj",
|
| 85 |
+
"o_proj",
|
| 86 |
+
"gate_proj",
|
| 87 |
+
"up_proj",
|
| 88 |
+
"down_proj"
|
| 89 |
+
],
|
| 90 |
+
"state_dropout_prob": 0.0,
|
| 91 |
+
"state_additive_noise_scale": 0.0,
|
| 92 |
+
"rtc_training_max_delay": 0,
|
| 93 |
+
"rtc_inference_mode": "none",
|
| 94 |
+
"rtc_inference_delay": 0,
|
| 95 |
+
"rtc_inference_exec_horizon": 0,
|
| 96 |
+
"rtc_jacobian_steps_only": 3,
|
| 97 |
+
"use_memory": false,
|
| 98 |
+
"memory_length": 4,
|
| 99 |
+
"memory_n_cog_tokens": null,
|
| 100 |
+
"concat_memory": false,
|
| 101 |
+
"memory_dropout_prob": 0.0,
|
| 102 |
+
"memory_stride": 16,
|
| 103 |
+
"memory_cfg": {
|
| 104 |
+
"hidden_size": 4096,
|
| 105 |
+
"intermediate_size": 16384,
|
| 106 |
+
"max_position_embeddings": 32,
|
| 107 |
+
"num_attention_heads": 16,
|
| 108 |
+
"num_hidden_layers": 2,
|
| 109 |
+
"num_key_value_heads": 16,
|
| 110 |
+
"rms_norm_eps": 1e-05,
|
| 111 |
+
"use_causal_attn": true,
|
| 112 |
+
"use_rope": true
|
| 113 |
+
},
|
| 114 |
+
"use_motion": false,
|
| 115 |
+
"motion_insert_layer": 9,
|
| 116 |
+
"motion_d_hid": 512,
|
| 117 |
+
"motion_window": [
|
| 118 |
+
5,
|
| 119 |
+
9,
|
| 120 |
+
9
|
| 121 |
+
],
|
| 122 |
+
"motion_ext_chnls": [
|
| 123 |
+
256
|
| 124 |
+
],
|
| 125 |
+
"motion_int_chnls": [
|
| 126 |
+
256,
|
| 127 |
+
256,
|
| 128 |
+
512
|
| 129 |
+
],
|
| 130 |
+
"motion_corr_func": "cosine",
|
| 131 |
+
"motion_n_encoders": 1,
|
| 132 |
+
"motion_use_layerscale": false,
|
| 133 |
+
"motion_layerscale_init": 1e-05,
|
| 134 |
+
"motion_use_layernorm": false,
|
| 135 |
+
"motion_use_syncbn": false,
|
| 136 |
+
"motion_injection_point": "vision_encoder",
|
| 137 |
+
"motion_pool_type": "avg",
|
| 138 |
+
"motion_drop": true,
|
| 139 |
+
"motion_gradient_check": false,
|
| 140 |
+
"motion_int_mode": "lite",
|
| 141 |
+
"use_video": true,
|
| 142 |
+
"video_length": 1,
|
| 143 |
+
"video_stride": 2,
|
| 144 |
+
"use_physics": false,
|
| 145 |
+
"physics_keys": [],
|
| 146 |
+
"physics_dims": [],
|
| 147 |
+
"physics_loss_weight": 0.1,
|
| 148 |
+
"allow_missing_physics": false,
|
| 149 |
+
"physics_delta_indices": null,
|
| 150 |
+
"physics_use_flow_matching": true,
|
| 151 |
+
"physics_dropout_prob": 0.0,
|
| 152 |
+
"max_num_embodiments": 36
|
| 153 |
+
}
|
experiment_cfg/final_processor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model-00001-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:943972daafa84c6e8320f48cf5fe233b58170d144a508cdaf5070b9a66bcf9db
|
| 3 |
+
size 4912540968
|
model-00002-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b91a0ff1bb5e351533e2946085f4232b71cf4d5968a5bdbd4dc806edf8c3ba6c
|
| 3 |
+
size 4446192352
|
model-00003-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf7444d513d3a226e563201175ff305ff4f848ddb8ca73b5c59af929aa196560
|
| 3 |
+
size 4467155576
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,730 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 6912894784,
|
| 4 |
+
"total_size": 13825789568
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"action_model.action_decoder.layer1.W": "model-00003-of-00003.safetensors",
|
| 8 |
+
"action_model.action_decoder.layer1.b": "model-00003-of-00003.safetensors",
|
| 9 |
+
"action_model.action_decoder.layer2.W": "model-00003-of-00003.safetensors",
|
| 10 |
+
"action_model.action_decoder.layer2.b": "model-00003-of-00003.safetensors",
|
| 11 |
+
"action_model.action_encoder.W1.W": "model-00003-of-00003.safetensors",
|
| 12 |
+
"action_model.action_encoder.W1.b": "model-00003-of-00003.safetensors",
|
| 13 |
+
"action_model.action_encoder.W2.W": "model-00003-of-00003.safetensors",
|
| 14 |
+
"action_model.action_encoder.W2.b": "model-00003-of-00003.safetensors",
|
| 15 |
+
"action_model.action_encoder.W3.W": "model-00003-of-00003.safetensors",
|
| 16 |
+
"action_model.action_encoder.W3.b": "model-00003-of-00003.safetensors",
|
| 17 |
+
"action_model.model.double_blocks.0.k_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 18 |
+
"action_model.model.double_blocks.0.k_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 19 |
+
"action_model.model.double_blocks.0.q_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 20 |
+
"action_model.model.double_blocks.0.q_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 21 |
+
"action_model.model.double_blocks.0.sa_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 22 |
+
"action_model.model.double_blocks.0.sa_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 23 |
+
"action_model.model.double_blocks.0.sa_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 24 |
+
"action_model.model.double_blocks.0.sa_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 25 |
+
"action_model.model.double_blocks.0.sa_proj.bias": "model-00003-of-00003.safetensors",
|
| 26 |
+
"action_model.model.double_blocks.0.sa_proj.weight": "model-00003-of-00003.safetensors",
|
| 27 |
+
"action_model.model.double_blocks.0.sa_qkv.bias": "model-00003-of-00003.safetensors",
|
| 28 |
+
"action_model.model.double_blocks.0.sa_qkv.weight": "model-00003-of-00003.safetensors",
|
| 29 |
+
"action_model.model.double_blocks.0.vl_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 30 |
+
"action_model.model.double_blocks.0.vl_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 31 |
+
"action_model.model.double_blocks.0.vl_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 32 |
+
"action_model.model.double_blocks.0.vl_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 33 |
+
"action_model.model.double_blocks.0.vl_proj.bias": "model-00003-of-00003.safetensors",
|
| 34 |
+
"action_model.model.double_blocks.0.vl_proj.weight": "model-00003-of-00003.safetensors",
|
| 35 |
+
"action_model.model.double_blocks.0.vl_qkv.bias": "model-00003-of-00003.safetensors",
|
| 36 |
+
"action_model.model.double_blocks.0.vl_qkv.weight": "model-00003-of-00003.safetensors",
|
| 37 |
+
"action_model.model.double_blocks.1.k_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 38 |
+
"action_model.model.double_blocks.1.k_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 39 |
+
"action_model.model.double_blocks.1.q_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 40 |
+
"action_model.model.double_blocks.1.q_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 41 |
+
"action_model.model.double_blocks.1.sa_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 42 |
+
"action_model.model.double_blocks.1.sa_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 43 |
+
"action_model.model.double_blocks.1.sa_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 44 |
+
"action_model.model.double_blocks.1.sa_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 45 |
+
"action_model.model.double_blocks.1.sa_proj.bias": "model-00003-of-00003.safetensors",
|
| 46 |
+
"action_model.model.double_blocks.1.sa_proj.weight": "model-00003-of-00003.safetensors",
|
| 47 |
+
"action_model.model.double_blocks.1.sa_qkv.bias": "model-00003-of-00003.safetensors",
|
| 48 |
+
"action_model.model.double_blocks.1.sa_qkv.weight": "model-00003-of-00003.safetensors",
|
| 49 |
+
"action_model.model.double_blocks.1.vl_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 50 |
+
"action_model.model.double_blocks.1.vl_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 51 |
+
"action_model.model.double_blocks.1.vl_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 52 |
+
"action_model.model.double_blocks.1.vl_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 53 |
+
"action_model.model.double_blocks.1.vl_proj.bias": "model-00003-of-00003.safetensors",
|
| 54 |
+
"action_model.model.double_blocks.1.vl_proj.weight": "model-00003-of-00003.safetensors",
|
| 55 |
+
"action_model.model.double_blocks.1.vl_qkv.bias": "model-00003-of-00003.safetensors",
|
| 56 |
+
"action_model.model.double_blocks.1.vl_qkv.weight": "model-00003-of-00003.safetensors",
|
| 57 |
+
"action_model.model.double_blocks.2.k_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 58 |
+
"action_model.model.double_blocks.2.k_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 59 |
+
"action_model.model.double_blocks.2.q_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 60 |
+
"action_model.model.double_blocks.2.q_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 61 |
+
"action_model.model.double_blocks.2.sa_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 62 |
+
"action_model.model.double_blocks.2.sa_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 63 |
+
"action_model.model.double_blocks.2.sa_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 64 |
+
"action_model.model.double_blocks.2.sa_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 65 |
+
"action_model.model.double_blocks.2.sa_proj.bias": "model-00003-of-00003.safetensors",
|
| 66 |
+
"action_model.model.double_blocks.2.sa_proj.weight": "model-00003-of-00003.safetensors",
|
| 67 |
+
"action_model.model.double_blocks.2.sa_qkv.bias": "model-00003-of-00003.safetensors",
|
| 68 |
+
"action_model.model.double_blocks.2.sa_qkv.weight": "model-00003-of-00003.safetensors",
|
| 69 |
+
"action_model.model.double_blocks.2.vl_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 70 |
+
"action_model.model.double_blocks.2.vl_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 71 |
+
"action_model.model.double_blocks.2.vl_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 72 |
+
"action_model.model.double_blocks.2.vl_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 73 |
+
"action_model.model.double_blocks.2.vl_proj.bias": "model-00003-of-00003.safetensors",
|
| 74 |
+
"action_model.model.double_blocks.2.vl_proj.weight": "model-00003-of-00003.safetensors",
|
| 75 |
+
"action_model.model.double_blocks.2.vl_qkv.bias": "model-00003-of-00003.safetensors",
|
| 76 |
+
"action_model.model.double_blocks.2.vl_qkv.weight": "model-00003-of-00003.safetensors",
|
| 77 |
+
"action_model.model.double_blocks.3.k_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 78 |
+
"action_model.model.double_blocks.3.k_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 79 |
+
"action_model.model.double_blocks.3.q_norm_sa.weight": "model-00003-of-00003.safetensors",
|
| 80 |
+
"action_model.model.double_blocks.3.q_norm_vl.weight": "model-00003-of-00003.safetensors",
|
| 81 |
+
"action_model.model.double_blocks.3.sa_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 82 |
+
"action_model.model.double_blocks.3.sa_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 83 |
+
"action_model.model.double_blocks.3.sa_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 84 |
+
"action_model.model.double_blocks.3.sa_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 85 |
+
"action_model.model.double_blocks.3.sa_proj.bias": "model-00003-of-00003.safetensors",
|
| 86 |
+
"action_model.model.double_blocks.3.sa_proj.weight": "model-00003-of-00003.safetensors",
|
| 87 |
+
"action_model.model.double_blocks.3.sa_qkv.bias": "model-00003-of-00003.safetensors",
|
| 88 |
+
"action_model.model.double_blocks.3.sa_qkv.weight": "model-00003-of-00003.safetensors",
|
| 89 |
+
"action_model.model.double_blocks.3.vl_mlp.w12.bias": "model-00003-of-00003.safetensors",
|
| 90 |
+
"action_model.model.double_blocks.3.vl_mlp.w12.weight": "model-00003-of-00003.safetensors",
|
| 91 |
+
"action_model.model.double_blocks.3.vl_mlp.w3.bias": "model-00003-of-00003.safetensors",
|
| 92 |
+
"action_model.model.double_blocks.3.vl_mlp.w3.weight": "model-00003-of-00003.safetensors",
|
| 93 |
+
"action_model.model.double_blocks.3.vl_proj.bias": "model-00003-of-00003.safetensors",
|
| 94 |
+
"action_model.model.double_blocks.3.vl_proj.weight": "model-00003-of-00003.safetensors",
|
| 95 |
+
"action_model.model.double_blocks.3.vl_qkv.bias": "model-00003-of-00003.safetensors",
|
| 96 |
+
"action_model.model.double_blocks.3.vl_qkv.weight": "model-00003-of-00003.safetensors",
|
| 97 |
+
"action_model.model.proj_out_1.bias": "model-00003-of-00003.safetensors",
|
| 98 |
+
"action_model.model.proj_out_1.weight": "model-00003-of-00003.safetensors",
|
| 99 |
+
"action_model.model.proj_out_2.bias": "model-00003-of-00003.safetensors",
|
| 100 |
+
"action_model.model.proj_out_2.weight": "model-00003-of-00003.safetensors",
|
| 101 |
+
"action_model.model.single_blocks.0.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 102 |
+
"action_model.model.single_blocks.0.linear1.bias": "model-00003-of-00003.safetensors",
|
| 103 |
+
"action_model.model.single_blocks.0.linear1.weight": "model-00003-of-00003.safetensors",
|
| 104 |
+
"action_model.model.single_blocks.0.linear2.bias": "model-00003-of-00003.safetensors",
|
| 105 |
+
"action_model.model.single_blocks.0.linear2.weight": "model-00003-of-00003.safetensors",
|
| 106 |
+
"action_model.model.single_blocks.0.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 107 |
+
"action_model.model.single_blocks.0.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 108 |
+
"action_model.model.single_blocks.0.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 109 |
+
"action_model.model.single_blocks.1.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 110 |
+
"action_model.model.single_blocks.1.linear1.bias": "model-00003-of-00003.safetensors",
|
| 111 |
+
"action_model.model.single_blocks.1.linear1.weight": "model-00003-of-00003.safetensors",
|
| 112 |
+
"action_model.model.single_blocks.1.linear2.bias": "model-00003-of-00003.safetensors",
|
| 113 |
+
"action_model.model.single_blocks.1.linear2.weight": "model-00003-of-00003.safetensors",
|
| 114 |
+
"action_model.model.single_blocks.1.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 115 |
+
"action_model.model.single_blocks.1.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 116 |
+
"action_model.model.single_blocks.1.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 117 |
+
"action_model.model.single_blocks.2.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 118 |
+
"action_model.model.single_blocks.2.linear1.bias": "model-00003-of-00003.safetensors",
|
| 119 |
+
"action_model.model.single_blocks.2.linear1.weight": "model-00003-of-00003.safetensors",
|
| 120 |
+
"action_model.model.single_blocks.2.linear2.bias": "model-00003-of-00003.safetensors",
|
| 121 |
+
"action_model.model.single_blocks.2.linear2.weight": "model-00003-of-00003.safetensors",
|
| 122 |
+
"action_model.model.single_blocks.2.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 123 |
+
"action_model.model.single_blocks.2.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 124 |
+
"action_model.model.single_blocks.2.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 125 |
+
"action_model.model.single_blocks.3.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 126 |
+
"action_model.model.single_blocks.3.linear1.bias": "model-00003-of-00003.safetensors",
|
| 127 |
+
"action_model.model.single_blocks.3.linear1.weight": "model-00003-of-00003.safetensors",
|
| 128 |
+
"action_model.model.single_blocks.3.linear2.bias": "model-00003-of-00003.safetensors",
|
| 129 |
+
"action_model.model.single_blocks.3.linear2.weight": "model-00003-of-00003.safetensors",
|
| 130 |
+
"action_model.model.single_blocks.3.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 131 |
+
"action_model.model.single_blocks.3.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 132 |
+
"action_model.model.single_blocks.3.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 133 |
+
"action_model.model.single_blocks.4.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 134 |
+
"action_model.model.single_blocks.4.linear1.bias": "model-00003-of-00003.safetensors",
|
| 135 |
+
"action_model.model.single_blocks.4.linear1.weight": "model-00003-of-00003.safetensors",
|
| 136 |
+
"action_model.model.single_blocks.4.linear2.bias": "model-00003-of-00003.safetensors",
|
| 137 |
+
"action_model.model.single_blocks.4.linear2.weight": "model-00003-of-00003.safetensors",
|
| 138 |
+
"action_model.model.single_blocks.4.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 139 |
+
"action_model.model.single_blocks.4.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 140 |
+
"action_model.model.single_blocks.4.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 141 |
+
"action_model.model.single_blocks.5.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 142 |
+
"action_model.model.single_blocks.5.linear1.bias": "model-00003-of-00003.safetensors",
|
| 143 |
+
"action_model.model.single_blocks.5.linear1.weight": "model-00003-of-00003.safetensors",
|
| 144 |
+
"action_model.model.single_blocks.5.linear2.bias": "model-00003-of-00003.safetensors",
|
| 145 |
+
"action_model.model.single_blocks.5.linear2.weight": "model-00003-of-00003.safetensors",
|
| 146 |
+
"action_model.model.single_blocks.5.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 147 |
+
"action_model.model.single_blocks.5.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 148 |
+
"action_model.model.single_blocks.5.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 149 |
+
"action_model.model.single_blocks.6.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 150 |
+
"action_model.model.single_blocks.6.linear1.bias": "model-00003-of-00003.safetensors",
|
| 151 |
+
"action_model.model.single_blocks.6.linear1.weight": "model-00003-of-00003.safetensors",
|
| 152 |
+
"action_model.model.single_blocks.6.linear2.bias": "model-00003-of-00003.safetensors",
|
| 153 |
+
"action_model.model.single_blocks.6.linear2.weight": "model-00003-of-00003.safetensors",
|
| 154 |
+
"action_model.model.single_blocks.6.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 155 |
+
"action_model.model.single_blocks.6.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 156 |
+
"action_model.model.single_blocks.6.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 157 |
+
"action_model.model.single_blocks.7.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 158 |
+
"action_model.model.single_blocks.7.linear1.bias": "model-00003-of-00003.safetensors",
|
| 159 |
+
"action_model.model.single_blocks.7.linear1.weight": "model-00003-of-00003.safetensors",
|
| 160 |
+
"action_model.model.single_blocks.7.linear2.bias": "model-00003-of-00003.safetensors",
|
| 161 |
+
"action_model.model.single_blocks.7.linear2.weight": "model-00003-of-00003.safetensors",
|
| 162 |
+
"action_model.model.single_blocks.7.mlp_proj.bias": "model-00003-of-00003.safetensors",
|
| 163 |
+
"action_model.model.single_blocks.7.mlp_proj.weight": "model-00003-of-00003.safetensors",
|
| 164 |
+
"action_model.model.single_blocks.7.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 165 |
+
"action_model.model.timestep_encoder.timestep_embedder.linear_1.bias": "model-00003-of-00003.safetensors",
|
| 166 |
+
"action_model.model.timestep_encoder.timestep_embedder.linear_1.weight": "model-00003-of-00003.safetensors",
|
| 167 |
+
"action_model.model.timestep_encoder.timestep_embedder.linear_2.bias": "model-00003-of-00003.safetensors",
|
| 168 |
+
"action_model.model.timestep_encoder.timestep_embedder.linear_2.weight": "model-00003-of-00003.safetensors",
|
| 169 |
+
"action_model.model.vl_proj_to_sa.bias": "model-00003-of-00003.safetensors",
|
| 170 |
+
"action_model.model.vl_proj_to_sa.weight": "model-00003-of-00003.safetensors",
|
| 171 |
+
"action_model.position_embedding.weight": "model-00003-of-00003.safetensors",
|
| 172 |
+
"action_model.state_encoder.layer1.W": "model-00003-of-00003.safetensors",
|
| 173 |
+
"action_model.state_encoder.layer1.b": "model-00003-of-00003.safetensors",
|
| 174 |
+
"action_model.state_encoder.layer2.W": "model-00003-of-00003.safetensors",
|
| 175 |
+
"action_model.state_encoder.layer2.b": "model-00003-of-00003.safetensors",
|
| 176 |
+
"backbone.cog_emb": "model-00001-of-00003.safetensors",
|
| 177 |
+
"backbone.qwen_model.lm_head.weight": "model-00003-of-00003.safetensors",
|
| 178 |
+
"backbone.qwen_model.model.language_model.embed_tokens.weight": "model-00001-of-00003.safetensors",
|
| 179 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 180 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 181 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 182 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 183 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 184 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 185 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 186 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 187 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 188 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 189 |
+
"backbone.qwen_model.model.language_model.layers.0.layer.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 190 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 191 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 192 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 193 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 194 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 195 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 196 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 197 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 198 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 199 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 200 |
+
"backbone.qwen_model.model.language_model.layers.1.layer.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 201 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 202 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 203 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 204 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 205 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 206 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 207 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 208 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 209 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 210 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 211 |
+
"backbone.qwen_model.model.language_model.layers.10.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 212 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 213 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 214 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 215 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 216 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 217 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 218 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 219 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 220 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 221 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 222 |
+
"backbone.qwen_model.model.language_model.layers.11.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 223 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 224 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 225 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 226 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 227 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 228 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 229 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 230 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 231 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 232 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 233 |
+
"backbone.qwen_model.model.language_model.layers.12.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 234 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 235 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 236 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 237 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 238 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 239 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 240 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 241 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 242 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 243 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 244 |
+
"backbone.qwen_model.model.language_model.layers.13.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 245 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 246 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 247 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 248 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 249 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 250 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 251 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 252 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 253 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 254 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 255 |
+
"backbone.qwen_model.model.language_model.layers.14.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 256 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 257 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 258 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 259 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 260 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 261 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 262 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 263 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 264 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 265 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 266 |
+
"backbone.qwen_model.model.language_model.layers.15.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 267 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 268 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 269 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 270 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 271 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 272 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 273 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 274 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 275 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 276 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 277 |
+
"backbone.qwen_model.model.language_model.layers.16.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 278 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 279 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 280 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 281 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 282 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 283 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 284 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 285 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 286 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 287 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 288 |
+
"backbone.qwen_model.model.language_model.layers.17.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 289 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 290 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 291 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 292 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 293 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 294 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 295 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 296 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 297 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 298 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 299 |
+
"backbone.qwen_model.model.language_model.layers.2.layer.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 300 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 301 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 302 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 303 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 304 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 305 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 306 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 307 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 308 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 309 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 310 |
+
"backbone.qwen_model.model.language_model.layers.3.layer.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 311 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 312 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 313 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 314 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 315 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 316 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 317 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 318 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 319 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 320 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 321 |
+
"backbone.qwen_model.model.language_model.layers.4.layer.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 322 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 323 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 324 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 325 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 326 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 327 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 328 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 329 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 330 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 331 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 332 |
+
"backbone.qwen_model.model.language_model.layers.5.layer.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 333 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 334 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 335 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 336 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 337 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 338 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 339 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 340 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 341 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 342 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 343 |
+
"backbone.qwen_model.model.language_model.layers.6.layer.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 344 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 345 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 346 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 347 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 348 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 349 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 350 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 351 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 352 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 353 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 354 |
+
"backbone.qwen_model.model.language_model.layers.7.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 355 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 356 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 357 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 358 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 359 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 360 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 361 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 362 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 363 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 364 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 365 |
+
"backbone.qwen_model.model.language_model.layers.8.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 366 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 367 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 368 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 369 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 370 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 371 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 372 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 373 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 374 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 375 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 376 |
+
"backbone.qwen_model.model.language_model.layers.9.layer.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 377 |
+
"backbone.qwen_model.model.language_model.norm.weight": "model-00002-of-00003.safetensors",
|
| 378 |
+
"backbone.qwen_model.model.visual.blocks.0.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 379 |
+
"backbone.qwen_model.model.visual.blocks.0.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 380 |
+
"backbone.qwen_model.model.visual.blocks.0.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 381 |
+
"backbone.qwen_model.model.visual.blocks.0.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 382 |
+
"backbone.qwen_model.model.visual.blocks.0.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 383 |
+
"backbone.qwen_model.model.visual.blocks.0.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 384 |
+
"backbone.qwen_model.model.visual.blocks.0.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 385 |
+
"backbone.qwen_model.model.visual.blocks.0.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 386 |
+
"backbone.qwen_model.model.visual.blocks.0.norm1.bias": "model-00001-of-00003.safetensors",
|
| 387 |
+
"backbone.qwen_model.model.visual.blocks.0.norm1.weight": "model-00001-of-00003.safetensors",
|
| 388 |
+
"backbone.qwen_model.model.visual.blocks.0.norm2.bias": "model-00001-of-00003.safetensors",
|
| 389 |
+
"backbone.qwen_model.model.visual.blocks.0.norm2.weight": "model-00001-of-00003.safetensors",
|
| 390 |
+
"backbone.qwen_model.model.visual.blocks.1.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 391 |
+
"backbone.qwen_model.model.visual.blocks.1.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 392 |
+
"backbone.qwen_model.model.visual.blocks.1.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 393 |
+
"backbone.qwen_model.model.visual.blocks.1.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 394 |
+
"backbone.qwen_model.model.visual.blocks.1.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 395 |
+
"backbone.qwen_model.model.visual.blocks.1.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 396 |
+
"backbone.qwen_model.model.visual.blocks.1.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 397 |
+
"backbone.qwen_model.model.visual.blocks.1.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 398 |
+
"backbone.qwen_model.model.visual.blocks.1.norm1.bias": "model-00001-of-00003.safetensors",
|
| 399 |
+
"backbone.qwen_model.model.visual.blocks.1.norm1.weight": "model-00001-of-00003.safetensors",
|
| 400 |
+
"backbone.qwen_model.model.visual.blocks.1.norm2.bias": "model-00001-of-00003.safetensors",
|
| 401 |
+
"backbone.qwen_model.model.visual.blocks.1.norm2.weight": "model-00001-of-00003.safetensors",
|
| 402 |
+
"backbone.qwen_model.model.visual.blocks.10.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 403 |
+
"backbone.qwen_model.model.visual.blocks.10.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 404 |
+
"backbone.qwen_model.model.visual.blocks.10.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 405 |
+
"backbone.qwen_model.model.visual.blocks.10.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 406 |
+
"backbone.qwen_model.model.visual.blocks.10.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 407 |
+
"backbone.qwen_model.model.visual.blocks.10.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 408 |
+
"backbone.qwen_model.model.visual.blocks.10.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 409 |
+
"backbone.qwen_model.model.visual.blocks.10.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 410 |
+
"backbone.qwen_model.model.visual.blocks.10.norm1.bias": "model-00001-of-00003.safetensors",
|
| 411 |
+
"backbone.qwen_model.model.visual.blocks.10.norm1.weight": "model-00001-of-00003.safetensors",
|
| 412 |
+
"backbone.qwen_model.model.visual.blocks.10.norm2.bias": "model-00001-of-00003.safetensors",
|
| 413 |
+
"backbone.qwen_model.model.visual.blocks.10.norm2.weight": "model-00001-of-00003.safetensors",
|
| 414 |
+
"backbone.qwen_model.model.visual.blocks.11.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 415 |
+
"backbone.qwen_model.model.visual.blocks.11.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 416 |
+
"backbone.qwen_model.model.visual.blocks.11.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 417 |
+
"backbone.qwen_model.model.visual.blocks.11.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 418 |
+
"backbone.qwen_model.model.visual.blocks.11.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 419 |
+
"backbone.qwen_model.model.visual.blocks.11.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 420 |
+
"backbone.qwen_model.model.visual.blocks.11.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 421 |
+
"backbone.qwen_model.model.visual.blocks.11.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 422 |
+
"backbone.qwen_model.model.visual.blocks.11.norm1.bias": "model-00001-of-00003.safetensors",
|
| 423 |
+
"backbone.qwen_model.model.visual.blocks.11.norm1.weight": "model-00001-of-00003.safetensors",
|
| 424 |
+
"backbone.qwen_model.model.visual.blocks.11.norm2.bias": "model-00001-of-00003.safetensors",
|
| 425 |
+
"backbone.qwen_model.model.visual.blocks.11.norm2.weight": "model-00001-of-00003.safetensors",
|
| 426 |
+
"backbone.qwen_model.model.visual.blocks.12.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 427 |
+
"backbone.qwen_model.model.visual.blocks.12.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 428 |
+
"backbone.qwen_model.model.visual.blocks.12.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 429 |
+
"backbone.qwen_model.model.visual.blocks.12.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 430 |
+
"backbone.qwen_model.model.visual.blocks.12.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 431 |
+
"backbone.qwen_model.model.visual.blocks.12.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 432 |
+
"backbone.qwen_model.model.visual.blocks.12.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 433 |
+
"backbone.qwen_model.model.visual.blocks.12.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 434 |
+
"backbone.qwen_model.model.visual.blocks.12.norm1.bias": "model-00001-of-00003.safetensors",
|
| 435 |
+
"backbone.qwen_model.model.visual.blocks.12.norm1.weight": "model-00001-of-00003.safetensors",
|
| 436 |
+
"backbone.qwen_model.model.visual.blocks.12.norm2.bias": "model-00001-of-00003.safetensors",
|
| 437 |
+
"backbone.qwen_model.model.visual.blocks.12.norm2.weight": "model-00001-of-00003.safetensors",
|
| 438 |
+
"backbone.qwen_model.model.visual.blocks.13.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 439 |
+
"backbone.qwen_model.model.visual.blocks.13.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 440 |
+
"backbone.qwen_model.model.visual.blocks.13.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 441 |
+
"backbone.qwen_model.model.visual.blocks.13.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 442 |
+
"backbone.qwen_model.model.visual.blocks.13.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 443 |
+
"backbone.qwen_model.model.visual.blocks.13.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 444 |
+
"backbone.qwen_model.model.visual.blocks.13.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 445 |
+
"backbone.qwen_model.model.visual.blocks.13.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 446 |
+
"backbone.qwen_model.model.visual.blocks.13.norm1.bias": "model-00001-of-00003.safetensors",
|
| 447 |
+
"backbone.qwen_model.model.visual.blocks.13.norm1.weight": "model-00001-of-00003.safetensors",
|
| 448 |
+
"backbone.qwen_model.model.visual.blocks.13.norm2.bias": "model-00001-of-00003.safetensors",
|
| 449 |
+
"backbone.qwen_model.model.visual.blocks.13.norm2.weight": "model-00001-of-00003.safetensors",
|
| 450 |
+
"backbone.qwen_model.model.visual.blocks.14.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 451 |
+
"backbone.qwen_model.model.visual.blocks.14.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 452 |
+
"backbone.qwen_model.model.visual.blocks.14.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 453 |
+
"backbone.qwen_model.model.visual.blocks.14.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 454 |
+
"backbone.qwen_model.model.visual.blocks.14.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 455 |
+
"backbone.qwen_model.model.visual.blocks.14.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 456 |
+
"backbone.qwen_model.model.visual.blocks.14.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 457 |
+
"backbone.qwen_model.model.visual.blocks.14.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 458 |
+
"backbone.qwen_model.model.visual.blocks.14.norm1.bias": "model-00001-of-00003.safetensors",
|
| 459 |
+
"backbone.qwen_model.model.visual.blocks.14.norm1.weight": "model-00001-of-00003.safetensors",
|
| 460 |
+
"backbone.qwen_model.model.visual.blocks.14.norm2.bias": "model-00001-of-00003.safetensors",
|
| 461 |
+
"backbone.qwen_model.model.visual.blocks.14.norm2.weight": "model-00001-of-00003.safetensors",
|
| 462 |
+
"backbone.qwen_model.model.visual.blocks.15.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 463 |
+
"backbone.qwen_model.model.visual.blocks.15.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 464 |
+
"backbone.qwen_model.model.visual.blocks.15.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 465 |
+
"backbone.qwen_model.model.visual.blocks.15.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 466 |
+
"backbone.qwen_model.model.visual.blocks.15.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 467 |
+
"backbone.qwen_model.model.visual.blocks.15.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 468 |
+
"backbone.qwen_model.model.visual.blocks.15.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 469 |
+
"backbone.qwen_model.model.visual.blocks.15.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 470 |
+
"backbone.qwen_model.model.visual.blocks.15.norm1.bias": "model-00001-of-00003.safetensors",
|
| 471 |
+
"backbone.qwen_model.model.visual.blocks.15.norm1.weight": "model-00001-of-00003.safetensors",
|
| 472 |
+
"backbone.qwen_model.model.visual.blocks.15.norm2.bias": "model-00001-of-00003.safetensors",
|
| 473 |
+
"backbone.qwen_model.model.visual.blocks.15.norm2.weight": "model-00001-of-00003.safetensors",
|
| 474 |
+
"backbone.qwen_model.model.visual.blocks.16.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 475 |
+
"backbone.qwen_model.model.visual.blocks.16.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 476 |
+
"backbone.qwen_model.model.visual.blocks.16.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 477 |
+
"backbone.qwen_model.model.visual.blocks.16.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 478 |
+
"backbone.qwen_model.model.visual.blocks.16.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 479 |
+
"backbone.qwen_model.model.visual.blocks.16.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 480 |
+
"backbone.qwen_model.model.visual.blocks.16.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 481 |
+
"backbone.qwen_model.model.visual.blocks.16.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 482 |
+
"backbone.qwen_model.model.visual.blocks.16.norm1.bias": "model-00001-of-00003.safetensors",
|
| 483 |
+
"backbone.qwen_model.model.visual.blocks.16.norm1.weight": "model-00001-of-00003.safetensors",
|
| 484 |
+
"backbone.qwen_model.model.visual.blocks.16.norm2.bias": "model-00001-of-00003.safetensors",
|
| 485 |
+
"backbone.qwen_model.model.visual.blocks.16.norm2.weight": "model-00001-of-00003.safetensors",
|
| 486 |
+
"backbone.qwen_model.model.visual.blocks.17.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 487 |
+
"backbone.qwen_model.model.visual.blocks.17.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 488 |
+
"backbone.qwen_model.model.visual.blocks.17.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 489 |
+
"backbone.qwen_model.model.visual.blocks.17.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 490 |
+
"backbone.qwen_model.model.visual.blocks.17.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 491 |
+
"backbone.qwen_model.model.visual.blocks.17.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 492 |
+
"backbone.qwen_model.model.visual.blocks.17.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 493 |
+
"backbone.qwen_model.model.visual.blocks.17.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 494 |
+
"backbone.qwen_model.model.visual.blocks.17.norm1.bias": "model-00001-of-00003.safetensors",
|
| 495 |
+
"backbone.qwen_model.model.visual.blocks.17.norm1.weight": "model-00001-of-00003.safetensors",
|
| 496 |
+
"backbone.qwen_model.model.visual.blocks.17.norm2.bias": "model-00001-of-00003.safetensors",
|
| 497 |
+
"backbone.qwen_model.model.visual.blocks.17.norm2.weight": "model-00001-of-00003.safetensors",
|
| 498 |
+
"backbone.qwen_model.model.visual.blocks.18.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 499 |
+
"backbone.qwen_model.model.visual.blocks.18.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 500 |
+
"backbone.qwen_model.model.visual.blocks.18.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 501 |
+
"backbone.qwen_model.model.visual.blocks.18.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 502 |
+
"backbone.qwen_model.model.visual.blocks.18.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 503 |
+
"backbone.qwen_model.model.visual.blocks.18.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 504 |
+
"backbone.qwen_model.model.visual.blocks.18.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 505 |
+
"backbone.qwen_model.model.visual.blocks.18.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 506 |
+
"backbone.qwen_model.model.visual.blocks.18.norm1.bias": "model-00001-of-00003.safetensors",
|
| 507 |
+
"backbone.qwen_model.model.visual.blocks.18.norm1.weight": "model-00001-of-00003.safetensors",
|
| 508 |
+
"backbone.qwen_model.model.visual.blocks.18.norm2.bias": "model-00001-of-00003.safetensors",
|
| 509 |
+
"backbone.qwen_model.model.visual.blocks.18.norm2.weight": "model-00001-of-00003.safetensors",
|
| 510 |
+
"backbone.qwen_model.model.visual.blocks.19.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 511 |
+
"backbone.qwen_model.model.visual.blocks.19.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 512 |
+
"backbone.qwen_model.model.visual.blocks.19.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 513 |
+
"backbone.qwen_model.model.visual.blocks.19.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 514 |
+
"backbone.qwen_model.model.visual.blocks.19.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 515 |
+
"backbone.qwen_model.model.visual.blocks.19.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 516 |
+
"backbone.qwen_model.model.visual.blocks.19.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 517 |
+
"backbone.qwen_model.model.visual.blocks.19.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 518 |
+
"backbone.qwen_model.model.visual.blocks.19.norm1.bias": "model-00001-of-00003.safetensors",
|
| 519 |
+
"backbone.qwen_model.model.visual.blocks.19.norm1.weight": "model-00001-of-00003.safetensors",
|
| 520 |
+
"backbone.qwen_model.model.visual.blocks.19.norm2.bias": "model-00001-of-00003.safetensors",
|
| 521 |
+
"backbone.qwen_model.model.visual.blocks.19.norm2.weight": "model-00001-of-00003.safetensors",
|
| 522 |
+
"backbone.qwen_model.model.visual.blocks.2.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 523 |
+
"backbone.qwen_model.model.visual.blocks.2.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 524 |
+
"backbone.qwen_model.model.visual.blocks.2.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 525 |
+
"backbone.qwen_model.model.visual.blocks.2.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 526 |
+
"backbone.qwen_model.model.visual.blocks.2.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 527 |
+
"backbone.qwen_model.model.visual.blocks.2.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 528 |
+
"backbone.qwen_model.model.visual.blocks.2.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 529 |
+
"backbone.qwen_model.model.visual.blocks.2.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 530 |
+
"backbone.qwen_model.model.visual.blocks.2.norm1.bias": "model-00001-of-00003.safetensors",
|
| 531 |
+
"backbone.qwen_model.model.visual.blocks.2.norm1.weight": "model-00001-of-00003.safetensors",
|
| 532 |
+
"backbone.qwen_model.model.visual.blocks.2.norm2.bias": "model-00001-of-00003.safetensors",
|
| 533 |
+
"backbone.qwen_model.model.visual.blocks.2.norm2.weight": "model-00001-of-00003.safetensors",
|
| 534 |
+
"backbone.qwen_model.model.visual.blocks.20.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 535 |
+
"backbone.qwen_model.model.visual.blocks.20.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 536 |
+
"backbone.qwen_model.model.visual.blocks.20.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 537 |
+
"backbone.qwen_model.model.visual.blocks.20.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 538 |
+
"backbone.qwen_model.model.visual.blocks.20.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 539 |
+
"backbone.qwen_model.model.visual.blocks.20.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 540 |
+
"backbone.qwen_model.model.visual.blocks.20.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 541 |
+
"backbone.qwen_model.model.visual.blocks.20.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 542 |
+
"backbone.qwen_model.model.visual.blocks.20.norm1.bias": "model-00001-of-00003.safetensors",
|
| 543 |
+
"backbone.qwen_model.model.visual.blocks.20.norm1.weight": "model-00001-of-00003.safetensors",
|
| 544 |
+
"backbone.qwen_model.model.visual.blocks.20.norm2.bias": "model-00001-of-00003.safetensors",
|
| 545 |
+
"backbone.qwen_model.model.visual.blocks.20.norm2.weight": "model-00001-of-00003.safetensors",
|
| 546 |
+
"backbone.qwen_model.model.visual.blocks.21.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 547 |
+
"backbone.qwen_model.model.visual.blocks.21.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 548 |
+
"backbone.qwen_model.model.visual.blocks.21.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 549 |
+
"backbone.qwen_model.model.visual.blocks.21.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 550 |
+
"backbone.qwen_model.model.visual.blocks.21.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 551 |
+
"backbone.qwen_model.model.visual.blocks.21.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 552 |
+
"backbone.qwen_model.model.visual.blocks.21.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 553 |
+
"backbone.qwen_model.model.visual.blocks.21.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 554 |
+
"backbone.qwen_model.model.visual.blocks.21.norm1.bias": "model-00001-of-00003.safetensors",
|
| 555 |
+
"backbone.qwen_model.model.visual.blocks.21.norm1.weight": "model-00001-of-00003.safetensors",
|
| 556 |
+
"backbone.qwen_model.model.visual.blocks.21.norm2.bias": "model-00001-of-00003.safetensors",
|
| 557 |
+
"backbone.qwen_model.model.visual.blocks.21.norm2.weight": "model-00001-of-00003.safetensors",
|
| 558 |
+
"backbone.qwen_model.model.visual.blocks.22.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 559 |
+
"backbone.qwen_model.model.visual.blocks.22.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 560 |
+
"backbone.qwen_model.model.visual.blocks.22.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 561 |
+
"backbone.qwen_model.model.visual.blocks.22.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 562 |
+
"backbone.qwen_model.model.visual.blocks.22.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 563 |
+
"backbone.qwen_model.model.visual.blocks.22.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 564 |
+
"backbone.qwen_model.model.visual.blocks.22.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 565 |
+
"backbone.qwen_model.model.visual.blocks.22.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 566 |
+
"backbone.qwen_model.model.visual.blocks.22.norm1.bias": "model-00001-of-00003.safetensors",
|
| 567 |
+
"backbone.qwen_model.model.visual.blocks.22.norm1.weight": "model-00001-of-00003.safetensors",
|
| 568 |
+
"backbone.qwen_model.model.visual.blocks.22.norm2.bias": "model-00001-of-00003.safetensors",
|
| 569 |
+
"backbone.qwen_model.model.visual.blocks.22.norm2.weight": "model-00001-of-00003.safetensors",
|
| 570 |
+
"backbone.qwen_model.model.visual.blocks.23.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 571 |
+
"backbone.qwen_model.model.visual.blocks.23.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 572 |
+
"backbone.qwen_model.model.visual.blocks.23.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 573 |
+
"backbone.qwen_model.model.visual.blocks.23.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 574 |
+
"backbone.qwen_model.model.visual.blocks.23.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 575 |
+
"backbone.qwen_model.model.visual.blocks.23.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 576 |
+
"backbone.qwen_model.model.visual.blocks.23.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 577 |
+
"backbone.qwen_model.model.visual.blocks.23.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 578 |
+
"backbone.qwen_model.model.visual.blocks.23.norm1.bias": "model-00001-of-00003.safetensors",
|
| 579 |
+
"backbone.qwen_model.model.visual.blocks.23.norm1.weight": "model-00001-of-00003.safetensors",
|
| 580 |
+
"backbone.qwen_model.model.visual.blocks.23.norm2.bias": "model-00001-of-00003.safetensors",
|
| 581 |
+
"backbone.qwen_model.model.visual.blocks.23.norm2.weight": "model-00001-of-00003.safetensors",
|
| 582 |
+
"backbone.qwen_model.model.visual.blocks.24.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 583 |
+
"backbone.qwen_model.model.visual.blocks.24.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 584 |
+
"backbone.qwen_model.model.visual.blocks.24.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 585 |
+
"backbone.qwen_model.model.visual.blocks.24.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 586 |
+
"backbone.qwen_model.model.visual.blocks.24.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 587 |
+
"backbone.qwen_model.model.visual.blocks.24.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 588 |
+
"backbone.qwen_model.model.visual.blocks.24.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 589 |
+
"backbone.qwen_model.model.visual.blocks.24.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 590 |
+
"backbone.qwen_model.model.visual.blocks.24.norm1.bias": "model-00001-of-00003.safetensors",
|
| 591 |
+
"backbone.qwen_model.model.visual.blocks.24.norm1.weight": "model-00001-of-00003.safetensors",
|
| 592 |
+
"backbone.qwen_model.model.visual.blocks.24.norm2.bias": "model-00001-of-00003.safetensors",
|
| 593 |
+
"backbone.qwen_model.model.visual.blocks.24.norm2.weight": "model-00001-of-00003.safetensors",
|
| 594 |
+
"backbone.qwen_model.model.visual.blocks.25.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 595 |
+
"backbone.qwen_model.model.visual.blocks.25.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 596 |
+
"backbone.qwen_model.model.visual.blocks.25.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 597 |
+
"backbone.qwen_model.model.visual.blocks.25.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 598 |
+
"backbone.qwen_model.model.visual.blocks.25.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 599 |
+
"backbone.qwen_model.model.visual.blocks.25.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 600 |
+
"backbone.qwen_model.model.visual.blocks.25.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 601 |
+
"backbone.qwen_model.model.visual.blocks.25.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 602 |
+
"backbone.qwen_model.model.visual.blocks.25.norm1.bias": "model-00001-of-00003.safetensors",
|
| 603 |
+
"backbone.qwen_model.model.visual.blocks.25.norm1.weight": "model-00001-of-00003.safetensors",
|
| 604 |
+
"backbone.qwen_model.model.visual.blocks.25.norm2.bias": "model-00001-of-00003.safetensors",
|
| 605 |
+
"backbone.qwen_model.model.visual.blocks.25.norm2.weight": "model-00001-of-00003.safetensors",
|
| 606 |
+
"backbone.qwen_model.model.visual.blocks.26.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 607 |
+
"backbone.qwen_model.model.visual.blocks.26.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 608 |
+
"backbone.qwen_model.model.visual.blocks.26.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 609 |
+
"backbone.qwen_model.model.visual.blocks.26.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 610 |
+
"backbone.qwen_model.model.visual.blocks.26.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 611 |
+
"backbone.qwen_model.model.visual.blocks.26.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 612 |
+
"backbone.qwen_model.model.visual.blocks.26.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 613 |
+
"backbone.qwen_model.model.visual.blocks.26.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 614 |
+
"backbone.qwen_model.model.visual.blocks.26.norm1.bias": "model-00001-of-00003.safetensors",
|
| 615 |
+
"backbone.qwen_model.model.visual.blocks.26.norm1.weight": "model-00001-of-00003.safetensors",
|
| 616 |
+
"backbone.qwen_model.model.visual.blocks.26.norm2.bias": "model-00001-of-00003.safetensors",
|
| 617 |
+
"backbone.qwen_model.model.visual.blocks.26.norm2.weight": "model-00001-of-00003.safetensors",
|
| 618 |
+
"backbone.qwen_model.model.visual.blocks.3.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 619 |
+
"backbone.qwen_model.model.visual.blocks.3.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 620 |
+
"backbone.qwen_model.model.visual.blocks.3.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 621 |
+
"backbone.qwen_model.model.visual.blocks.3.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 622 |
+
"backbone.qwen_model.model.visual.blocks.3.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 623 |
+
"backbone.qwen_model.model.visual.blocks.3.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 624 |
+
"backbone.qwen_model.model.visual.blocks.3.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 625 |
+
"backbone.qwen_model.model.visual.blocks.3.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 626 |
+
"backbone.qwen_model.model.visual.blocks.3.norm1.bias": "model-00001-of-00003.safetensors",
|
| 627 |
+
"backbone.qwen_model.model.visual.blocks.3.norm1.weight": "model-00001-of-00003.safetensors",
|
| 628 |
+
"backbone.qwen_model.model.visual.blocks.3.norm2.bias": "model-00001-of-00003.safetensors",
|
| 629 |
+
"backbone.qwen_model.model.visual.blocks.3.norm2.weight": "model-00001-of-00003.safetensors",
|
| 630 |
+
"backbone.qwen_model.model.visual.blocks.4.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 631 |
+
"backbone.qwen_model.model.visual.blocks.4.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 632 |
+
"backbone.qwen_model.model.visual.blocks.4.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 633 |
+
"backbone.qwen_model.model.visual.blocks.4.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 634 |
+
"backbone.qwen_model.model.visual.blocks.4.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 635 |
+
"backbone.qwen_model.model.visual.blocks.4.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 636 |
+
"backbone.qwen_model.model.visual.blocks.4.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 637 |
+
"backbone.qwen_model.model.visual.blocks.4.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 638 |
+
"backbone.qwen_model.model.visual.blocks.4.norm1.bias": "model-00001-of-00003.safetensors",
|
| 639 |
+
"backbone.qwen_model.model.visual.blocks.4.norm1.weight": "model-00001-of-00003.safetensors",
|
| 640 |
+
"backbone.qwen_model.model.visual.blocks.4.norm2.bias": "model-00001-of-00003.safetensors",
|
| 641 |
+
"backbone.qwen_model.model.visual.blocks.4.norm2.weight": "model-00001-of-00003.safetensors",
|
| 642 |
+
"backbone.qwen_model.model.visual.blocks.5.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 643 |
+
"backbone.qwen_model.model.visual.blocks.5.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 644 |
+
"backbone.qwen_model.model.visual.blocks.5.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 645 |
+
"backbone.qwen_model.model.visual.blocks.5.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 646 |
+
"backbone.qwen_model.model.visual.blocks.5.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 647 |
+
"backbone.qwen_model.model.visual.blocks.5.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 648 |
+
"backbone.qwen_model.model.visual.blocks.5.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 649 |
+
"backbone.qwen_model.model.visual.blocks.5.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 650 |
+
"backbone.qwen_model.model.visual.blocks.5.norm1.bias": "model-00001-of-00003.safetensors",
|
| 651 |
+
"backbone.qwen_model.model.visual.blocks.5.norm1.weight": "model-00001-of-00003.safetensors",
|
| 652 |
+
"backbone.qwen_model.model.visual.blocks.5.norm2.bias": "model-00001-of-00003.safetensors",
|
| 653 |
+
"backbone.qwen_model.model.visual.blocks.5.norm2.weight": "model-00001-of-00003.safetensors",
|
| 654 |
+
"backbone.qwen_model.model.visual.blocks.6.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 655 |
+
"backbone.qwen_model.model.visual.blocks.6.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 656 |
+
"backbone.qwen_model.model.visual.blocks.6.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 657 |
+
"backbone.qwen_model.model.visual.blocks.6.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 658 |
+
"backbone.qwen_model.model.visual.blocks.6.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 659 |
+
"backbone.qwen_model.model.visual.blocks.6.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 660 |
+
"backbone.qwen_model.model.visual.blocks.6.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 661 |
+
"backbone.qwen_model.model.visual.blocks.6.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 662 |
+
"backbone.qwen_model.model.visual.blocks.6.norm1.bias": "model-00001-of-00003.safetensors",
|
| 663 |
+
"backbone.qwen_model.model.visual.blocks.6.norm1.weight": "model-00001-of-00003.safetensors",
|
| 664 |
+
"backbone.qwen_model.model.visual.blocks.6.norm2.bias": "model-00001-of-00003.safetensors",
|
| 665 |
+
"backbone.qwen_model.model.visual.blocks.6.norm2.weight": "model-00001-of-00003.safetensors",
|
| 666 |
+
"backbone.qwen_model.model.visual.blocks.7.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 667 |
+
"backbone.qwen_model.model.visual.blocks.7.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 668 |
+
"backbone.qwen_model.model.visual.blocks.7.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 669 |
+
"backbone.qwen_model.model.visual.blocks.7.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 670 |
+
"backbone.qwen_model.model.visual.blocks.7.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 671 |
+
"backbone.qwen_model.model.visual.blocks.7.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 672 |
+
"backbone.qwen_model.model.visual.blocks.7.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 673 |
+
"backbone.qwen_model.model.visual.blocks.7.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 674 |
+
"backbone.qwen_model.model.visual.blocks.7.norm1.bias": "model-00001-of-00003.safetensors",
|
| 675 |
+
"backbone.qwen_model.model.visual.blocks.7.norm1.weight": "model-00001-of-00003.safetensors",
|
| 676 |
+
"backbone.qwen_model.model.visual.blocks.7.norm2.bias": "model-00001-of-00003.safetensors",
|
| 677 |
+
"backbone.qwen_model.model.visual.blocks.7.norm2.weight": "model-00001-of-00003.safetensors",
|
| 678 |
+
"backbone.qwen_model.model.visual.blocks.8.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 679 |
+
"backbone.qwen_model.model.visual.blocks.8.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 680 |
+
"backbone.qwen_model.model.visual.blocks.8.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 681 |
+
"backbone.qwen_model.model.visual.blocks.8.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 682 |
+
"backbone.qwen_model.model.visual.blocks.8.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 683 |
+
"backbone.qwen_model.model.visual.blocks.8.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 684 |
+
"backbone.qwen_model.model.visual.blocks.8.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 685 |
+
"backbone.qwen_model.model.visual.blocks.8.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 686 |
+
"backbone.qwen_model.model.visual.blocks.8.norm1.bias": "model-00001-of-00003.safetensors",
|
| 687 |
+
"backbone.qwen_model.model.visual.blocks.8.norm1.weight": "model-00001-of-00003.safetensors",
|
| 688 |
+
"backbone.qwen_model.model.visual.blocks.8.norm2.bias": "model-00001-of-00003.safetensors",
|
| 689 |
+
"backbone.qwen_model.model.visual.blocks.8.norm2.weight": "model-00001-of-00003.safetensors",
|
| 690 |
+
"backbone.qwen_model.model.visual.blocks.9.attn.proj.bias": "model-00001-of-00003.safetensors",
|
| 691 |
+
"backbone.qwen_model.model.visual.blocks.9.attn.proj.weight": "model-00001-of-00003.safetensors",
|
| 692 |
+
"backbone.qwen_model.model.visual.blocks.9.attn.qkv.bias": "model-00001-of-00003.safetensors",
|
| 693 |
+
"backbone.qwen_model.model.visual.blocks.9.attn.qkv.weight": "model-00001-of-00003.safetensors",
|
| 694 |
+
"backbone.qwen_model.model.visual.blocks.9.mlp.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 695 |
+
"backbone.qwen_model.model.visual.blocks.9.mlp.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 696 |
+
"backbone.qwen_model.model.visual.blocks.9.mlp.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 697 |
+
"backbone.qwen_model.model.visual.blocks.9.mlp.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 698 |
+
"backbone.qwen_model.model.visual.blocks.9.norm1.bias": "model-00001-of-00003.safetensors",
|
| 699 |
+
"backbone.qwen_model.model.visual.blocks.9.norm1.weight": "model-00001-of-00003.safetensors",
|
| 700 |
+
"backbone.qwen_model.model.visual.blocks.9.norm2.bias": "model-00001-of-00003.safetensors",
|
| 701 |
+
"backbone.qwen_model.model.visual.blocks.9.norm2.weight": "model-00001-of-00003.safetensors",
|
| 702 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.0.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 703 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.0.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 704 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.0.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 705 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.0.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 706 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.0.norm.bias": "model-00001-of-00003.safetensors",
|
| 707 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.0.norm.weight": "model-00001-of-00003.safetensors",
|
| 708 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.1.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 709 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.1.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 710 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.1.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 711 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.1.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 712 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.1.norm.bias": "model-00001-of-00003.safetensors",
|
| 713 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.1.norm.weight": "model-00001-of-00003.safetensors",
|
| 714 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.2.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 715 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.2.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 716 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.2.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 717 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.2.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 718 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.2.norm.bias": "model-00001-of-00003.safetensors",
|
| 719 |
+
"backbone.qwen_model.model.visual.deepstack_merger_list.2.norm.weight": "model-00001-of-00003.safetensors",
|
| 720 |
+
"backbone.qwen_model.model.visual.merger.linear_fc1.bias": "model-00001-of-00003.safetensors",
|
| 721 |
+
"backbone.qwen_model.model.visual.merger.linear_fc1.weight": "model-00001-of-00003.safetensors",
|
| 722 |
+
"backbone.qwen_model.model.visual.merger.linear_fc2.bias": "model-00001-of-00003.safetensors",
|
| 723 |
+
"backbone.qwen_model.model.visual.merger.linear_fc2.weight": "model-00001-of-00003.safetensors",
|
| 724 |
+
"backbone.qwen_model.model.visual.merger.norm.bias": "model-00001-of-00003.safetensors",
|
| 725 |
+
"backbone.qwen_model.model.visual.merger.norm.weight": "model-00001-of-00003.safetensors",
|
| 726 |
+
"backbone.qwen_model.model.visual.patch_embed.proj.bias": "model-00001-of-00003.safetensors",
|
| 727 |
+
"backbone.qwen_model.model.visual.patch_embed.proj.weight": "model-00001-of-00003.safetensors",
|
| 728 |
+
"backbone.qwen_model.model.visual.pos_embed.weight": "model-00001-of-00003.safetensors"
|
| 729 |
+
}
|
| 730 |
+
}
|
processor/embodiment_id.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"general_embodiment": 0,
|
| 3 |
+
"fractal20220817_data": 1,
|
| 4 |
+
"kuka": 2,
|
| 5 |
+
"bridge_orig": 3,
|
| 6 |
+
"taco_play": 4,
|
| 7 |
+
"jaco_play": 5,
|
| 8 |
+
"berkeley_cable_routing": 6,
|
| 9 |
+
"roboturk": 7,
|
| 10 |
+
"viola": 8,
|
| 11 |
+
"berkeley_autolab_ur5": 9,
|
| 12 |
+
"toto": 10,
|
| 13 |
+
"language_table": 11,
|
| 14 |
+
"stanford_hydra_dataset_converted_externally_to_rlds": 12,
|
| 15 |
+
"austin_buds_dataset_converted_externally_to_rlds": 13,
|
| 16 |
+
"nyu_franka_play_dataset_converted_externally_to_rlds": 14,
|
| 17 |
+
"furniture_bench_dataset_converted_externally_to_rlds": 15,
|
| 18 |
+
"ucsd_kitchen_dataset_converted_externally_to_rlds": 16,
|
| 19 |
+
"austin_sailor_dataset_converted_externally_to_rlds": 17,
|
| 20 |
+
"austin_sirius_dataset_converted_externally_to_rlds": 18,
|
| 21 |
+
"dlr_edan_shared_control_converted_externally_to_rlds": 19,
|
| 22 |
+
"iamlab_cmu_pickup_insert_converted_externally_to_rlds": 20,
|
| 23 |
+
"utaustin_mutex": 21,
|
| 24 |
+
"berkeley_fanuc_manipulation": 22,
|
| 25 |
+
"cmu_stretch": 23,
|
| 26 |
+
"bc_z": 24,
|
| 27 |
+
"fmb_dataset": 25,
|
| 28 |
+
"dobbe": 26,
|
| 29 |
+
"droid": 27,
|
| 30 |
+
"agibot_dexhand": 28,
|
| 31 |
+
"agibot_gripper": 29,
|
| 32 |
+
"galaxea": 30,
|
| 33 |
+
"humanoid_everyday_g1": 31,
|
| 34 |
+
"humanoid_everyday_h1": 32,
|
| 35 |
+
"action_net": 33,
|
| 36 |
+
"neural_gr1": 34,
|
| 37 |
+
"new_embodiment": 35
|
| 38 |
+
}
|
processor/processor_config.json
ADDED
|
@@ -0,0 +1,2980 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"processor_class": "RLDXProcessor",
|
| 3 |
+
"processor_kwargs": {
|
| 4 |
+
"modality_configs": {
|
| 5 |
+
"agibot_gripper": {
|
| 6 |
+
"video": {
|
| 7 |
+
"delta_indices": [
|
| 8 |
+
0
|
| 9 |
+
],
|
| 10 |
+
"modality_keys": [
|
| 11 |
+
"primary",
|
| 12 |
+
"wrist_left",
|
| 13 |
+
"wrist_right"
|
| 14 |
+
],
|
| 15 |
+
"sin_cos_embedding_keys": null,
|
| 16 |
+
"mean_std_embedding_keys": null,
|
| 17 |
+
"action_configs": null
|
| 18 |
+
},
|
| 19 |
+
"state": {
|
| 20 |
+
"delta_indices": [
|
| 21 |
+
0
|
| 22 |
+
],
|
| 23 |
+
"modality_keys": [
|
| 24 |
+
"state"
|
| 25 |
+
],
|
| 26 |
+
"sin_cos_embedding_keys": null,
|
| 27 |
+
"mean_std_embedding_keys": null,
|
| 28 |
+
"action_configs": null
|
| 29 |
+
},
|
| 30 |
+
"action": {
|
| 31 |
+
"delta_indices": [
|
| 32 |
+
0,
|
| 33 |
+
1,
|
| 34 |
+
2,
|
| 35 |
+
3,
|
| 36 |
+
4,
|
| 37 |
+
5,
|
| 38 |
+
6,
|
| 39 |
+
7,
|
| 40 |
+
8,
|
| 41 |
+
9,
|
| 42 |
+
10,
|
| 43 |
+
11,
|
| 44 |
+
12,
|
| 45 |
+
13,
|
| 46 |
+
14,
|
| 47 |
+
15
|
| 48 |
+
],
|
| 49 |
+
"modality_keys": [
|
| 50 |
+
"action"
|
| 51 |
+
],
|
| 52 |
+
"sin_cos_embedding_keys": null,
|
| 53 |
+
"mean_std_embedding_keys": null,
|
| 54 |
+
"action_configs": [
|
| 55 |
+
{
|
| 56 |
+
"rep": "ABSOLUTE",
|
| 57 |
+
"type": "NON_EEF",
|
| 58 |
+
"format": "DEFAULT",
|
| 59 |
+
"state_key": null
|
| 60 |
+
}
|
| 61 |
+
]
|
| 62 |
+
},
|
| 63 |
+
"language": {
|
| 64 |
+
"delta_indices": [
|
| 65 |
+
0
|
| 66 |
+
],
|
| 67 |
+
"modality_keys": [
|
| 68 |
+
"annotation.human.action.task_description"
|
| 69 |
+
],
|
| 70 |
+
"sin_cos_embedding_keys": null,
|
| 71 |
+
"mean_std_embedding_keys": null,
|
| 72 |
+
"action_configs": null
|
| 73 |
+
}
|
| 74 |
+
},
|
| 75 |
+
"dlr_edan_shared_control_converted_externally_to_rlds": {
|
| 76 |
+
"video": {
|
| 77 |
+
"delta_indices": [
|
| 78 |
+
0
|
| 79 |
+
],
|
| 80 |
+
"modality_keys": [
|
| 81 |
+
"primary"
|
| 82 |
+
],
|
| 83 |
+
"sin_cos_embedding_keys": null,
|
| 84 |
+
"mean_std_embedding_keys": null,
|
| 85 |
+
"action_configs": null
|
| 86 |
+
},
|
| 87 |
+
"state": {
|
| 88 |
+
"delta_indices": [
|
| 89 |
+
0
|
| 90 |
+
],
|
| 91 |
+
"modality_keys": [
|
| 92 |
+
"end_effector_position",
|
| 93 |
+
"end_effector_rotation",
|
| 94 |
+
"gripper_position"
|
| 95 |
+
],
|
| 96 |
+
"sin_cos_embedding_keys": null,
|
| 97 |
+
"mean_std_embedding_keys": null,
|
| 98 |
+
"action_configs": null
|
| 99 |
+
},
|
| 100 |
+
"action": {
|
| 101 |
+
"delta_indices": [
|
| 102 |
+
0,
|
| 103 |
+
1,
|
| 104 |
+
2,
|
| 105 |
+
3,
|
| 106 |
+
4,
|
| 107 |
+
5,
|
| 108 |
+
6,
|
| 109 |
+
7,
|
| 110 |
+
8,
|
| 111 |
+
9,
|
| 112 |
+
10,
|
| 113 |
+
11,
|
| 114 |
+
12,
|
| 115 |
+
13,
|
| 116 |
+
14,
|
| 117 |
+
15
|
| 118 |
+
],
|
| 119 |
+
"modality_keys": [
|
| 120 |
+
"end_effector_position",
|
| 121 |
+
"end_effector_rotation",
|
| 122 |
+
"gripper_close"
|
| 123 |
+
],
|
| 124 |
+
"sin_cos_embedding_keys": null,
|
| 125 |
+
"mean_std_embedding_keys": null,
|
| 126 |
+
"action_configs": [
|
| 127 |
+
{
|
| 128 |
+
"rep": "DELTA",
|
| 129 |
+
"type": "EEF",
|
| 130 |
+
"format": "DEFAULT",
|
| 131 |
+
"state_key": null
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"rep": "DELTA",
|
| 135 |
+
"type": "EEF",
|
| 136 |
+
"format": "DEFAULT",
|
| 137 |
+
"state_key": null
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"rep": "ABSOLUTE",
|
| 141 |
+
"type": "NON_EEF",
|
| 142 |
+
"format": "DEFAULT",
|
| 143 |
+
"state_key": null
|
| 144 |
+
}
|
| 145 |
+
]
|
| 146 |
+
},
|
| 147 |
+
"language": {
|
| 148 |
+
"delta_indices": [
|
| 149 |
+
0
|
| 150 |
+
],
|
| 151 |
+
"modality_keys": [
|
| 152 |
+
"annotation.human.action.task_description"
|
| 153 |
+
],
|
| 154 |
+
"sin_cos_embedding_keys": null,
|
| 155 |
+
"mean_std_embedding_keys": null,
|
| 156 |
+
"action_configs": null
|
| 157 |
+
}
|
| 158 |
+
},
|
| 159 |
+
"humanoid_everyday_g1": {
|
| 160 |
+
"video": {
|
| 161 |
+
"delta_indices": [
|
| 162 |
+
0
|
| 163 |
+
],
|
| 164 |
+
"modality_keys": [
|
| 165 |
+
"egocentric_resized"
|
| 166 |
+
],
|
| 167 |
+
"sin_cos_embedding_keys": null,
|
| 168 |
+
"mean_std_embedding_keys": null,
|
| 169 |
+
"action_configs": null
|
| 170 |
+
},
|
| 171 |
+
"state": {
|
| 172 |
+
"delta_indices": [
|
| 173 |
+
0
|
| 174 |
+
],
|
| 175 |
+
"modality_keys": [
|
| 176 |
+
"left_arm",
|
| 177 |
+
"left_hand",
|
| 178 |
+
"right_arm",
|
| 179 |
+
"right_hand"
|
| 180 |
+
],
|
| 181 |
+
"sin_cos_embedding_keys": null,
|
| 182 |
+
"mean_std_embedding_keys": null,
|
| 183 |
+
"action_configs": null
|
| 184 |
+
},
|
| 185 |
+
"action": {
|
| 186 |
+
"delta_indices": [
|
| 187 |
+
0,
|
| 188 |
+
1,
|
| 189 |
+
2,
|
| 190 |
+
3,
|
| 191 |
+
4,
|
| 192 |
+
5,
|
| 193 |
+
6,
|
| 194 |
+
7,
|
| 195 |
+
8,
|
| 196 |
+
9,
|
| 197 |
+
10,
|
| 198 |
+
11,
|
| 199 |
+
12,
|
| 200 |
+
13,
|
| 201 |
+
14,
|
| 202 |
+
15
|
| 203 |
+
],
|
| 204 |
+
"modality_keys": [
|
| 205 |
+
"left_arm",
|
| 206 |
+
"left_hand",
|
| 207 |
+
"right_arm",
|
| 208 |
+
"right_hand"
|
| 209 |
+
],
|
| 210 |
+
"sin_cos_embedding_keys": null,
|
| 211 |
+
"mean_std_embedding_keys": null,
|
| 212 |
+
"action_configs": [
|
| 213 |
+
{
|
| 214 |
+
"rep": "ABSOLUTE",
|
| 215 |
+
"type": "NON_EEF",
|
| 216 |
+
"format": "DEFAULT",
|
| 217 |
+
"state_key": null
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"rep": "ABSOLUTE",
|
| 221 |
+
"type": "NON_EEF",
|
| 222 |
+
"format": "DEFAULT",
|
| 223 |
+
"state_key": null
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"rep": "ABSOLUTE",
|
| 227 |
+
"type": "NON_EEF",
|
| 228 |
+
"format": "DEFAULT",
|
| 229 |
+
"state_key": null
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"rep": "ABSOLUTE",
|
| 233 |
+
"type": "NON_EEF",
|
| 234 |
+
"format": "DEFAULT",
|
| 235 |
+
"state_key": null
|
| 236 |
+
}
|
| 237 |
+
]
|
| 238 |
+
},
|
| 239 |
+
"language": {
|
| 240 |
+
"delta_indices": [
|
| 241 |
+
0
|
| 242 |
+
],
|
| 243 |
+
"modality_keys": [
|
| 244 |
+
"annotation.human.action.task_description"
|
| 245 |
+
],
|
| 246 |
+
"sin_cos_embedding_keys": null,
|
| 247 |
+
"mean_std_embedding_keys": null,
|
| 248 |
+
"action_configs": null
|
| 249 |
+
}
|
| 250 |
+
},
|
| 251 |
+
"austin_sirius_dataset_converted_externally_to_rlds": {
|
| 252 |
+
"video": {
|
| 253 |
+
"delta_indices": [
|
| 254 |
+
0
|
| 255 |
+
],
|
| 256 |
+
"modality_keys": [
|
| 257 |
+
"primary",
|
| 258 |
+
"wrist"
|
| 259 |
+
],
|
| 260 |
+
"sin_cos_embedding_keys": null,
|
| 261 |
+
"mean_std_embedding_keys": null,
|
| 262 |
+
"action_configs": null
|
| 263 |
+
},
|
| 264 |
+
"state": {
|
| 265 |
+
"delta_indices": [
|
| 266 |
+
0
|
| 267 |
+
],
|
| 268 |
+
"modality_keys": [
|
| 269 |
+
"end_effector_position",
|
| 270 |
+
"end_effector_rotation",
|
| 271 |
+
"gripper_position"
|
| 272 |
+
],
|
| 273 |
+
"sin_cos_embedding_keys": null,
|
| 274 |
+
"mean_std_embedding_keys": null,
|
| 275 |
+
"action_configs": null
|
| 276 |
+
},
|
| 277 |
+
"action": {
|
| 278 |
+
"delta_indices": [
|
| 279 |
+
0,
|
| 280 |
+
1,
|
| 281 |
+
2,
|
| 282 |
+
3,
|
| 283 |
+
4,
|
| 284 |
+
5,
|
| 285 |
+
6,
|
| 286 |
+
7,
|
| 287 |
+
8,
|
| 288 |
+
9,
|
| 289 |
+
10,
|
| 290 |
+
11,
|
| 291 |
+
12,
|
| 292 |
+
13,
|
| 293 |
+
14,
|
| 294 |
+
15
|
| 295 |
+
],
|
| 296 |
+
"modality_keys": [
|
| 297 |
+
"end_effector_position",
|
| 298 |
+
"end_effector_rotation",
|
| 299 |
+
"gripper_close"
|
| 300 |
+
],
|
| 301 |
+
"sin_cos_embedding_keys": null,
|
| 302 |
+
"mean_std_embedding_keys": null,
|
| 303 |
+
"action_configs": [
|
| 304 |
+
{
|
| 305 |
+
"rep": "DELTA",
|
| 306 |
+
"type": "EEF",
|
| 307 |
+
"format": "DEFAULT",
|
| 308 |
+
"state_key": null
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"rep": "DELTA",
|
| 312 |
+
"type": "EEF",
|
| 313 |
+
"format": "DEFAULT",
|
| 314 |
+
"state_key": null
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"rep": "ABSOLUTE",
|
| 318 |
+
"type": "NON_EEF",
|
| 319 |
+
"format": "DEFAULT",
|
| 320 |
+
"state_key": null
|
| 321 |
+
}
|
| 322 |
+
]
|
| 323 |
+
},
|
| 324 |
+
"language": {
|
| 325 |
+
"delta_indices": [
|
| 326 |
+
0
|
| 327 |
+
],
|
| 328 |
+
"modality_keys": [
|
| 329 |
+
"annotation.human.action.task_description"
|
| 330 |
+
],
|
| 331 |
+
"sin_cos_embedding_keys": null,
|
| 332 |
+
"mean_std_embedding_keys": null,
|
| 333 |
+
"action_configs": null
|
| 334 |
+
}
|
| 335 |
+
},
|
| 336 |
+
"jaco_play": {
|
| 337 |
+
"video": {
|
| 338 |
+
"delta_indices": [
|
| 339 |
+
0
|
| 340 |
+
],
|
| 341 |
+
"modality_keys": [
|
| 342 |
+
"primary",
|
| 343 |
+
"wrist"
|
| 344 |
+
],
|
| 345 |
+
"sin_cos_embedding_keys": null,
|
| 346 |
+
"mean_std_embedding_keys": null,
|
| 347 |
+
"action_configs": null
|
| 348 |
+
},
|
| 349 |
+
"state": {
|
| 350 |
+
"delta_indices": [
|
| 351 |
+
0
|
| 352 |
+
],
|
| 353 |
+
"modality_keys": [
|
| 354 |
+
"end_effector_position",
|
| 355 |
+
"end_effector_rotation",
|
| 356 |
+
"gripper_position"
|
| 357 |
+
],
|
| 358 |
+
"sin_cos_embedding_keys": null,
|
| 359 |
+
"mean_std_embedding_keys": null,
|
| 360 |
+
"action_configs": null
|
| 361 |
+
},
|
| 362 |
+
"action": {
|
| 363 |
+
"delta_indices": [
|
| 364 |
+
0,
|
| 365 |
+
1,
|
| 366 |
+
2,
|
| 367 |
+
3,
|
| 368 |
+
4,
|
| 369 |
+
5,
|
| 370 |
+
6,
|
| 371 |
+
7,
|
| 372 |
+
8,
|
| 373 |
+
9,
|
| 374 |
+
10,
|
| 375 |
+
11,
|
| 376 |
+
12,
|
| 377 |
+
13,
|
| 378 |
+
14,
|
| 379 |
+
15
|
| 380 |
+
],
|
| 381 |
+
"modality_keys": [
|
| 382 |
+
"end_effector_position",
|
| 383 |
+
"end_effector_rotation",
|
| 384 |
+
"gripper_close"
|
| 385 |
+
],
|
| 386 |
+
"sin_cos_embedding_keys": null,
|
| 387 |
+
"mean_std_embedding_keys": null,
|
| 388 |
+
"action_configs": [
|
| 389 |
+
{
|
| 390 |
+
"rep": "DELTA",
|
| 391 |
+
"type": "EEF",
|
| 392 |
+
"format": "DEFAULT",
|
| 393 |
+
"state_key": null
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"rep": "DELTA",
|
| 397 |
+
"type": "EEF",
|
| 398 |
+
"format": "DEFAULT",
|
| 399 |
+
"state_key": null
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"rep": "ABSOLUTE",
|
| 403 |
+
"type": "NON_EEF",
|
| 404 |
+
"format": "DEFAULT",
|
| 405 |
+
"state_key": null
|
| 406 |
+
}
|
| 407 |
+
]
|
| 408 |
+
},
|
| 409 |
+
"language": {
|
| 410 |
+
"delta_indices": [
|
| 411 |
+
0
|
| 412 |
+
],
|
| 413 |
+
"modality_keys": [
|
| 414 |
+
"annotation.human.action.task_description"
|
| 415 |
+
],
|
| 416 |
+
"sin_cos_embedding_keys": null,
|
| 417 |
+
"mean_std_embedding_keys": null,
|
| 418 |
+
"action_configs": null
|
| 419 |
+
}
|
| 420 |
+
},
|
| 421 |
+
"cmu_stretch": {
|
| 422 |
+
"video": {
|
| 423 |
+
"delta_indices": [
|
| 424 |
+
0
|
| 425 |
+
],
|
| 426 |
+
"modality_keys": [
|
| 427 |
+
"primary"
|
| 428 |
+
],
|
| 429 |
+
"sin_cos_embedding_keys": null,
|
| 430 |
+
"mean_std_embedding_keys": null,
|
| 431 |
+
"action_configs": null
|
| 432 |
+
},
|
| 433 |
+
"state": {
|
| 434 |
+
"delta_indices": [
|
| 435 |
+
0
|
| 436 |
+
],
|
| 437 |
+
"modality_keys": [
|
| 438 |
+
"end_effector_position",
|
| 439 |
+
"end_effector_rotation",
|
| 440 |
+
"gripper_position"
|
| 441 |
+
],
|
| 442 |
+
"sin_cos_embedding_keys": null,
|
| 443 |
+
"mean_std_embedding_keys": null,
|
| 444 |
+
"action_configs": null
|
| 445 |
+
},
|
| 446 |
+
"action": {
|
| 447 |
+
"delta_indices": [
|
| 448 |
+
0,
|
| 449 |
+
1,
|
| 450 |
+
2,
|
| 451 |
+
3,
|
| 452 |
+
4,
|
| 453 |
+
5,
|
| 454 |
+
6,
|
| 455 |
+
7,
|
| 456 |
+
8,
|
| 457 |
+
9,
|
| 458 |
+
10,
|
| 459 |
+
11,
|
| 460 |
+
12,
|
| 461 |
+
13,
|
| 462 |
+
14,
|
| 463 |
+
15
|
| 464 |
+
],
|
| 465 |
+
"modality_keys": [
|
| 466 |
+
"end_effector_position",
|
| 467 |
+
"end_effector_rotation",
|
| 468 |
+
"gripper_close"
|
| 469 |
+
],
|
| 470 |
+
"sin_cos_embedding_keys": null,
|
| 471 |
+
"mean_std_embedding_keys": null,
|
| 472 |
+
"action_configs": [
|
| 473 |
+
{
|
| 474 |
+
"rep": "DELTA",
|
| 475 |
+
"type": "EEF",
|
| 476 |
+
"format": "DEFAULT",
|
| 477 |
+
"state_key": null
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"rep": "DELTA",
|
| 481 |
+
"type": "EEF",
|
| 482 |
+
"format": "DEFAULT",
|
| 483 |
+
"state_key": null
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"rep": "ABSOLUTE",
|
| 487 |
+
"type": "NON_EEF",
|
| 488 |
+
"format": "DEFAULT",
|
| 489 |
+
"state_key": null
|
| 490 |
+
}
|
| 491 |
+
]
|
| 492 |
+
},
|
| 493 |
+
"language": {
|
| 494 |
+
"delta_indices": [
|
| 495 |
+
0
|
| 496 |
+
],
|
| 497 |
+
"modality_keys": [
|
| 498 |
+
"annotation.human.action.task_description"
|
| 499 |
+
],
|
| 500 |
+
"sin_cos_embedding_keys": null,
|
| 501 |
+
"mean_std_embedding_keys": null,
|
| 502 |
+
"action_configs": null
|
| 503 |
+
}
|
| 504 |
+
},
|
| 505 |
+
"fractal20220817_data": {
|
| 506 |
+
"video": {
|
| 507 |
+
"delta_indices": [
|
| 508 |
+
0
|
| 509 |
+
],
|
| 510 |
+
"modality_keys": [
|
| 511 |
+
"primary"
|
| 512 |
+
],
|
| 513 |
+
"sin_cos_embedding_keys": null,
|
| 514 |
+
"mean_std_embedding_keys": null,
|
| 515 |
+
"action_configs": null
|
| 516 |
+
},
|
| 517 |
+
"state": {
|
| 518 |
+
"delta_indices": [
|
| 519 |
+
0
|
| 520 |
+
],
|
| 521 |
+
"modality_keys": [
|
| 522 |
+
"end_effector_position",
|
| 523 |
+
"end_effector_rotation",
|
| 524 |
+
"gripper_position"
|
| 525 |
+
],
|
| 526 |
+
"sin_cos_embedding_keys": null,
|
| 527 |
+
"mean_std_embedding_keys": null,
|
| 528 |
+
"action_configs": null
|
| 529 |
+
},
|
| 530 |
+
"action": {
|
| 531 |
+
"delta_indices": [
|
| 532 |
+
0,
|
| 533 |
+
1,
|
| 534 |
+
2,
|
| 535 |
+
3,
|
| 536 |
+
4,
|
| 537 |
+
5,
|
| 538 |
+
6,
|
| 539 |
+
7,
|
| 540 |
+
8,
|
| 541 |
+
9,
|
| 542 |
+
10,
|
| 543 |
+
11,
|
| 544 |
+
12,
|
| 545 |
+
13,
|
| 546 |
+
14,
|
| 547 |
+
15
|
| 548 |
+
],
|
| 549 |
+
"modality_keys": [
|
| 550 |
+
"end_effector_position",
|
| 551 |
+
"end_effector_rotation",
|
| 552 |
+
"gripper_close"
|
| 553 |
+
],
|
| 554 |
+
"sin_cos_embedding_keys": null,
|
| 555 |
+
"mean_std_embedding_keys": null,
|
| 556 |
+
"action_configs": [
|
| 557 |
+
{
|
| 558 |
+
"rep": "DELTA",
|
| 559 |
+
"type": "EEF",
|
| 560 |
+
"format": "DEFAULT",
|
| 561 |
+
"state_key": null
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"rep": "DELTA",
|
| 565 |
+
"type": "EEF",
|
| 566 |
+
"format": "DEFAULT",
|
| 567 |
+
"state_key": null
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"rep": "ABSOLUTE",
|
| 571 |
+
"type": "NON_EEF",
|
| 572 |
+
"format": "DEFAULT",
|
| 573 |
+
"state_key": null
|
| 574 |
+
}
|
| 575 |
+
]
|
| 576 |
+
},
|
| 577 |
+
"language": {
|
| 578 |
+
"delta_indices": [
|
| 579 |
+
0
|
| 580 |
+
],
|
| 581 |
+
"modality_keys": [
|
| 582 |
+
"annotation.human.action.task_description"
|
| 583 |
+
],
|
| 584 |
+
"sin_cos_embedding_keys": null,
|
| 585 |
+
"mean_std_embedding_keys": null,
|
| 586 |
+
"action_configs": null
|
| 587 |
+
}
|
| 588 |
+
},
|
| 589 |
+
"berkeley_autolab_ur5": {
|
| 590 |
+
"video": {
|
| 591 |
+
"delta_indices": [
|
| 592 |
+
0
|
| 593 |
+
],
|
| 594 |
+
"modality_keys": [
|
| 595 |
+
"primary",
|
| 596 |
+
"wrist"
|
| 597 |
+
],
|
| 598 |
+
"sin_cos_embedding_keys": null,
|
| 599 |
+
"mean_std_embedding_keys": null,
|
| 600 |
+
"action_configs": null
|
| 601 |
+
},
|
| 602 |
+
"state": {
|
| 603 |
+
"delta_indices": [
|
| 604 |
+
0
|
| 605 |
+
],
|
| 606 |
+
"modality_keys": [
|
| 607 |
+
"end_effector_position",
|
| 608 |
+
"end_effector_rotation",
|
| 609 |
+
"gripper_position"
|
| 610 |
+
],
|
| 611 |
+
"sin_cos_embedding_keys": null,
|
| 612 |
+
"mean_std_embedding_keys": null,
|
| 613 |
+
"action_configs": null
|
| 614 |
+
},
|
| 615 |
+
"action": {
|
| 616 |
+
"delta_indices": [
|
| 617 |
+
0,
|
| 618 |
+
1,
|
| 619 |
+
2,
|
| 620 |
+
3,
|
| 621 |
+
4,
|
| 622 |
+
5,
|
| 623 |
+
6,
|
| 624 |
+
7,
|
| 625 |
+
8,
|
| 626 |
+
9,
|
| 627 |
+
10,
|
| 628 |
+
11,
|
| 629 |
+
12,
|
| 630 |
+
13,
|
| 631 |
+
14,
|
| 632 |
+
15
|
| 633 |
+
],
|
| 634 |
+
"modality_keys": [
|
| 635 |
+
"end_effector_position",
|
| 636 |
+
"end_effector_rotation",
|
| 637 |
+
"gripper_close"
|
| 638 |
+
],
|
| 639 |
+
"sin_cos_embedding_keys": null,
|
| 640 |
+
"mean_std_embedding_keys": null,
|
| 641 |
+
"action_configs": [
|
| 642 |
+
{
|
| 643 |
+
"rep": "DELTA",
|
| 644 |
+
"type": "EEF",
|
| 645 |
+
"format": "DEFAULT",
|
| 646 |
+
"state_key": null
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"rep": "DELTA",
|
| 650 |
+
"type": "EEF",
|
| 651 |
+
"format": "DEFAULT",
|
| 652 |
+
"state_key": null
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"rep": "ABSOLUTE",
|
| 656 |
+
"type": "NON_EEF",
|
| 657 |
+
"format": "DEFAULT",
|
| 658 |
+
"state_key": null
|
| 659 |
+
}
|
| 660 |
+
]
|
| 661 |
+
},
|
| 662 |
+
"language": {
|
| 663 |
+
"delta_indices": [
|
| 664 |
+
0
|
| 665 |
+
],
|
| 666 |
+
"modality_keys": [
|
| 667 |
+
"annotation.human.action.task_description"
|
| 668 |
+
],
|
| 669 |
+
"sin_cos_embedding_keys": null,
|
| 670 |
+
"mean_std_embedding_keys": null,
|
| 671 |
+
"action_configs": null
|
| 672 |
+
}
|
| 673 |
+
},
|
| 674 |
+
"language_table": {
|
| 675 |
+
"video": {
|
| 676 |
+
"delta_indices": [
|
| 677 |
+
0
|
| 678 |
+
],
|
| 679 |
+
"modality_keys": [
|
| 680 |
+
"primary"
|
| 681 |
+
],
|
| 682 |
+
"sin_cos_embedding_keys": null,
|
| 683 |
+
"mean_std_embedding_keys": null,
|
| 684 |
+
"action_configs": null
|
| 685 |
+
},
|
| 686 |
+
"state": {
|
| 687 |
+
"delta_indices": [
|
| 688 |
+
0
|
| 689 |
+
],
|
| 690 |
+
"modality_keys": [
|
| 691 |
+
"end_effector_position"
|
| 692 |
+
],
|
| 693 |
+
"sin_cos_embedding_keys": null,
|
| 694 |
+
"mean_std_embedding_keys": null,
|
| 695 |
+
"action_configs": null
|
| 696 |
+
},
|
| 697 |
+
"action": {
|
| 698 |
+
"delta_indices": [
|
| 699 |
+
0,
|
| 700 |
+
1,
|
| 701 |
+
2,
|
| 702 |
+
3,
|
| 703 |
+
4,
|
| 704 |
+
5,
|
| 705 |
+
6,
|
| 706 |
+
7,
|
| 707 |
+
8,
|
| 708 |
+
9,
|
| 709 |
+
10,
|
| 710 |
+
11,
|
| 711 |
+
12,
|
| 712 |
+
13,
|
| 713 |
+
14,
|
| 714 |
+
15
|
| 715 |
+
],
|
| 716 |
+
"modality_keys": [
|
| 717 |
+
"end_effector_position"
|
| 718 |
+
],
|
| 719 |
+
"sin_cos_embedding_keys": null,
|
| 720 |
+
"mean_std_embedding_keys": null,
|
| 721 |
+
"action_configs": [
|
| 722 |
+
{
|
| 723 |
+
"rep": "DELTA",
|
| 724 |
+
"type": "EEF",
|
| 725 |
+
"format": "DEFAULT",
|
| 726 |
+
"state_key": null
|
| 727 |
+
}
|
| 728 |
+
]
|
| 729 |
+
},
|
| 730 |
+
"language": {
|
| 731 |
+
"delta_indices": [
|
| 732 |
+
0
|
| 733 |
+
],
|
| 734 |
+
"modality_keys": [
|
| 735 |
+
"annotation.human.action.task_description"
|
| 736 |
+
],
|
| 737 |
+
"sin_cos_embedding_keys": null,
|
| 738 |
+
"mean_std_embedding_keys": null,
|
| 739 |
+
"action_configs": null
|
| 740 |
+
}
|
| 741 |
+
},
|
| 742 |
+
"viola": {
|
| 743 |
+
"video": {
|
| 744 |
+
"delta_indices": [
|
| 745 |
+
0
|
| 746 |
+
],
|
| 747 |
+
"modality_keys": [
|
| 748 |
+
"primary",
|
| 749 |
+
"wrist"
|
| 750 |
+
],
|
| 751 |
+
"sin_cos_embedding_keys": null,
|
| 752 |
+
"mean_std_embedding_keys": null,
|
| 753 |
+
"action_configs": null
|
| 754 |
+
},
|
| 755 |
+
"state": {
|
| 756 |
+
"delta_indices": [
|
| 757 |
+
0
|
| 758 |
+
],
|
| 759 |
+
"modality_keys": [
|
| 760 |
+
"joint_position",
|
| 761 |
+
"gripper_position"
|
| 762 |
+
],
|
| 763 |
+
"sin_cos_embedding_keys": null,
|
| 764 |
+
"mean_std_embedding_keys": null,
|
| 765 |
+
"action_configs": null
|
| 766 |
+
},
|
| 767 |
+
"action": {
|
| 768 |
+
"delta_indices": [
|
| 769 |
+
0,
|
| 770 |
+
1,
|
| 771 |
+
2,
|
| 772 |
+
3,
|
| 773 |
+
4,
|
| 774 |
+
5,
|
| 775 |
+
6,
|
| 776 |
+
7,
|
| 777 |
+
8,
|
| 778 |
+
9,
|
| 779 |
+
10,
|
| 780 |
+
11,
|
| 781 |
+
12,
|
| 782 |
+
13,
|
| 783 |
+
14,
|
| 784 |
+
15
|
| 785 |
+
],
|
| 786 |
+
"modality_keys": [
|
| 787 |
+
"end_effector_position",
|
| 788 |
+
"end_effector_rotation",
|
| 789 |
+
"gripper_close"
|
| 790 |
+
],
|
| 791 |
+
"sin_cos_embedding_keys": null,
|
| 792 |
+
"mean_std_embedding_keys": null,
|
| 793 |
+
"action_configs": [
|
| 794 |
+
{
|
| 795 |
+
"rep": "DELTA",
|
| 796 |
+
"type": "EEF",
|
| 797 |
+
"format": "DEFAULT",
|
| 798 |
+
"state_key": null
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"rep": "DELTA",
|
| 802 |
+
"type": "EEF",
|
| 803 |
+
"format": "DEFAULT",
|
| 804 |
+
"state_key": null
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"rep": "ABSOLUTE",
|
| 808 |
+
"type": "NON_EEF",
|
| 809 |
+
"format": "DEFAULT",
|
| 810 |
+
"state_key": null
|
| 811 |
+
}
|
| 812 |
+
]
|
| 813 |
+
},
|
| 814 |
+
"language": {
|
| 815 |
+
"delta_indices": [
|
| 816 |
+
0
|
| 817 |
+
],
|
| 818 |
+
"modality_keys": [
|
| 819 |
+
"annotation.human.action.task_description"
|
| 820 |
+
],
|
| 821 |
+
"sin_cos_embedding_keys": null,
|
| 822 |
+
"mean_std_embedding_keys": null,
|
| 823 |
+
"action_configs": null
|
| 824 |
+
}
|
| 825 |
+
},
|
| 826 |
+
"furniture_bench_dataset_converted_externally_to_rlds": {
|
| 827 |
+
"video": {
|
| 828 |
+
"delta_indices": [
|
| 829 |
+
0
|
| 830 |
+
],
|
| 831 |
+
"modality_keys": [
|
| 832 |
+
"primary",
|
| 833 |
+
"wrist"
|
| 834 |
+
],
|
| 835 |
+
"sin_cos_embedding_keys": null,
|
| 836 |
+
"mean_std_embedding_keys": null,
|
| 837 |
+
"action_configs": null
|
| 838 |
+
},
|
| 839 |
+
"state": {
|
| 840 |
+
"delta_indices": [
|
| 841 |
+
0
|
| 842 |
+
],
|
| 843 |
+
"modality_keys": [
|
| 844 |
+
"end_effector_position",
|
| 845 |
+
"end_effector_rotation",
|
| 846 |
+
"gripper_position"
|
| 847 |
+
],
|
| 848 |
+
"sin_cos_embedding_keys": null,
|
| 849 |
+
"mean_std_embedding_keys": null,
|
| 850 |
+
"action_configs": null
|
| 851 |
+
},
|
| 852 |
+
"action": {
|
| 853 |
+
"delta_indices": [
|
| 854 |
+
0,
|
| 855 |
+
1,
|
| 856 |
+
2,
|
| 857 |
+
3,
|
| 858 |
+
4,
|
| 859 |
+
5,
|
| 860 |
+
6,
|
| 861 |
+
7,
|
| 862 |
+
8,
|
| 863 |
+
9,
|
| 864 |
+
10,
|
| 865 |
+
11,
|
| 866 |
+
12,
|
| 867 |
+
13,
|
| 868 |
+
14,
|
| 869 |
+
15
|
| 870 |
+
],
|
| 871 |
+
"modality_keys": [
|
| 872 |
+
"end_effector_position",
|
| 873 |
+
"end_effector_rotation",
|
| 874 |
+
"gripper_close"
|
| 875 |
+
],
|
| 876 |
+
"sin_cos_embedding_keys": null,
|
| 877 |
+
"mean_std_embedding_keys": null,
|
| 878 |
+
"action_configs": [
|
| 879 |
+
{
|
| 880 |
+
"rep": "DELTA",
|
| 881 |
+
"type": "EEF",
|
| 882 |
+
"format": "DEFAULT",
|
| 883 |
+
"state_key": null
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"rep": "DELTA",
|
| 887 |
+
"type": "EEF",
|
| 888 |
+
"format": "DEFAULT",
|
| 889 |
+
"state_key": null
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"rep": "ABSOLUTE",
|
| 893 |
+
"type": "NON_EEF",
|
| 894 |
+
"format": "DEFAULT",
|
| 895 |
+
"state_key": null
|
| 896 |
+
}
|
| 897 |
+
]
|
| 898 |
+
},
|
| 899 |
+
"language": {
|
| 900 |
+
"delta_indices": [
|
| 901 |
+
0
|
| 902 |
+
],
|
| 903 |
+
"modality_keys": [
|
| 904 |
+
"annotation.human.action.task_description"
|
| 905 |
+
],
|
| 906 |
+
"sin_cos_embedding_keys": null,
|
| 907 |
+
"mean_std_embedding_keys": null,
|
| 908 |
+
"action_configs": null
|
| 909 |
+
}
|
| 910 |
+
},
|
| 911 |
+
"agibot_dexhand": {
|
| 912 |
+
"video": {
|
| 913 |
+
"delta_indices": [
|
| 914 |
+
0
|
| 915 |
+
],
|
| 916 |
+
"modality_keys": [
|
| 917 |
+
"primary"
|
| 918 |
+
],
|
| 919 |
+
"sin_cos_embedding_keys": null,
|
| 920 |
+
"mean_std_embedding_keys": null,
|
| 921 |
+
"action_configs": null
|
| 922 |
+
},
|
| 923 |
+
"state": {
|
| 924 |
+
"delta_indices": [
|
| 925 |
+
0
|
| 926 |
+
],
|
| 927 |
+
"modality_keys": [
|
| 928 |
+
"state"
|
| 929 |
+
],
|
| 930 |
+
"sin_cos_embedding_keys": null,
|
| 931 |
+
"mean_std_embedding_keys": null,
|
| 932 |
+
"action_configs": null
|
| 933 |
+
},
|
| 934 |
+
"action": {
|
| 935 |
+
"delta_indices": [
|
| 936 |
+
0,
|
| 937 |
+
1,
|
| 938 |
+
2,
|
| 939 |
+
3,
|
| 940 |
+
4,
|
| 941 |
+
5,
|
| 942 |
+
6,
|
| 943 |
+
7,
|
| 944 |
+
8,
|
| 945 |
+
9,
|
| 946 |
+
10,
|
| 947 |
+
11,
|
| 948 |
+
12,
|
| 949 |
+
13,
|
| 950 |
+
14,
|
| 951 |
+
15
|
| 952 |
+
],
|
| 953 |
+
"modality_keys": [
|
| 954 |
+
"action"
|
| 955 |
+
],
|
| 956 |
+
"sin_cos_embedding_keys": null,
|
| 957 |
+
"mean_std_embedding_keys": null,
|
| 958 |
+
"action_configs": [
|
| 959 |
+
{
|
| 960 |
+
"rep": "ABSOLUTE",
|
| 961 |
+
"type": "NON_EEF",
|
| 962 |
+
"format": "DEFAULT",
|
| 963 |
+
"state_key": null
|
| 964 |
+
}
|
| 965 |
+
]
|
| 966 |
+
},
|
| 967 |
+
"language": {
|
| 968 |
+
"delta_indices": [
|
| 969 |
+
0
|
| 970 |
+
],
|
| 971 |
+
"modality_keys": [
|
| 972 |
+
"annotation.human.action.task_description"
|
| 973 |
+
],
|
| 974 |
+
"sin_cos_embedding_keys": null,
|
| 975 |
+
"mean_std_embedding_keys": null,
|
| 976 |
+
"action_configs": null
|
| 977 |
+
}
|
| 978 |
+
},
|
| 979 |
+
"taco_play": {
|
| 980 |
+
"video": {
|
| 981 |
+
"delta_indices": [
|
| 982 |
+
0
|
| 983 |
+
],
|
| 984 |
+
"modality_keys": [
|
| 985 |
+
"primary",
|
| 986 |
+
"wrist"
|
| 987 |
+
],
|
| 988 |
+
"sin_cos_embedding_keys": null,
|
| 989 |
+
"mean_std_embedding_keys": null,
|
| 990 |
+
"action_configs": null
|
| 991 |
+
},
|
| 992 |
+
"state": {
|
| 993 |
+
"delta_indices": [
|
| 994 |
+
0
|
| 995 |
+
],
|
| 996 |
+
"modality_keys": [
|
| 997 |
+
"end_effector_position",
|
| 998 |
+
"end_effector_rotation",
|
| 999 |
+
"gripper_position"
|
| 1000 |
+
],
|
| 1001 |
+
"sin_cos_embedding_keys": null,
|
| 1002 |
+
"mean_std_embedding_keys": null,
|
| 1003 |
+
"action_configs": null
|
| 1004 |
+
},
|
| 1005 |
+
"action": {
|
| 1006 |
+
"delta_indices": [
|
| 1007 |
+
0,
|
| 1008 |
+
1,
|
| 1009 |
+
2,
|
| 1010 |
+
3,
|
| 1011 |
+
4,
|
| 1012 |
+
5,
|
| 1013 |
+
6,
|
| 1014 |
+
7,
|
| 1015 |
+
8,
|
| 1016 |
+
9,
|
| 1017 |
+
10,
|
| 1018 |
+
11,
|
| 1019 |
+
12,
|
| 1020 |
+
13,
|
| 1021 |
+
14,
|
| 1022 |
+
15
|
| 1023 |
+
],
|
| 1024 |
+
"modality_keys": [
|
| 1025 |
+
"end_effector_position",
|
| 1026 |
+
"end_effector_rotation",
|
| 1027 |
+
"gripper_close"
|
| 1028 |
+
],
|
| 1029 |
+
"sin_cos_embedding_keys": null,
|
| 1030 |
+
"mean_std_embedding_keys": null,
|
| 1031 |
+
"action_configs": [
|
| 1032 |
+
{
|
| 1033 |
+
"rep": "DELTA",
|
| 1034 |
+
"type": "EEF",
|
| 1035 |
+
"format": "DEFAULT",
|
| 1036 |
+
"state_key": null
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"rep": "DELTA",
|
| 1040 |
+
"type": "EEF",
|
| 1041 |
+
"format": "DEFAULT",
|
| 1042 |
+
"state_key": null
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"rep": "ABSOLUTE",
|
| 1046 |
+
"type": "NON_EEF",
|
| 1047 |
+
"format": "DEFAULT",
|
| 1048 |
+
"state_key": null
|
| 1049 |
+
}
|
| 1050 |
+
]
|
| 1051 |
+
},
|
| 1052 |
+
"language": {
|
| 1053 |
+
"delta_indices": [
|
| 1054 |
+
0
|
| 1055 |
+
],
|
| 1056 |
+
"modality_keys": [
|
| 1057 |
+
"annotation.human.action.task_description"
|
| 1058 |
+
],
|
| 1059 |
+
"sin_cos_embedding_keys": null,
|
| 1060 |
+
"mean_std_embedding_keys": null,
|
| 1061 |
+
"action_configs": null
|
| 1062 |
+
}
|
| 1063 |
+
},
|
| 1064 |
+
"berkeley_fanuc_manipulation": {
|
| 1065 |
+
"video": {
|
| 1066 |
+
"delta_indices": [
|
| 1067 |
+
0
|
| 1068 |
+
],
|
| 1069 |
+
"modality_keys": [
|
| 1070 |
+
"primary",
|
| 1071 |
+
"wrist"
|
| 1072 |
+
],
|
| 1073 |
+
"sin_cos_embedding_keys": null,
|
| 1074 |
+
"mean_std_embedding_keys": null,
|
| 1075 |
+
"action_configs": null
|
| 1076 |
+
},
|
| 1077 |
+
"state": {
|
| 1078 |
+
"delta_indices": [
|
| 1079 |
+
0
|
| 1080 |
+
],
|
| 1081 |
+
"modality_keys": [
|
| 1082 |
+
"joint_position",
|
| 1083 |
+
"gripper_position"
|
| 1084 |
+
],
|
| 1085 |
+
"sin_cos_embedding_keys": null,
|
| 1086 |
+
"mean_std_embedding_keys": null,
|
| 1087 |
+
"action_configs": null
|
| 1088 |
+
},
|
| 1089 |
+
"action": {
|
| 1090 |
+
"delta_indices": [
|
| 1091 |
+
0,
|
| 1092 |
+
1,
|
| 1093 |
+
2,
|
| 1094 |
+
3,
|
| 1095 |
+
4,
|
| 1096 |
+
5,
|
| 1097 |
+
6,
|
| 1098 |
+
7,
|
| 1099 |
+
8,
|
| 1100 |
+
9,
|
| 1101 |
+
10,
|
| 1102 |
+
11,
|
| 1103 |
+
12,
|
| 1104 |
+
13,
|
| 1105 |
+
14,
|
| 1106 |
+
15
|
| 1107 |
+
],
|
| 1108 |
+
"modality_keys": [
|
| 1109 |
+
"end_effector_position",
|
| 1110 |
+
"end_effector_rotation",
|
| 1111 |
+
"gripper_close"
|
| 1112 |
+
],
|
| 1113 |
+
"sin_cos_embedding_keys": null,
|
| 1114 |
+
"mean_std_embedding_keys": null,
|
| 1115 |
+
"action_configs": [
|
| 1116 |
+
{
|
| 1117 |
+
"rep": "DELTA",
|
| 1118 |
+
"type": "EEF",
|
| 1119 |
+
"format": "DEFAULT",
|
| 1120 |
+
"state_key": null
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"rep": "DELTA",
|
| 1124 |
+
"type": "EEF",
|
| 1125 |
+
"format": "DEFAULT",
|
| 1126 |
+
"state_key": null
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"rep": "ABSOLUTE",
|
| 1130 |
+
"type": "NON_EEF",
|
| 1131 |
+
"format": "DEFAULT",
|
| 1132 |
+
"state_key": null
|
| 1133 |
+
}
|
| 1134 |
+
]
|
| 1135 |
+
},
|
| 1136 |
+
"language": {
|
| 1137 |
+
"delta_indices": [
|
| 1138 |
+
0
|
| 1139 |
+
],
|
| 1140 |
+
"modality_keys": [
|
| 1141 |
+
"annotation.human.action.task_description"
|
| 1142 |
+
],
|
| 1143 |
+
"sin_cos_embedding_keys": null,
|
| 1144 |
+
"mean_std_embedding_keys": null,
|
| 1145 |
+
"action_configs": null
|
| 1146 |
+
}
|
| 1147 |
+
},
|
| 1148 |
+
"bc_z": {
|
| 1149 |
+
"video": {
|
| 1150 |
+
"delta_indices": [
|
| 1151 |
+
0
|
| 1152 |
+
],
|
| 1153 |
+
"modality_keys": [
|
| 1154 |
+
"primary"
|
| 1155 |
+
],
|
| 1156 |
+
"sin_cos_embedding_keys": null,
|
| 1157 |
+
"mean_std_embedding_keys": null,
|
| 1158 |
+
"action_configs": null
|
| 1159 |
+
},
|
| 1160 |
+
"state": {
|
| 1161 |
+
"delta_indices": [
|
| 1162 |
+
0
|
| 1163 |
+
],
|
| 1164 |
+
"modality_keys": [
|
| 1165 |
+
"end_effector_position",
|
| 1166 |
+
"end_effector_rotation",
|
| 1167 |
+
"gripper_position"
|
| 1168 |
+
],
|
| 1169 |
+
"sin_cos_embedding_keys": null,
|
| 1170 |
+
"mean_std_embedding_keys": null,
|
| 1171 |
+
"action_configs": null
|
| 1172 |
+
},
|
| 1173 |
+
"action": {
|
| 1174 |
+
"delta_indices": [
|
| 1175 |
+
0,
|
| 1176 |
+
1,
|
| 1177 |
+
2,
|
| 1178 |
+
3,
|
| 1179 |
+
4,
|
| 1180 |
+
5,
|
| 1181 |
+
6,
|
| 1182 |
+
7,
|
| 1183 |
+
8,
|
| 1184 |
+
9,
|
| 1185 |
+
10,
|
| 1186 |
+
11,
|
| 1187 |
+
12,
|
| 1188 |
+
13,
|
| 1189 |
+
14,
|
| 1190 |
+
15
|
| 1191 |
+
],
|
| 1192 |
+
"modality_keys": [
|
| 1193 |
+
"end_effector_position",
|
| 1194 |
+
"end_effector_rotation",
|
| 1195 |
+
"gripper_close"
|
| 1196 |
+
],
|
| 1197 |
+
"sin_cos_embedding_keys": null,
|
| 1198 |
+
"mean_std_embedding_keys": null,
|
| 1199 |
+
"action_configs": [
|
| 1200 |
+
{
|
| 1201 |
+
"rep": "DELTA",
|
| 1202 |
+
"type": "EEF",
|
| 1203 |
+
"format": "DEFAULT",
|
| 1204 |
+
"state_key": null
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"rep": "DELTA",
|
| 1208 |
+
"type": "EEF",
|
| 1209 |
+
"format": "DEFAULT",
|
| 1210 |
+
"state_key": null
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"rep": "ABSOLUTE",
|
| 1214 |
+
"type": "NON_EEF",
|
| 1215 |
+
"format": "DEFAULT",
|
| 1216 |
+
"state_key": null
|
| 1217 |
+
}
|
| 1218 |
+
]
|
| 1219 |
+
},
|
| 1220 |
+
"language": {
|
| 1221 |
+
"delta_indices": [
|
| 1222 |
+
0
|
| 1223 |
+
],
|
| 1224 |
+
"modality_keys": [
|
| 1225 |
+
"annotation.human.action.task_description"
|
| 1226 |
+
],
|
| 1227 |
+
"sin_cos_embedding_keys": null,
|
| 1228 |
+
"mean_std_embedding_keys": null,
|
| 1229 |
+
"action_configs": null
|
| 1230 |
+
}
|
| 1231 |
+
},
|
| 1232 |
+
"toto": {
|
| 1233 |
+
"video": {
|
| 1234 |
+
"delta_indices": [
|
| 1235 |
+
0
|
| 1236 |
+
],
|
| 1237 |
+
"modality_keys": [
|
| 1238 |
+
"primary"
|
| 1239 |
+
],
|
| 1240 |
+
"sin_cos_embedding_keys": null,
|
| 1241 |
+
"mean_std_embedding_keys": null,
|
| 1242 |
+
"action_configs": null
|
| 1243 |
+
},
|
| 1244 |
+
"state": {
|
| 1245 |
+
"delta_indices": [
|
| 1246 |
+
0
|
| 1247 |
+
],
|
| 1248 |
+
"modality_keys": [
|
| 1249 |
+
"joint_position",
|
| 1250 |
+
"gripper_position"
|
| 1251 |
+
],
|
| 1252 |
+
"sin_cos_embedding_keys": null,
|
| 1253 |
+
"mean_std_embedding_keys": null,
|
| 1254 |
+
"action_configs": null
|
| 1255 |
+
},
|
| 1256 |
+
"action": {
|
| 1257 |
+
"delta_indices": [
|
| 1258 |
+
0,
|
| 1259 |
+
1,
|
| 1260 |
+
2,
|
| 1261 |
+
3,
|
| 1262 |
+
4,
|
| 1263 |
+
5,
|
| 1264 |
+
6,
|
| 1265 |
+
7,
|
| 1266 |
+
8,
|
| 1267 |
+
9,
|
| 1268 |
+
10,
|
| 1269 |
+
11,
|
| 1270 |
+
12,
|
| 1271 |
+
13,
|
| 1272 |
+
14,
|
| 1273 |
+
15
|
| 1274 |
+
],
|
| 1275 |
+
"modality_keys": [
|
| 1276 |
+
"end_effector_position",
|
| 1277 |
+
"end_effector_rotation",
|
| 1278 |
+
"gripper_close"
|
| 1279 |
+
],
|
| 1280 |
+
"sin_cos_embedding_keys": null,
|
| 1281 |
+
"mean_std_embedding_keys": null,
|
| 1282 |
+
"action_configs": [
|
| 1283 |
+
{
|
| 1284 |
+
"rep": "DELTA",
|
| 1285 |
+
"type": "EEF",
|
| 1286 |
+
"format": "DEFAULT",
|
| 1287 |
+
"state_key": null
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"rep": "DELTA",
|
| 1291 |
+
"type": "EEF",
|
| 1292 |
+
"format": "DEFAULT",
|
| 1293 |
+
"state_key": null
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"rep": "ABSOLUTE",
|
| 1297 |
+
"type": "NON_EEF",
|
| 1298 |
+
"format": "DEFAULT",
|
| 1299 |
+
"state_key": null
|
| 1300 |
+
}
|
| 1301 |
+
]
|
| 1302 |
+
},
|
| 1303 |
+
"language": {
|
| 1304 |
+
"delta_indices": [
|
| 1305 |
+
0
|
| 1306 |
+
],
|
| 1307 |
+
"modality_keys": [
|
| 1308 |
+
"annotation.human.action.task_description"
|
| 1309 |
+
],
|
| 1310 |
+
"sin_cos_embedding_keys": null,
|
| 1311 |
+
"mean_std_embedding_keys": null,
|
| 1312 |
+
"action_configs": null
|
| 1313 |
+
}
|
| 1314 |
+
},
|
| 1315 |
+
"dobbe": {
|
| 1316 |
+
"video": {
|
| 1317 |
+
"delta_indices": [
|
| 1318 |
+
0
|
| 1319 |
+
],
|
| 1320 |
+
"modality_keys": [
|
| 1321 |
+
"wrist"
|
| 1322 |
+
],
|
| 1323 |
+
"sin_cos_embedding_keys": null,
|
| 1324 |
+
"mean_std_embedding_keys": null,
|
| 1325 |
+
"action_configs": null
|
| 1326 |
+
},
|
| 1327 |
+
"state": {
|
| 1328 |
+
"delta_indices": [
|
| 1329 |
+
0
|
| 1330 |
+
],
|
| 1331 |
+
"modality_keys": [
|
| 1332 |
+
"end_effector_position",
|
| 1333 |
+
"end_effector_rotation",
|
| 1334 |
+
"gripper_position"
|
| 1335 |
+
],
|
| 1336 |
+
"sin_cos_embedding_keys": null,
|
| 1337 |
+
"mean_std_embedding_keys": null,
|
| 1338 |
+
"action_configs": null
|
| 1339 |
+
},
|
| 1340 |
+
"action": {
|
| 1341 |
+
"delta_indices": [
|
| 1342 |
+
0,
|
| 1343 |
+
1,
|
| 1344 |
+
2,
|
| 1345 |
+
3,
|
| 1346 |
+
4,
|
| 1347 |
+
5,
|
| 1348 |
+
6,
|
| 1349 |
+
7,
|
| 1350 |
+
8,
|
| 1351 |
+
9,
|
| 1352 |
+
10,
|
| 1353 |
+
11,
|
| 1354 |
+
12,
|
| 1355 |
+
13,
|
| 1356 |
+
14,
|
| 1357 |
+
15
|
| 1358 |
+
],
|
| 1359 |
+
"modality_keys": [
|
| 1360 |
+
"end_effector_position",
|
| 1361 |
+
"end_effector_rotation",
|
| 1362 |
+
"gripper_close"
|
| 1363 |
+
],
|
| 1364 |
+
"sin_cos_embedding_keys": null,
|
| 1365 |
+
"mean_std_embedding_keys": null,
|
| 1366 |
+
"action_configs": [
|
| 1367 |
+
{
|
| 1368 |
+
"rep": "DELTA",
|
| 1369 |
+
"type": "EEF",
|
| 1370 |
+
"format": "DEFAULT",
|
| 1371 |
+
"state_key": null
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"rep": "DELTA",
|
| 1375 |
+
"type": "EEF",
|
| 1376 |
+
"format": "DEFAULT",
|
| 1377 |
+
"state_key": null
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"rep": "ABSOLUTE",
|
| 1381 |
+
"type": "NON_EEF",
|
| 1382 |
+
"format": "DEFAULT",
|
| 1383 |
+
"state_key": null
|
| 1384 |
+
}
|
| 1385 |
+
]
|
| 1386 |
+
},
|
| 1387 |
+
"language": {
|
| 1388 |
+
"delta_indices": [
|
| 1389 |
+
0
|
| 1390 |
+
],
|
| 1391 |
+
"modality_keys": [
|
| 1392 |
+
"annotation.human.action.task_description"
|
| 1393 |
+
],
|
| 1394 |
+
"sin_cos_embedding_keys": null,
|
| 1395 |
+
"mean_std_embedding_keys": null,
|
| 1396 |
+
"action_configs": null
|
| 1397 |
+
}
|
| 1398 |
+
},
|
| 1399 |
+
"droid": {
|
| 1400 |
+
"video": {
|
| 1401 |
+
"delta_indices": [
|
| 1402 |
+
0
|
| 1403 |
+
],
|
| 1404 |
+
"modality_keys": [
|
| 1405 |
+
"primary",
|
| 1406 |
+
"secondary",
|
| 1407 |
+
"wrist"
|
| 1408 |
+
],
|
| 1409 |
+
"sin_cos_embedding_keys": null,
|
| 1410 |
+
"mean_std_embedding_keys": null,
|
| 1411 |
+
"action_configs": null
|
| 1412 |
+
},
|
| 1413 |
+
"state": {
|
| 1414 |
+
"delta_indices": [
|
| 1415 |
+
0
|
| 1416 |
+
],
|
| 1417 |
+
"modality_keys": [
|
| 1418 |
+
"end_effector_position",
|
| 1419 |
+
"end_effector_rotation",
|
| 1420 |
+
"gripper_position"
|
| 1421 |
+
],
|
| 1422 |
+
"sin_cos_embedding_keys": null,
|
| 1423 |
+
"mean_std_embedding_keys": null,
|
| 1424 |
+
"action_configs": null
|
| 1425 |
+
},
|
| 1426 |
+
"action": {
|
| 1427 |
+
"delta_indices": [
|
| 1428 |
+
0,
|
| 1429 |
+
1,
|
| 1430 |
+
2,
|
| 1431 |
+
3,
|
| 1432 |
+
4,
|
| 1433 |
+
5,
|
| 1434 |
+
6,
|
| 1435 |
+
7,
|
| 1436 |
+
8,
|
| 1437 |
+
9,
|
| 1438 |
+
10,
|
| 1439 |
+
11,
|
| 1440 |
+
12,
|
| 1441 |
+
13,
|
| 1442 |
+
14,
|
| 1443 |
+
15
|
| 1444 |
+
],
|
| 1445 |
+
"modality_keys": [
|
| 1446 |
+
"end_effector_position",
|
| 1447 |
+
"end_effector_rotation",
|
| 1448 |
+
"gripper_close"
|
| 1449 |
+
],
|
| 1450 |
+
"sin_cos_embedding_keys": null,
|
| 1451 |
+
"mean_std_embedding_keys": null,
|
| 1452 |
+
"action_configs": [
|
| 1453 |
+
{
|
| 1454 |
+
"rep": "DELTA",
|
| 1455 |
+
"type": "EEF",
|
| 1456 |
+
"format": "DEFAULT",
|
| 1457 |
+
"state_key": null
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"rep": "DELTA",
|
| 1461 |
+
"type": "EEF",
|
| 1462 |
+
"format": "DEFAULT",
|
| 1463 |
+
"state_key": null
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"rep": "ABSOLUTE",
|
| 1467 |
+
"type": "NON_EEF",
|
| 1468 |
+
"format": "DEFAULT",
|
| 1469 |
+
"state_key": null
|
| 1470 |
+
}
|
| 1471 |
+
]
|
| 1472 |
+
},
|
| 1473 |
+
"language": {
|
| 1474 |
+
"delta_indices": [
|
| 1475 |
+
0
|
| 1476 |
+
],
|
| 1477 |
+
"modality_keys": [
|
| 1478 |
+
"annotation.human.action.task_description"
|
| 1479 |
+
],
|
| 1480 |
+
"sin_cos_embedding_keys": null,
|
| 1481 |
+
"mean_std_embedding_keys": null,
|
| 1482 |
+
"action_configs": null
|
| 1483 |
+
}
|
| 1484 |
+
},
|
| 1485 |
+
"ucsd_kitchen_dataset_converted_externally_to_rlds": {
|
| 1486 |
+
"video": {
|
| 1487 |
+
"delta_indices": [
|
| 1488 |
+
0
|
| 1489 |
+
],
|
| 1490 |
+
"modality_keys": [
|
| 1491 |
+
"primary"
|
| 1492 |
+
],
|
| 1493 |
+
"sin_cos_embedding_keys": null,
|
| 1494 |
+
"mean_std_embedding_keys": null,
|
| 1495 |
+
"action_configs": null
|
| 1496 |
+
},
|
| 1497 |
+
"state": {
|
| 1498 |
+
"delta_indices": [
|
| 1499 |
+
0
|
| 1500 |
+
],
|
| 1501 |
+
"modality_keys": [
|
| 1502 |
+
"joint_position"
|
| 1503 |
+
],
|
| 1504 |
+
"sin_cos_embedding_keys": null,
|
| 1505 |
+
"mean_std_embedding_keys": null,
|
| 1506 |
+
"action_configs": null
|
| 1507 |
+
},
|
| 1508 |
+
"action": {
|
| 1509 |
+
"delta_indices": [
|
| 1510 |
+
0,
|
| 1511 |
+
1,
|
| 1512 |
+
2,
|
| 1513 |
+
3,
|
| 1514 |
+
4,
|
| 1515 |
+
5,
|
| 1516 |
+
6,
|
| 1517 |
+
7,
|
| 1518 |
+
8,
|
| 1519 |
+
9,
|
| 1520 |
+
10,
|
| 1521 |
+
11,
|
| 1522 |
+
12,
|
| 1523 |
+
13,
|
| 1524 |
+
14,
|
| 1525 |
+
15
|
| 1526 |
+
],
|
| 1527 |
+
"modality_keys": [
|
| 1528 |
+
"end_effector_position",
|
| 1529 |
+
"end_effector_rotation",
|
| 1530 |
+
"gripper_close"
|
| 1531 |
+
],
|
| 1532 |
+
"sin_cos_embedding_keys": null,
|
| 1533 |
+
"mean_std_embedding_keys": null,
|
| 1534 |
+
"action_configs": [
|
| 1535 |
+
{
|
| 1536 |
+
"rep": "DELTA",
|
| 1537 |
+
"type": "EEF",
|
| 1538 |
+
"format": "DEFAULT",
|
| 1539 |
+
"state_key": null
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"rep": "DELTA",
|
| 1543 |
+
"type": "EEF",
|
| 1544 |
+
"format": "DEFAULT",
|
| 1545 |
+
"state_key": null
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"rep": "ABSOLUTE",
|
| 1549 |
+
"type": "NON_EEF",
|
| 1550 |
+
"format": "DEFAULT",
|
| 1551 |
+
"state_key": null
|
| 1552 |
+
}
|
| 1553 |
+
]
|
| 1554 |
+
},
|
| 1555 |
+
"language": {
|
| 1556 |
+
"delta_indices": [
|
| 1557 |
+
0
|
| 1558 |
+
],
|
| 1559 |
+
"modality_keys": [
|
| 1560 |
+
"annotation.human.action.task_description"
|
| 1561 |
+
],
|
| 1562 |
+
"sin_cos_embedding_keys": null,
|
| 1563 |
+
"mean_std_embedding_keys": null,
|
| 1564 |
+
"action_configs": null
|
| 1565 |
+
}
|
| 1566 |
+
},
|
| 1567 |
+
"stanford_hydra_dataset_converted_externally_to_rlds": {
|
| 1568 |
+
"video": {
|
| 1569 |
+
"delta_indices": [
|
| 1570 |
+
0
|
| 1571 |
+
],
|
| 1572 |
+
"modality_keys": [
|
| 1573 |
+
"primary",
|
| 1574 |
+
"wrist"
|
| 1575 |
+
],
|
| 1576 |
+
"sin_cos_embedding_keys": null,
|
| 1577 |
+
"mean_std_embedding_keys": null,
|
| 1578 |
+
"action_configs": null
|
| 1579 |
+
},
|
| 1580 |
+
"state": {
|
| 1581 |
+
"delta_indices": [
|
| 1582 |
+
0
|
| 1583 |
+
],
|
| 1584 |
+
"modality_keys": [
|
| 1585 |
+
"end_effector_position",
|
| 1586 |
+
"end_effector_rotation",
|
| 1587 |
+
"gripper_position"
|
| 1588 |
+
],
|
| 1589 |
+
"sin_cos_embedding_keys": null,
|
| 1590 |
+
"mean_std_embedding_keys": null,
|
| 1591 |
+
"action_configs": null
|
| 1592 |
+
},
|
| 1593 |
+
"action": {
|
| 1594 |
+
"delta_indices": [
|
| 1595 |
+
0,
|
| 1596 |
+
1,
|
| 1597 |
+
2,
|
| 1598 |
+
3,
|
| 1599 |
+
4,
|
| 1600 |
+
5,
|
| 1601 |
+
6,
|
| 1602 |
+
7,
|
| 1603 |
+
8,
|
| 1604 |
+
9,
|
| 1605 |
+
10,
|
| 1606 |
+
11,
|
| 1607 |
+
12,
|
| 1608 |
+
13,
|
| 1609 |
+
14,
|
| 1610 |
+
15
|
| 1611 |
+
],
|
| 1612 |
+
"modality_keys": [
|
| 1613 |
+
"end_effector_position",
|
| 1614 |
+
"end_effector_rotation",
|
| 1615 |
+
"gripper_close"
|
| 1616 |
+
],
|
| 1617 |
+
"sin_cos_embedding_keys": null,
|
| 1618 |
+
"mean_std_embedding_keys": null,
|
| 1619 |
+
"action_configs": [
|
| 1620 |
+
{
|
| 1621 |
+
"rep": "DELTA",
|
| 1622 |
+
"type": "EEF",
|
| 1623 |
+
"format": "DEFAULT",
|
| 1624 |
+
"state_key": null
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"rep": "DELTA",
|
| 1628 |
+
"type": "EEF",
|
| 1629 |
+
"format": "DEFAULT",
|
| 1630 |
+
"state_key": null
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"rep": "ABSOLUTE",
|
| 1634 |
+
"type": "NON_EEF",
|
| 1635 |
+
"format": "DEFAULT",
|
| 1636 |
+
"state_key": null
|
| 1637 |
+
}
|
| 1638 |
+
]
|
| 1639 |
+
},
|
| 1640 |
+
"language": {
|
| 1641 |
+
"delta_indices": [
|
| 1642 |
+
0
|
| 1643 |
+
],
|
| 1644 |
+
"modality_keys": [
|
| 1645 |
+
"annotation.human.action.task_description"
|
| 1646 |
+
],
|
| 1647 |
+
"sin_cos_embedding_keys": null,
|
| 1648 |
+
"mean_std_embedding_keys": null,
|
| 1649 |
+
"action_configs": null
|
| 1650 |
+
}
|
| 1651 |
+
},
|
| 1652 |
+
"iamlab_cmu_pickup_insert_converted_externally_to_rlds": {
|
| 1653 |
+
"video": {
|
| 1654 |
+
"delta_indices": [
|
| 1655 |
+
0
|
| 1656 |
+
],
|
| 1657 |
+
"modality_keys": [
|
| 1658 |
+
"primary",
|
| 1659 |
+
"wrist"
|
| 1660 |
+
],
|
| 1661 |
+
"sin_cos_embedding_keys": null,
|
| 1662 |
+
"mean_std_embedding_keys": null,
|
| 1663 |
+
"action_configs": null
|
| 1664 |
+
},
|
| 1665 |
+
"state": {
|
| 1666 |
+
"delta_indices": [
|
| 1667 |
+
0
|
| 1668 |
+
],
|
| 1669 |
+
"modality_keys": [
|
| 1670 |
+
"end_effector_position",
|
| 1671 |
+
"end_effector_rotation",
|
| 1672 |
+
"gripper_position"
|
| 1673 |
+
],
|
| 1674 |
+
"sin_cos_embedding_keys": null,
|
| 1675 |
+
"mean_std_embedding_keys": null,
|
| 1676 |
+
"action_configs": null
|
| 1677 |
+
},
|
| 1678 |
+
"action": {
|
| 1679 |
+
"delta_indices": [
|
| 1680 |
+
0,
|
| 1681 |
+
1,
|
| 1682 |
+
2,
|
| 1683 |
+
3,
|
| 1684 |
+
4,
|
| 1685 |
+
5,
|
| 1686 |
+
6,
|
| 1687 |
+
7,
|
| 1688 |
+
8,
|
| 1689 |
+
9,
|
| 1690 |
+
10,
|
| 1691 |
+
11,
|
| 1692 |
+
12,
|
| 1693 |
+
13,
|
| 1694 |
+
14,
|
| 1695 |
+
15
|
| 1696 |
+
],
|
| 1697 |
+
"modality_keys": [
|
| 1698 |
+
"end_effector_position",
|
| 1699 |
+
"end_effector_rotation",
|
| 1700 |
+
"gripper_close"
|
| 1701 |
+
],
|
| 1702 |
+
"sin_cos_embedding_keys": null,
|
| 1703 |
+
"mean_std_embedding_keys": null,
|
| 1704 |
+
"action_configs": [
|
| 1705 |
+
{
|
| 1706 |
+
"rep": "DELTA",
|
| 1707 |
+
"type": "EEF",
|
| 1708 |
+
"format": "DEFAULT",
|
| 1709 |
+
"state_key": null
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"rep": "DELTA",
|
| 1713 |
+
"type": "EEF",
|
| 1714 |
+
"format": "DEFAULT",
|
| 1715 |
+
"state_key": null
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"rep": "ABSOLUTE",
|
| 1719 |
+
"type": "NON_EEF",
|
| 1720 |
+
"format": "DEFAULT",
|
| 1721 |
+
"state_key": null
|
| 1722 |
+
}
|
| 1723 |
+
]
|
| 1724 |
+
},
|
| 1725 |
+
"language": {
|
| 1726 |
+
"delta_indices": [
|
| 1727 |
+
0
|
| 1728 |
+
],
|
| 1729 |
+
"modality_keys": [
|
| 1730 |
+
"annotation.human.action.task_description"
|
| 1731 |
+
],
|
| 1732 |
+
"sin_cos_embedding_keys": null,
|
| 1733 |
+
"mean_std_embedding_keys": null,
|
| 1734 |
+
"action_configs": null
|
| 1735 |
+
}
|
| 1736 |
+
},
|
| 1737 |
+
"nyu_franka_play_dataset_converted_externally_to_rlds": {
|
| 1738 |
+
"video": {
|
| 1739 |
+
"delta_indices": [
|
| 1740 |
+
0
|
| 1741 |
+
],
|
| 1742 |
+
"modality_keys": [
|
| 1743 |
+
"primary",
|
| 1744 |
+
"secondary"
|
| 1745 |
+
],
|
| 1746 |
+
"sin_cos_embedding_keys": null,
|
| 1747 |
+
"mean_std_embedding_keys": null,
|
| 1748 |
+
"action_configs": null
|
| 1749 |
+
},
|
| 1750 |
+
"state": {
|
| 1751 |
+
"delta_indices": [
|
| 1752 |
+
0
|
| 1753 |
+
],
|
| 1754 |
+
"modality_keys": [
|
| 1755 |
+
"end_effector_position",
|
| 1756 |
+
"end_effector_rotation",
|
| 1757 |
+
"gripper_position"
|
| 1758 |
+
],
|
| 1759 |
+
"sin_cos_embedding_keys": null,
|
| 1760 |
+
"mean_std_embedding_keys": null,
|
| 1761 |
+
"action_configs": null
|
| 1762 |
+
},
|
| 1763 |
+
"action": {
|
| 1764 |
+
"delta_indices": [
|
| 1765 |
+
0,
|
| 1766 |
+
1,
|
| 1767 |
+
2,
|
| 1768 |
+
3,
|
| 1769 |
+
4,
|
| 1770 |
+
5,
|
| 1771 |
+
6,
|
| 1772 |
+
7,
|
| 1773 |
+
8,
|
| 1774 |
+
9,
|
| 1775 |
+
10,
|
| 1776 |
+
11,
|
| 1777 |
+
12,
|
| 1778 |
+
13,
|
| 1779 |
+
14,
|
| 1780 |
+
15
|
| 1781 |
+
],
|
| 1782 |
+
"modality_keys": [
|
| 1783 |
+
"end_effector_position",
|
| 1784 |
+
"end_effector_rotation",
|
| 1785 |
+
"gripper_close"
|
| 1786 |
+
],
|
| 1787 |
+
"sin_cos_embedding_keys": null,
|
| 1788 |
+
"mean_std_embedding_keys": null,
|
| 1789 |
+
"action_configs": [
|
| 1790 |
+
{
|
| 1791 |
+
"rep": "DELTA",
|
| 1792 |
+
"type": "EEF",
|
| 1793 |
+
"format": "DEFAULT",
|
| 1794 |
+
"state_key": null
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"rep": "DELTA",
|
| 1798 |
+
"type": "EEF",
|
| 1799 |
+
"format": "DEFAULT",
|
| 1800 |
+
"state_key": null
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"rep": "ABSOLUTE",
|
| 1804 |
+
"type": "NON_EEF",
|
| 1805 |
+
"format": "DEFAULT",
|
| 1806 |
+
"state_key": null
|
| 1807 |
+
}
|
| 1808 |
+
]
|
| 1809 |
+
},
|
| 1810 |
+
"language": {
|
| 1811 |
+
"delta_indices": [
|
| 1812 |
+
0
|
| 1813 |
+
],
|
| 1814 |
+
"modality_keys": [
|
| 1815 |
+
"annotation.human.action.task_description"
|
| 1816 |
+
],
|
| 1817 |
+
"sin_cos_embedding_keys": null,
|
| 1818 |
+
"mean_std_embedding_keys": null,
|
| 1819 |
+
"action_configs": null
|
| 1820 |
+
}
|
| 1821 |
+
},
|
| 1822 |
+
"austin_sailor_dataset_converted_externally_to_rlds": {
|
| 1823 |
+
"video": {
|
| 1824 |
+
"delta_indices": [
|
| 1825 |
+
0
|
| 1826 |
+
],
|
| 1827 |
+
"modality_keys": [
|
| 1828 |
+
"primary",
|
| 1829 |
+
"wrist"
|
| 1830 |
+
],
|
| 1831 |
+
"sin_cos_embedding_keys": null,
|
| 1832 |
+
"mean_std_embedding_keys": null,
|
| 1833 |
+
"action_configs": null
|
| 1834 |
+
},
|
| 1835 |
+
"state": {
|
| 1836 |
+
"delta_indices": [
|
| 1837 |
+
0
|
| 1838 |
+
],
|
| 1839 |
+
"modality_keys": [
|
| 1840 |
+
"end_effector_position",
|
| 1841 |
+
"end_effector_rotation",
|
| 1842 |
+
"gripper_position"
|
| 1843 |
+
],
|
| 1844 |
+
"sin_cos_embedding_keys": null,
|
| 1845 |
+
"mean_std_embedding_keys": null,
|
| 1846 |
+
"action_configs": null
|
| 1847 |
+
},
|
| 1848 |
+
"action": {
|
| 1849 |
+
"delta_indices": [
|
| 1850 |
+
0,
|
| 1851 |
+
1,
|
| 1852 |
+
2,
|
| 1853 |
+
3,
|
| 1854 |
+
4,
|
| 1855 |
+
5,
|
| 1856 |
+
6,
|
| 1857 |
+
7,
|
| 1858 |
+
8,
|
| 1859 |
+
9,
|
| 1860 |
+
10,
|
| 1861 |
+
11,
|
| 1862 |
+
12,
|
| 1863 |
+
13,
|
| 1864 |
+
14,
|
| 1865 |
+
15
|
| 1866 |
+
],
|
| 1867 |
+
"modality_keys": [
|
| 1868 |
+
"end_effector_position",
|
| 1869 |
+
"end_effector_rotation",
|
| 1870 |
+
"gripper_close"
|
| 1871 |
+
],
|
| 1872 |
+
"sin_cos_embedding_keys": null,
|
| 1873 |
+
"mean_std_embedding_keys": null,
|
| 1874 |
+
"action_configs": [
|
| 1875 |
+
{
|
| 1876 |
+
"rep": "DELTA",
|
| 1877 |
+
"type": "EEF",
|
| 1878 |
+
"format": "DEFAULT",
|
| 1879 |
+
"state_key": null
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"rep": "DELTA",
|
| 1883 |
+
"type": "EEF",
|
| 1884 |
+
"format": "DEFAULT",
|
| 1885 |
+
"state_key": null
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"rep": "ABSOLUTE",
|
| 1889 |
+
"type": "NON_EEF",
|
| 1890 |
+
"format": "DEFAULT",
|
| 1891 |
+
"state_key": null
|
| 1892 |
+
}
|
| 1893 |
+
]
|
| 1894 |
+
},
|
| 1895 |
+
"language": {
|
| 1896 |
+
"delta_indices": [
|
| 1897 |
+
0
|
| 1898 |
+
],
|
| 1899 |
+
"modality_keys": [
|
| 1900 |
+
"annotation.human.action.task_description"
|
| 1901 |
+
],
|
| 1902 |
+
"sin_cos_embedding_keys": null,
|
| 1903 |
+
"mean_std_embedding_keys": null,
|
| 1904 |
+
"action_configs": null
|
| 1905 |
+
}
|
| 1906 |
+
},
|
| 1907 |
+
"utaustin_mutex": {
|
| 1908 |
+
"video": {
|
| 1909 |
+
"delta_indices": [
|
| 1910 |
+
0
|
| 1911 |
+
],
|
| 1912 |
+
"modality_keys": [
|
| 1913 |
+
"primary",
|
| 1914 |
+
"wrist"
|
| 1915 |
+
],
|
| 1916 |
+
"sin_cos_embedding_keys": null,
|
| 1917 |
+
"mean_std_embedding_keys": null,
|
| 1918 |
+
"action_configs": null
|
| 1919 |
+
},
|
| 1920 |
+
"state": {
|
| 1921 |
+
"delta_indices": [
|
| 1922 |
+
0
|
| 1923 |
+
],
|
| 1924 |
+
"modality_keys": [
|
| 1925 |
+
"joint_position",
|
| 1926 |
+
"gripper_position"
|
| 1927 |
+
],
|
| 1928 |
+
"sin_cos_embedding_keys": null,
|
| 1929 |
+
"mean_std_embedding_keys": null,
|
| 1930 |
+
"action_configs": null
|
| 1931 |
+
},
|
| 1932 |
+
"action": {
|
| 1933 |
+
"delta_indices": [
|
| 1934 |
+
0,
|
| 1935 |
+
1,
|
| 1936 |
+
2,
|
| 1937 |
+
3,
|
| 1938 |
+
4,
|
| 1939 |
+
5,
|
| 1940 |
+
6,
|
| 1941 |
+
7,
|
| 1942 |
+
8,
|
| 1943 |
+
9,
|
| 1944 |
+
10,
|
| 1945 |
+
11,
|
| 1946 |
+
12,
|
| 1947 |
+
13,
|
| 1948 |
+
14,
|
| 1949 |
+
15
|
| 1950 |
+
],
|
| 1951 |
+
"modality_keys": [
|
| 1952 |
+
"end_effector_position",
|
| 1953 |
+
"end_effector_rotation",
|
| 1954 |
+
"gripper_close"
|
| 1955 |
+
],
|
| 1956 |
+
"sin_cos_embedding_keys": null,
|
| 1957 |
+
"mean_std_embedding_keys": null,
|
| 1958 |
+
"action_configs": [
|
| 1959 |
+
{
|
| 1960 |
+
"rep": "DELTA",
|
| 1961 |
+
"type": "EEF",
|
| 1962 |
+
"format": "DEFAULT",
|
| 1963 |
+
"state_key": null
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"rep": "DELTA",
|
| 1967 |
+
"type": "EEF",
|
| 1968 |
+
"format": "DEFAULT",
|
| 1969 |
+
"state_key": null
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"rep": "ABSOLUTE",
|
| 1973 |
+
"type": "NON_EEF",
|
| 1974 |
+
"format": "DEFAULT",
|
| 1975 |
+
"state_key": null
|
| 1976 |
+
}
|
| 1977 |
+
]
|
| 1978 |
+
},
|
| 1979 |
+
"language": {
|
| 1980 |
+
"delta_indices": [
|
| 1981 |
+
0
|
| 1982 |
+
],
|
| 1983 |
+
"modality_keys": [
|
| 1984 |
+
"annotation.human.action.task_description"
|
| 1985 |
+
],
|
| 1986 |
+
"sin_cos_embedding_keys": null,
|
| 1987 |
+
"mean_std_embedding_keys": null,
|
| 1988 |
+
"action_configs": null
|
| 1989 |
+
}
|
| 1990 |
+
},
|
| 1991 |
+
"fmb_dataset": {
|
| 1992 |
+
"video": {
|
| 1993 |
+
"delta_indices": [
|
| 1994 |
+
0
|
| 1995 |
+
],
|
| 1996 |
+
"modality_keys": [
|
| 1997 |
+
"primary",
|
| 1998 |
+
"secondary",
|
| 1999 |
+
"wrist"
|
| 2000 |
+
],
|
| 2001 |
+
"sin_cos_embedding_keys": null,
|
| 2002 |
+
"mean_std_embedding_keys": null,
|
| 2003 |
+
"action_configs": null
|
| 2004 |
+
},
|
| 2005 |
+
"state": {
|
| 2006 |
+
"delta_indices": [
|
| 2007 |
+
0
|
| 2008 |
+
],
|
| 2009 |
+
"modality_keys": [
|
| 2010 |
+
"end_effector_position",
|
| 2011 |
+
"end_effector_rotation",
|
| 2012 |
+
"gripper_position"
|
| 2013 |
+
],
|
| 2014 |
+
"sin_cos_embedding_keys": null,
|
| 2015 |
+
"mean_std_embedding_keys": null,
|
| 2016 |
+
"action_configs": null
|
| 2017 |
+
},
|
| 2018 |
+
"action": {
|
| 2019 |
+
"delta_indices": [
|
| 2020 |
+
0,
|
| 2021 |
+
1,
|
| 2022 |
+
2,
|
| 2023 |
+
3,
|
| 2024 |
+
4,
|
| 2025 |
+
5,
|
| 2026 |
+
6,
|
| 2027 |
+
7,
|
| 2028 |
+
8,
|
| 2029 |
+
9,
|
| 2030 |
+
10,
|
| 2031 |
+
11,
|
| 2032 |
+
12,
|
| 2033 |
+
13,
|
| 2034 |
+
14,
|
| 2035 |
+
15
|
| 2036 |
+
],
|
| 2037 |
+
"modality_keys": [
|
| 2038 |
+
"end_effector_position",
|
| 2039 |
+
"end_effector_rotation",
|
| 2040 |
+
"gripper_close"
|
| 2041 |
+
],
|
| 2042 |
+
"sin_cos_embedding_keys": null,
|
| 2043 |
+
"mean_std_embedding_keys": null,
|
| 2044 |
+
"action_configs": [
|
| 2045 |
+
{
|
| 2046 |
+
"rep": "DELTA",
|
| 2047 |
+
"type": "EEF",
|
| 2048 |
+
"format": "DEFAULT",
|
| 2049 |
+
"state_key": null
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"rep": "DELTA",
|
| 2053 |
+
"type": "EEF",
|
| 2054 |
+
"format": "DEFAULT",
|
| 2055 |
+
"state_key": null
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"rep": "ABSOLUTE",
|
| 2059 |
+
"type": "NON_EEF",
|
| 2060 |
+
"format": "DEFAULT",
|
| 2061 |
+
"state_key": null
|
| 2062 |
+
}
|
| 2063 |
+
]
|
| 2064 |
+
},
|
| 2065 |
+
"language": {
|
| 2066 |
+
"delta_indices": [
|
| 2067 |
+
0
|
| 2068 |
+
],
|
| 2069 |
+
"modality_keys": [
|
| 2070 |
+
"annotation.human.action.task_description"
|
| 2071 |
+
],
|
| 2072 |
+
"sin_cos_embedding_keys": null,
|
| 2073 |
+
"mean_std_embedding_keys": null,
|
| 2074 |
+
"action_configs": null
|
| 2075 |
+
}
|
| 2076 |
+
},
|
| 2077 |
+
"bridge_orig": {
|
| 2078 |
+
"video": {
|
| 2079 |
+
"delta_indices": [
|
| 2080 |
+
0
|
| 2081 |
+
],
|
| 2082 |
+
"modality_keys": [
|
| 2083 |
+
"primary",
|
| 2084 |
+
"secondary"
|
| 2085 |
+
],
|
| 2086 |
+
"sin_cos_embedding_keys": null,
|
| 2087 |
+
"mean_std_embedding_keys": null,
|
| 2088 |
+
"action_configs": null
|
| 2089 |
+
},
|
| 2090 |
+
"state": {
|
| 2091 |
+
"delta_indices": [
|
| 2092 |
+
0
|
| 2093 |
+
],
|
| 2094 |
+
"modality_keys": [
|
| 2095 |
+
"end_effector_position",
|
| 2096 |
+
"end_effector_rotation",
|
| 2097 |
+
"gripper_position"
|
| 2098 |
+
],
|
| 2099 |
+
"sin_cos_embedding_keys": null,
|
| 2100 |
+
"mean_std_embedding_keys": null,
|
| 2101 |
+
"action_configs": null
|
| 2102 |
+
},
|
| 2103 |
+
"action": {
|
| 2104 |
+
"delta_indices": [
|
| 2105 |
+
0,
|
| 2106 |
+
1,
|
| 2107 |
+
2,
|
| 2108 |
+
3,
|
| 2109 |
+
4,
|
| 2110 |
+
5,
|
| 2111 |
+
6,
|
| 2112 |
+
7,
|
| 2113 |
+
8,
|
| 2114 |
+
9,
|
| 2115 |
+
10,
|
| 2116 |
+
11,
|
| 2117 |
+
12,
|
| 2118 |
+
13,
|
| 2119 |
+
14,
|
| 2120 |
+
15
|
| 2121 |
+
],
|
| 2122 |
+
"modality_keys": [
|
| 2123 |
+
"end_effector_position",
|
| 2124 |
+
"end_effector_rotation",
|
| 2125 |
+
"gripper_close"
|
| 2126 |
+
],
|
| 2127 |
+
"sin_cos_embedding_keys": null,
|
| 2128 |
+
"mean_std_embedding_keys": null,
|
| 2129 |
+
"action_configs": [
|
| 2130 |
+
{
|
| 2131 |
+
"rep": "DELTA",
|
| 2132 |
+
"type": "EEF",
|
| 2133 |
+
"format": "DEFAULT",
|
| 2134 |
+
"state_key": null
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"rep": "DELTA",
|
| 2138 |
+
"type": "EEF",
|
| 2139 |
+
"format": "DEFAULT",
|
| 2140 |
+
"state_key": null
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"rep": "ABSOLUTE",
|
| 2144 |
+
"type": "NON_EEF",
|
| 2145 |
+
"format": "DEFAULT",
|
| 2146 |
+
"state_key": null
|
| 2147 |
+
}
|
| 2148 |
+
]
|
| 2149 |
+
},
|
| 2150 |
+
"language": {
|
| 2151 |
+
"delta_indices": [
|
| 2152 |
+
0
|
| 2153 |
+
],
|
| 2154 |
+
"modality_keys": [
|
| 2155 |
+
"annotation.human.action.task_description"
|
| 2156 |
+
],
|
| 2157 |
+
"sin_cos_embedding_keys": null,
|
| 2158 |
+
"mean_std_embedding_keys": null,
|
| 2159 |
+
"action_configs": null
|
| 2160 |
+
}
|
| 2161 |
+
},
|
| 2162 |
+
"austin_buds_dataset_converted_externally_to_rlds": {
|
| 2163 |
+
"video": {
|
| 2164 |
+
"delta_indices": [
|
| 2165 |
+
0
|
| 2166 |
+
],
|
| 2167 |
+
"modality_keys": [
|
| 2168 |
+
"primary",
|
| 2169 |
+
"wrist"
|
| 2170 |
+
],
|
| 2171 |
+
"sin_cos_embedding_keys": null,
|
| 2172 |
+
"mean_std_embedding_keys": null,
|
| 2173 |
+
"action_configs": null
|
| 2174 |
+
},
|
| 2175 |
+
"state": {
|
| 2176 |
+
"delta_indices": [
|
| 2177 |
+
0
|
| 2178 |
+
],
|
| 2179 |
+
"modality_keys": [
|
| 2180 |
+
"joint_position",
|
| 2181 |
+
"gripper_position"
|
| 2182 |
+
],
|
| 2183 |
+
"sin_cos_embedding_keys": null,
|
| 2184 |
+
"mean_std_embedding_keys": null,
|
| 2185 |
+
"action_configs": null
|
| 2186 |
+
},
|
| 2187 |
+
"action": {
|
| 2188 |
+
"delta_indices": [
|
| 2189 |
+
0,
|
| 2190 |
+
1,
|
| 2191 |
+
2,
|
| 2192 |
+
3,
|
| 2193 |
+
4,
|
| 2194 |
+
5,
|
| 2195 |
+
6,
|
| 2196 |
+
7,
|
| 2197 |
+
8,
|
| 2198 |
+
9,
|
| 2199 |
+
10,
|
| 2200 |
+
11,
|
| 2201 |
+
12,
|
| 2202 |
+
13,
|
| 2203 |
+
14,
|
| 2204 |
+
15
|
| 2205 |
+
],
|
| 2206 |
+
"modality_keys": [
|
| 2207 |
+
"end_effector_position",
|
| 2208 |
+
"end_effector_rotation",
|
| 2209 |
+
"gripper_close"
|
| 2210 |
+
],
|
| 2211 |
+
"sin_cos_embedding_keys": null,
|
| 2212 |
+
"mean_std_embedding_keys": null,
|
| 2213 |
+
"action_configs": [
|
| 2214 |
+
{
|
| 2215 |
+
"rep": "DELTA",
|
| 2216 |
+
"type": "EEF",
|
| 2217 |
+
"format": "DEFAULT",
|
| 2218 |
+
"state_key": null
|
| 2219 |
+
},
|
| 2220 |
+
{
|
| 2221 |
+
"rep": "DELTA",
|
| 2222 |
+
"type": "EEF",
|
| 2223 |
+
"format": "DEFAULT",
|
| 2224 |
+
"state_key": null
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"rep": "ABSOLUTE",
|
| 2228 |
+
"type": "NON_EEF",
|
| 2229 |
+
"format": "DEFAULT",
|
| 2230 |
+
"state_key": null
|
| 2231 |
+
}
|
| 2232 |
+
]
|
| 2233 |
+
},
|
| 2234 |
+
"language": {
|
| 2235 |
+
"delta_indices": [
|
| 2236 |
+
0
|
| 2237 |
+
],
|
| 2238 |
+
"modality_keys": [
|
| 2239 |
+
"annotation.human.action.task_description"
|
| 2240 |
+
],
|
| 2241 |
+
"sin_cos_embedding_keys": null,
|
| 2242 |
+
"mean_std_embedding_keys": null,
|
| 2243 |
+
"action_configs": null
|
| 2244 |
+
}
|
| 2245 |
+
},
|
| 2246 |
+
"neural_gr1": {
|
| 2247 |
+
"video": {
|
| 2248 |
+
"delta_indices": [
|
| 2249 |
+
0
|
| 2250 |
+
],
|
| 2251 |
+
"modality_keys": [
|
| 2252 |
+
"ego_view"
|
| 2253 |
+
],
|
| 2254 |
+
"sin_cos_embedding_keys": null,
|
| 2255 |
+
"mean_std_embedding_keys": null,
|
| 2256 |
+
"action_configs": null
|
| 2257 |
+
},
|
| 2258 |
+
"state": {
|
| 2259 |
+
"delta_indices": [
|
| 2260 |
+
0
|
| 2261 |
+
],
|
| 2262 |
+
"modality_keys": [
|
| 2263 |
+
"left_arm",
|
| 2264 |
+
"left_hand",
|
| 2265 |
+
"left_leg",
|
| 2266 |
+
"neck",
|
| 2267 |
+
"right_arm",
|
| 2268 |
+
"right_hand",
|
| 2269 |
+
"right_leg",
|
| 2270 |
+
"waist"
|
| 2271 |
+
],
|
| 2272 |
+
"sin_cos_embedding_keys": null,
|
| 2273 |
+
"mean_std_embedding_keys": null,
|
| 2274 |
+
"action_configs": null
|
| 2275 |
+
},
|
| 2276 |
+
"action": {
|
| 2277 |
+
"delta_indices": [
|
| 2278 |
+
0,
|
| 2279 |
+
1,
|
| 2280 |
+
2,
|
| 2281 |
+
3,
|
| 2282 |
+
4,
|
| 2283 |
+
5,
|
| 2284 |
+
6,
|
| 2285 |
+
7,
|
| 2286 |
+
8,
|
| 2287 |
+
9,
|
| 2288 |
+
10,
|
| 2289 |
+
11,
|
| 2290 |
+
12,
|
| 2291 |
+
13,
|
| 2292 |
+
14,
|
| 2293 |
+
15
|
| 2294 |
+
],
|
| 2295 |
+
"modality_keys": [
|
| 2296 |
+
"left_arm",
|
| 2297 |
+
"left_hand",
|
| 2298 |
+
"left_leg",
|
| 2299 |
+
"neck",
|
| 2300 |
+
"right_arm",
|
| 2301 |
+
"right_hand",
|
| 2302 |
+
"right_leg",
|
| 2303 |
+
"waist"
|
| 2304 |
+
],
|
| 2305 |
+
"sin_cos_embedding_keys": null,
|
| 2306 |
+
"mean_std_embedding_keys": null,
|
| 2307 |
+
"action_configs": [
|
| 2308 |
+
{
|
| 2309 |
+
"rep": "ABSOLUTE",
|
| 2310 |
+
"type": "NON_EEF",
|
| 2311 |
+
"format": "DEFAULT",
|
| 2312 |
+
"state_key": null
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"rep": "ABSOLUTE",
|
| 2316 |
+
"type": "NON_EEF",
|
| 2317 |
+
"format": "DEFAULT",
|
| 2318 |
+
"state_key": null
|
| 2319 |
+
},
|
| 2320 |
+
{
|
| 2321 |
+
"rep": "ABSOLUTE",
|
| 2322 |
+
"type": "NON_EEF",
|
| 2323 |
+
"format": "DEFAULT",
|
| 2324 |
+
"state_key": null
|
| 2325 |
+
},
|
| 2326 |
+
{
|
| 2327 |
+
"rep": "ABSOLUTE",
|
| 2328 |
+
"type": "NON_EEF",
|
| 2329 |
+
"format": "DEFAULT",
|
| 2330 |
+
"state_key": null
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"rep": "ABSOLUTE",
|
| 2334 |
+
"type": "NON_EEF",
|
| 2335 |
+
"format": "DEFAULT",
|
| 2336 |
+
"state_key": null
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"rep": "ABSOLUTE",
|
| 2340 |
+
"type": "NON_EEF",
|
| 2341 |
+
"format": "DEFAULT",
|
| 2342 |
+
"state_key": null
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"rep": "ABSOLUTE",
|
| 2346 |
+
"type": "NON_EEF",
|
| 2347 |
+
"format": "DEFAULT",
|
| 2348 |
+
"state_key": null
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"rep": "ABSOLUTE",
|
| 2352 |
+
"type": "NON_EEF",
|
| 2353 |
+
"format": "DEFAULT",
|
| 2354 |
+
"state_key": null
|
| 2355 |
+
}
|
| 2356 |
+
]
|
| 2357 |
+
},
|
| 2358 |
+
"language": {
|
| 2359 |
+
"delta_indices": [
|
| 2360 |
+
0
|
| 2361 |
+
],
|
| 2362 |
+
"modality_keys": [
|
| 2363 |
+
"annotation.human.action.task_description"
|
| 2364 |
+
],
|
| 2365 |
+
"sin_cos_embedding_keys": null,
|
| 2366 |
+
"mean_std_embedding_keys": null,
|
| 2367 |
+
"action_configs": null
|
| 2368 |
+
}
|
| 2369 |
+
},
|
| 2370 |
+
"galaxea": {
|
| 2371 |
+
"video": {
|
| 2372 |
+
"delta_indices": [
|
| 2373 |
+
0
|
| 2374 |
+
],
|
| 2375 |
+
"modality_keys": [
|
| 2376 |
+
"primary",
|
| 2377 |
+
"wrist_left",
|
| 2378 |
+
"wrist_right"
|
| 2379 |
+
],
|
| 2380 |
+
"sin_cos_embedding_keys": null,
|
| 2381 |
+
"mean_std_embedding_keys": null,
|
| 2382 |
+
"action_configs": null
|
| 2383 |
+
},
|
| 2384 |
+
"state": {
|
| 2385 |
+
"delta_indices": [
|
| 2386 |
+
0
|
| 2387 |
+
],
|
| 2388 |
+
"modality_keys": [
|
| 2389 |
+
"state"
|
| 2390 |
+
],
|
| 2391 |
+
"sin_cos_embedding_keys": null,
|
| 2392 |
+
"mean_std_embedding_keys": null,
|
| 2393 |
+
"action_configs": null
|
| 2394 |
+
},
|
| 2395 |
+
"action": {
|
| 2396 |
+
"delta_indices": [
|
| 2397 |
+
0,
|
| 2398 |
+
1,
|
| 2399 |
+
2,
|
| 2400 |
+
3,
|
| 2401 |
+
4,
|
| 2402 |
+
5,
|
| 2403 |
+
6,
|
| 2404 |
+
7,
|
| 2405 |
+
8,
|
| 2406 |
+
9,
|
| 2407 |
+
10,
|
| 2408 |
+
11,
|
| 2409 |
+
12,
|
| 2410 |
+
13,
|
| 2411 |
+
14,
|
| 2412 |
+
15
|
| 2413 |
+
],
|
| 2414 |
+
"modality_keys": [
|
| 2415 |
+
"action"
|
| 2416 |
+
],
|
| 2417 |
+
"sin_cos_embedding_keys": null,
|
| 2418 |
+
"mean_std_embedding_keys": null,
|
| 2419 |
+
"action_configs": [
|
| 2420 |
+
{
|
| 2421 |
+
"rep": "ABSOLUTE",
|
| 2422 |
+
"type": "NON_EEF",
|
| 2423 |
+
"format": "DEFAULT",
|
| 2424 |
+
"state_key": null
|
| 2425 |
+
}
|
| 2426 |
+
]
|
| 2427 |
+
},
|
| 2428 |
+
"language": {
|
| 2429 |
+
"delta_indices": [
|
| 2430 |
+
0
|
| 2431 |
+
],
|
| 2432 |
+
"modality_keys": [
|
| 2433 |
+
"annotation.human.action.task_description"
|
| 2434 |
+
],
|
| 2435 |
+
"sin_cos_embedding_keys": null,
|
| 2436 |
+
"mean_std_embedding_keys": null,
|
| 2437 |
+
"action_configs": null
|
| 2438 |
+
}
|
| 2439 |
+
},
|
| 2440 |
+
"roboturk": {
|
| 2441 |
+
"video": {
|
| 2442 |
+
"delta_indices": [
|
| 2443 |
+
0
|
| 2444 |
+
],
|
| 2445 |
+
"modality_keys": [
|
| 2446 |
+
"primary"
|
| 2447 |
+
],
|
| 2448 |
+
"sin_cos_embedding_keys": null,
|
| 2449 |
+
"mean_std_embedding_keys": null,
|
| 2450 |
+
"action_configs": null
|
| 2451 |
+
},
|
| 2452 |
+
"state": {
|
| 2453 |
+
"delta_indices": [
|
| 2454 |
+
0
|
| 2455 |
+
],
|
| 2456 |
+
"modality_keys": [
|
| 2457 |
+
"none"
|
| 2458 |
+
],
|
| 2459 |
+
"sin_cos_embedding_keys": null,
|
| 2460 |
+
"mean_std_embedding_keys": null,
|
| 2461 |
+
"action_configs": null
|
| 2462 |
+
},
|
| 2463 |
+
"action": {
|
| 2464 |
+
"delta_indices": [
|
| 2465 |
+
0,
|
| 2466 |
+
1,
|
| 2467 |
+
2,
|
| 2468 |
+
3,
|
| 2469 |
+
4,
|
| 2470 |
+
5,
|
| 2471 |
+
6,
|
| 2472 |
+
7,
|
| 2473 |
+
8,
|
| 2474 |
+
9,
|
| 2475 |
+
10,
|
| 2476 |
+
11,
|
| 2477 |
+
12,
|
| 2478 |
+
13,
|
| 2479 |
+
14,
|
| 2480 |
+
15
|
| 2481 |
+
],
|
| 2482 |
+
"modality_keys": [
|
| 2483 |
+
"end_effector_position",
|
| 2484 |
+
"end_effector_rotation",
|
| 2485 |
+
"gripper_close"
|
| 2486 |
+
],
|
| 2487 |
+
"sin_cos_embedding_keys": null,
|
| 2488 |
+
"mean_std_embedding_keys": null,
|
| 2489 |
+
"action_configs": [
|
| 2490 |
+
{
|
| 2491 |
+
"rep": "DELTA",
|
| 2492 |
+
"type": "EEF",
|
| 2493 |
+
"format": "DEFAULT",
|
| 2494 |
+
"state_key": null
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"rep": "DELTA",
|
| 2498 |
+
"type": "EEF",
|
| 2499 |
+
"format": "DEFAULT",
|
| 2500 |
+
"state_key": null
|
| 2501 |
+
},
|
| 2502 |
+
{
|
| 2503 |
+
"rep": "ABSOLUTE",
|
| 2504 |
+
"type": "NON_EEF",
|
| 2505 |
+
"format": "DEFAULT",
|
| 2506 |
+
"state_key": null
|
| 2507 |
+
}
|
| 2508 |
+
]
|
| 2509 |
+
},
|
| 2510 |
+
"language": {
|
| 2511 |
+
"delta_indices": [
|
| 2512 |
+
0
|
| 2513 |
+
],
|
| 2514 |
+
"modality_keys": [
|
| 2515 |
+
"annotation.human.action.task_description"
|
| 2516 |
+
],
|
| 2517 |
+
"sin_cos_embedding_keys": null,
|
| 2518 |
+
"mean_std_embedding_keys": null,
|
| 2519 |
+
"action_configs": null
|
| 2520 |
+
}
|
| 2521 |
+
},
|
| 2522 |
+
"berkeley_cable_routing": {
|
| 2523 |
+
"video": {
|
| 2524 |
+
"delta_indices": [
|
| 2525 |
+
0
|
| 2526 |
+
],
|
| 2527 |
+
"modality_keys": [
|
| 2528 |
+
"primary",
|
| 2529 |
+
"secondary",
|
| 2530 |
+
"wrist"
|
| 2531 |
+
],
|
| 2532 |
+
"sin_cos_embedding_keys": null,
|
| 2533 |
+
"mean_std_embedding_keys": null,
|
| 2534 |
+
"action_configs": null
|
| 2535 |
+
},
|
| 2536 |
+
"state": {
|
| 2537 |
+
"delta_indices": [
|
| 2538 |
+
0
|
| 2539 |
+
],
|
| 2540 |
+
"modality_keys": [
|
| 2541 |
+
"joint_position"
|
| 2542 |
+
],
|
| 2543 |
+
"sin_cos_embedding_keys": null,
|
| 2544 |
+
"mean_std_embedding_keys": null,
|
| 2545 |
+
"action_configs": null
|
| 2546 |
+
},
|
| 2547 |
+
"action": {
|
| 2548 |
+
"delta_indices": [
|
| 2549 |
+
0,
|
| 2550 |
+
1,
|
| 2551 |
+
2,
|
| 2552 |
+
3,
|
| 2553 |
+
4,
|
| 2554 |
+
5,
|
| 2555 |
+
6,
|
| 2556 |
+
7,
|
| 2557 |
+
8,
|
| 2558 |
+
9,
|
| 2559 |
+
10,
|
| 2560 |
+
11,
|
| 2561 |
+
12,
|
| 2562 |
+
13,
|
| 2563 |
+
14,
|
| 2564 |
+
15
|
| 2565 |
+
],
|
| 2566 |
+
"modality_keys": [
|
| 2567 |
+
"end_effector_position",
|
| 2568 |
+
"end_effector_rotation",
|
| 2569 |
+
"gripper_close"
|
| 2570 |
+
],
|
| 2571 |
+
"sin_cos_embedding_keys": null,
|
| 2572 |
+
"mean_std_embedding_keys": null,
|
| 2573 |
+
"action_configs": [
|
| 2574 |
+
{
|
| 2575 |
+
"rep": "DELTA",
|
| 2576 |
+
"type": "EEF",
|
| 2577 |
+
"format": "DEFAULT",
|
| 2578 |
+
"state_key": null
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"rep": "DELTA",
|
| 2582 |
+
"type": "EEF",
|
| 2583 |
+
"format": "DEFAULT",
|
| 2584 |
+
"state_key": null
|
| 2585 |
+
},
|
| 2586 |
+
{
|
| 2587 |
+
"rep": "ABSOLUTE",
|
| 2588 |
+
"type": "NON_EEF",
|
| 2589 |
+
"format": "DEFAULT",
|
| 2590 |
+
"state_key": null
|
| 2591 |
+
}
|
| 2592 |
+
]
|
| 2593 |
+
},
|
| 2594 |
+
"language": {
|
| 2595 |
+
"delta_indices": [
|
| 2596 |
+
0
|
| 2597 |
+
],
|
| 2598 |
+
"modality_keys": [
|
| 2599 |
+
"annotation.human.action.task_description"
|
| 2600 |
+
],
|
| 2601 |
+
"sin_cos_embedding_keys": null,
|
| 2602 |
+
"mean_std_embedding_keys": null,
|
| 2603 |
+
"action_configs": null
|
| 2604 |
+
}
|
| 2605 |
+
},
|
| 2606 |
+
"action_net": {
|
| 2607 |
+
"video": {
|
| 2608 |
+
"delta_indices": [
|
| 2609 |
+
0
|
| 2610 |
+
],
|
| 2611 |
+
"modality_keys": [
|
| 2612 |
+
"primary"
|
| 2613 |
+
],
|
| 2614 |
+
"sin_cos_embedding_keys": null,
|
| 2615 |
+
"mean_std_embedding_keys": null,
|
| 2616 |
+
"action_configs": null
|
| 2617 |
+
},
|
| 2618 |
+
"state": {
|
| 2619 |
+
"delta_indices": [
|
| 2620 |
+
0
|
| 2621 |
+
],
|
| 2622 |
+
"modality_keys": [
|
| 2623 |
+
"state"
|
| 2624 |
+
],
|
| 2625 |
+
"sin_cos_embedding_keys": null,
|
| 2626 |
+
"mean_std_embedding_keys": null,
|
| 2627 |
+
"action_configs": null
|
| 2628 |
+
},
|
| 2629 |
+
"action": {
|
| 2630 |
+
"delta_indices": [
|
| 2631 |
+
0,
|
| 2632 |
+
1,
|
| 2633 |
+
2,
|
| 2634 |
+
3,
|
| 2635 |
+
4,
|
| 2636 |
+
5,
|
| 2637 |
+
6,
|
| 2638 |
+
7,
|
| 2639 |
+
8,
|
| 2640 |
+
9,
|
| 2641 |
+
10,
|
| 2642 |
+
11,
|
| 2643 |
+
12,
|
| 2644 |
+
13,
|
| 2645 |
+
14,
|
| 2646 |
+
15
|
| 2647 |
+
],
|
| 2648 |
+
"modality_keys": [
|
| 2649 |
+
"action"
|
| 2650 |
+
],
|
| 2651 |
+
"sin_cos_embedding_keys": null,
|
| 2652 |
+
"mean_std_embedding_keys": null,
|
| 2653 |
+
"action_configs": [
|
| 2654 |
+
{
|
| 2655 |
+
"rep": "ABSOLUTE",
|
| 2656 |
+
"type": "NON_EEF",
|
| 2657 |
+
"format": "DEFAULT",
|
| 2658 |
+
"state_key": null
|
| 2659 |
+
}
|
| 2660 |
+
]
|
| 2661 |
+
},
|
| 2662 |
+
"language": {
|
| 2663 |
+
"delta_indices": [
|
| 2664 |
+
0
|
| 2665 |
+
],
|
| 2666 |
+
"modality_keys": [
|
| 2667 |
+
"annotation.human.action.task_description"
|
| 2668 |
+
],
|
| 2669 |
+
"sin_cos_embedding_keys": null,
|
| 2670 |
+
"mean_std_embedding_keys": null,
|
| 2671 |
+
"action_configs": null
|
| 2672 |
+
}
|
| 2673 |
+
},
|
| 2674 |
+
"kuka": {
|
| 2675 |
+
"video": {
|
| 2676 |
+
"delta_indices": [
|
| 2677 |
+
0
|
| 2678 |
+
],
|
| 2679 |
+
"modality_keys": [
|
| 2680 |
+
"primary"
|
| 2681 |
+
],
|
| 2682 |
+
"sin_cos_embedding_keys": null,
|
| 2683 |
+
"mean_std_embedding_keys": null,
|
| 2684 |
+
"action_configs": null
|
| 2685 |
+
},
|
| 2686 |
+
"state": {
|
| 2687 |
+
"delta_indices": [
|
| 2688 |
+
0
|
| 2689 |
+
],
|
| 2690 |
+
"modality_keys": [
|
| 2691 |
+
"end_effector_position",
|
| 2692 |
+
"end_effector_rotation",
|
| 2693 |
+
"gripper_position"
|
| 2694 |
+
],
|
| 2695 |
+
"sin_cos_embedding_keys": null,
|
| 2696 |
+
"mean_std_embedding_keys": null,
|
| 2697 |
+
"action_configs": null
|
| 2698 |
+
},
|
| 2699 |
+
"action": {
|
| 2700 |
+
"delta_indices": [
|
| 2701 |
+
0,
|
| 2702 |
+
1,
|
| 2703 |
+
2,
|
| 2704 |
+
3,
|
| 2705 |
+
4,
|
| 2706 |
+
5,
|
| 2707 |
+
6,
|
| 2708 |
+
7,
|
| 2709 |
+
8,
|
| 2710 |
+
9,
|
| 2711 |
+
10,
|
| 2712 |
+
11,
|
| 2713 |
+
12,
|
| 2714 |
+
13,
|
| 2715 |
+
14,
|
| 2716 |
+
15
|
| 2717 |
+
],
|
| 2718 |
+
"modality_keys": [
|
| 2719 |
+
"end_effector_position",
|
| 2720 |
+
"end_effector_rotation",
|
| 2721 |
+
"gripper_close"
|
| 2722 |
+
],
|
| 2723 |
+
"sin_cos_embedding_keys": null,
|
| 2724 |
+
"mean_std_embedding_keys": null,
|
| 2725 |
+
"action_configs": [
|
| 2726 |
+
{
|
| 2727 |
+
"rep": "DELTA",
|
| 2728 |
+
"type": "EEF",
|
| 2729 |
+
"format": "DEFAULT",
|
| 2730 |
+
"state_key": null
|
| 2731 |
+
},
|
| 2732 |
+
{
|
| 2733 |
+
"rep": "DELTA",
|
| 2734 |
+
"type": "EEF",
|
| 2735 |
+
"format": "DEFAULT",
|
| 2736 |
+
"state_key": null
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"rep": "ABSOLUTE",
|
| 2740 |
+
"type": "NON_EEF",
|
| 2741 |
+
"format": "DEFAULT",
|
| 2742 |
+
"state_key": null
|
| 2743 |
+
}
|
| 2744 |
+
]
|
| 2745 |
+
},
|
| 2746 |
+
"language": {
|
| 2747 |
+
"delta_indices": [
|
| 2748 |
+
0
|
| 2749 |
+
],
|
| 2750 |
+
"modality_keys": [
|
| 2751 |
+
"annotation.human.action.task_description"
|
| 2752 |
+
],
|
| 2753 |
+
"sin_cos_embedding_keys": null,
|
| 2754 |
+
"mean_std_embedding_keys": null,
|
| 2755 |
+
"action_configs": null
|
| 2756 |
+
}
|
| 2757 |
+
},
|
| 2758 |
+
"humanoid_everyday_h1": {
|
| 2759 |
+
"video": {
|
| 2760 |
+
"delta_indices": [
|
| 2761 |
+
0
|
| 2762 |
+
],
|
| 2763 |
+
"modality_keys": [
|
| 2764 |
+
"egocentric_resized"
|
| 2765 |
+
],
|
| 2766 |
+
"sin_cos_embedding_keys": null,
|
| 2767 |
+
"mean_std_embedding_keys": null,
|
| 2768 |
+
"action_configs": null
|
| 2769 |
+
},
|
| 2770 |
+
"state": {
|
| 2771 |
+
"delta_indices": [
|
| 2772 |
+
0
|
| 2773 |
+
],
|
| 2774 |
+
"modality_keys": [
|
| 2775 |
+
"left_arm",
|
| 2776 |
+
"left_hand",
|
| 2777 |
+
"right_arm",
|
| 2778 |
+
"right_hand"
|
| 2779 |
+
],
|
| 2780 |
+
"sin_cos_embedding_keys": null,
|
| 2781 |
+
"mean_std_embedding_keys": null,
|
| 2782 |
+
"action_configs": null
|
| 2783 |
+
},
|
| 2784 |
+
"action": {
|
| 2785 |
+
"delta_indices": [
|
| 2786 |
+
0,
|
| 2787 |
+
1,
|
| 2788 |
+
2,
|
| 2789 |
+
3,
|
| 2790 |
+
4,
|
| 2791 |
+
5,
|
| 2792 |
+
6,
|
| 2793 |
+
7,
|
| 2794 |
+
8,
|
| 2795 |
+
9,
|
| 2796 |
+
10,
|
| 2797 |
+
11,
|
| 2798 |
+
12,
|
| 2799 |
+
13,
|
| 2800 |
+
14,
|
| 2801 |
+
15
|
| 2802 |
+
],
|
| 2803 |
+
"modality_keys": [
|
| 2804 |
+
"left_arm",
|
| 2805 |
+
"left_hand",
|
| 2806 |
+
"right_arm",
|
| 2807 |
+
"right_hand"
|
| 2808 |
+
],
|
| 2809 |
+
"sin_cos_embedding_keys": null,
|
| 2810 |
+
"mean_std_embedding_keys": null,
|
| 2811 |
+
"action_configs": [
|
| 2812 |
+
{
|
| 2813 |
+
"rep": "ABSOLUTE",
|
| 2814 |
+
"type": "NON_EEF",
|
| 2815 |
+
"format": "DEFAULT",
|
| 2816 |
+
"state_key": null
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"rep": "ABSOLUTE",
|
| 2820 |
+
"type": "NON_EEF",
|
| 2821 |
+
"format": "DEFAULT",
|
| 2822 |
+
"state_key": null
|
| 2823 |
+
},
|
| 2824 |
+
{
|
| 2825 |
+
"rep": "ABSOLUTE",
|
| 2826 |
+
"type": "NON_EEF",
|
| 2827 |
+
"format": "DEFAULT",
|
| 2828 |
+
"state_key": null
|
| 2829 |
+
},
|
| 2830 |
+
{
|
| 2831 |
+
"rep": "ABSOLUTE",
|
| 2832 |
+
"type": "NON_EEF",
|
| 2833 |
+
"format": "DEFAULT",
|
| 2834 |
+
"state_key": null
|
| 2835 |
+
}
|
| 2836 |
+
]
|
| 2837 |
+
},
|
| 2838 |
+
"language": {
|
| 2839 |
+
"delta_indices": [
|
| 2840 |
+
0
|
| 2841 |
+
],
|
| 2842 |
+
"modality_keys": [
|
| 2843 |
+
"annotation.human.action.task_description"
|
| 2844 |
+
],
|
| 2845 |
+
"sin_cos_embedding_keys": null,
|
| 2846 |
+
"mean_std_embedding_keys": null,
|
| 2847 |
+
"action_configs": null
|
| 2848 |
+
}
|
| 2849 |
+
},
|
| 2850 |
+
"general_embodiment": {
|
| 2851 |
+
"video": {
|
| 2852 |
+
"delta_indices": [
|
| 2853 |
+
0
|
| 2854 |
+
],
|
| 2855 |
+
"modality_keys": [
|
| 2856 |
+
"left_view",
|
| 2857 |
+
"right_view",
|
| 2858 |
+
"wrist_view"
|
| 2859 |
+
],
|
| 2860 |
+
"sin_cos_embedding_keys": null,
|
| 2861 |
+
"mean_std_embedding_keys": null,
|
| 2862 |
+
"action_configs": null
|
| 2863 |
+
},
|
| 2864 |
+
"state": {
|
| 2865 |
+
"delta_indices": [
|
| 2866 |
+
0
|
| 2867 |
+
],
|
| 2868 |
+
"modality_keys": [
|
| 2869 |
+
"end_effector_position_relative",
|
| 2870 |
+
"end_effector_rotation_relative",
|
| 2871 |
+
"gripper_qpos",
|
| 2872 |
+
"base_position",
|
| 2873 |
+
"base_rotation"
|
| 2874 |
+
],
|
| 2875 |
+
"sin_cos_embedding_keys": null,
|
| 2876 |
+
"mean_std_embedding_keys": null,
|
| 2877 |
+
"action_configs": null
|
| 2878 |
+
},
|
| 2879 |
+
"action": {
|
| 2880 |
+
"delta_indices": [
|
| 2881 |
+
0,
|
| 2882 |
+
1,
|
| 2883 |
+
2,
|
| 2884 |
+
3,
|
| 2885 |
+
4,
|
| 2886 |
+
5,
|
| 2887 |
+
6,
|
| 2888 |
+
7,
|
| 2889 |
+
8,
|
| 2890 |
+
9,
|
| 2891 |
+
10,
|
| 2892 |
+
11,
|
| 2893 |
+
12,
|
| 2894 |
+
13,
|
| 2895 |
+
14,
|
| 2896 |
+
15
|
| 2897 |
+
],
|
| 2898 |
+
"modality_keys": [
|
| 2899 |
+
"end_effector_position",
|
| 2900 |
+
"end_effector_rotation",
|
| 2901 |
+
"gripper_close",
|
| 2902 |
+
"base_motion",
|
| 2903 |
+
"control_mode"
|
| 2904 |
+
],
|
| 2905 |
+
"sin_cos_embedding_keys": null,
|
| 2906 |
+
"mean_std_embedding_keys": null,
|
| 2907 |
+
"action_configs": [
|
| 2908 |
+
{
|
| 2909 |
+
"rep": "DELTA",
|
| 2910 |
+
"type": "EEF",
|
| 2911 |
+
"format": "DEFAULT",
|
| 2912 |
+
"state_key": null
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"rep": "DELTA",
|
| 2916 |
+
"type": "EEF",
|
| 2917 |
+
"format": "DEFAULT",
|
| 2918 |
+
"state_key": null
|
| 2919 |
+
},
|
| 2920 |
+
{
|
| 2921 |
+
"rep": "ABSOLUTE",
|
| 2922 |
+
"type": "NON_EEF",
|
| 2923 |
+
"format": "DEFAULT",
|
| 2924 |
+
"state_key": null
|
| 2925 |
+
},
|
| 2926 |
+
{
|
| 2927 |
+
"rep": "DELTA",
|
| 2928 |
+
"type": "NON_EEF",
|
| 2929 |
+
"format": "DEFAULT",
|
| 2930 |
+
"state_key": null
|
| 2931 |
+
},
|
| 2932 |
+
{
|
| 2933 |
+
"rep": "ABSOLUTE",
|
| 2934 |
+
"type": "NON_EEF",
|
| 2935 |
+
"format": "DEFAULT",
|
| 2936 |
+
"state_key": null
|
| 2937 |
+
}
|
| 2938 |
+
]
|
| 2939 |
+
},
|
| 2940 |
+
"language": {
|
| 2941 |
+
"delta_indices": [
|
| 2942 |
+
0
|
| 2943 |
+
],
|
| 2944 |
+
"modality_keys": [
|
| 2945 |
+
"annotation.human.action.task_description"
|
| 2946 |
+
],
|
| 2947 |
+
"sin_cos_embedding_keys": null,
|
| 2948 |
+
"mean_std_embedding_keys": null,
|
| 2949 |
+
"action_configs": null
|
| 2950 |
+
}
|
| 2951 |
+
}
|
| 2952 |
+
},
|
| 2953 |
+
"image_max_area": 65536,
|
| 2954 |
+
"image_resize_m": 32,
|
| 2955 |
+
"random_crop_fraction": null,
|
| 2956 |
+
"random_rotation_angle": null,
|
| 2957 |
+
"color_jitter_params": {
|
| 2958 |
+
"brightness": 0.3,
|
| 2959 |
+
"contrast": 0.4,
|
| 2960 |
+
"saturation": 0.5,
|
| 2961 |
+
"hue": 0.08
|
| 2962 |
+
},
|
| 2963 |
+
"model_name": "/data/taeyoung/models/RLDX-1-VLM",
|
| 2964 |
+
"model_type": "vtc_qwen3_vl",
|
| 2965 |
+
"formalize_language": true,
|
| 2966 |
+
"max_state_dim": 64,
|
| 2967 |
+
"max_action_dim": 64,
|
| 2968 |
+
"max_action_horizon": 16,
|
| 2969 |
+
"use_percentiles": true,
|
| 2970 |
+
"clip_outliers": true,
|
| 2971 |
+
"apply_sincos_state_encoding": false,
|
| 2972 |
+
"use_relative_action": true,
|
| 2973 |
+
"memory_length": 1,
|
| 2974 |
+
"general_embodiment_train_ratio": 0,
|
| 2975 |
+
"allow_missing_physics": false,
|
| 2976 |
+
"physics_keys": null,
|
| 2977 |
+
"physics_dims": null,
|
| 2978 |
+
"conversation_image_first": false
|
| 2979 |
+
}
|
| 2980 |
+
}
|
processor/statistics.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e154263f71e1bb31204674d171a0cbaa45c0645eafa0310a74bc99b6e8760a9
|
| 3 |
+
size 8017
|
wandb_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"project": "rldx-finetune", "run_id": "rldx1_image_ft_robocasa300_bsz1024_60k"}
|
zero_to_fp32.py
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python
|
| 2 |
+
|
| 3 |
+
# Copyright (c) Microsoft Corporation.
|
| 4 |
+
# SPDX-License-Identifier: Apache-2.0
|
| 5 |
+
|
| 6 |
+
# DeepSpeed Team
|
| 7 |
+
|
| 8 |
+
# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
|
| 9 |
+
# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
|
| 10 |
+
# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
|
| 11 |
+
# application.
|
| 12 |
+
#
|
| 13 |
+
# example:
|
| 14 |
+
# python zero_to_fp32.py . output_dir/
|
| 15 |
+
# or
|
| 16 |
+
# python zero_to_fp32.py . output_dir/ --safe_serialization
|
| 17 |
+
|
| 18 |
+
import argparse
|
| 19 |
+
import torch
|
| 20 |
+
import glob
|
| 21 |
+
import math
|
| 22 |
+
import os
|
| 23 |
+
import re
|
| 24 |
+
import gc
|
| 25 |
+
import json
|
| 26 |
+
import numpy as np
|
| 27 |
+
from tqdm import tqdm
|
| 28 |
+
from collections import OrderedDict
|
| 29 |
+
from dataclasses import dataclass
|
| 30 |
+
|
| 31 |
+
# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
|
| 32 |
+
# DeepSpeed data structures it has to be available in the current python environment.
|
| 33 |
+
from deepspeed.utils import logger
|
| 34 |
+
from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
|
| 35 |
+
FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
|
| 36 |
+
FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
@dataclass
|
| 40 |
+
class zero_model_state:
|
| 41 |
+
buffers: dict()
|
| 42 |
+
param_shapes: dict()
|
| 43 |
+
shared_params: list
|
| 44 |
+
ds_version: int
|
| 45 |
+
frozen_param_shapes: dict()
|
| 46 |
+
frozen_param_fragments: dict()
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
debug = 0
|
| 50 |
+
|
| 51 |
+
# load to cpu
|
| 52 |
+
device = torch.device('cpu')
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
def atoi(text):
|
| 56 |
+
return int(text) if text.isdigit() else text
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
def natural_keys(text):
|
| 60 |
+
'''
|
| 61 |
+
alist.sort(key=natural_keys) sorts in human order
|
| 62 |
+
http://nedbatchelder.com/blog/200712/human_sorting.html
|
| 63 |
+
(See Toothy's implementation in the comments)
|
| 64 |
+
'''
|
| 65 |
+
return [atoi(c) for c in re.split(r'(\d+)', text)]
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
def get_model_state_file(checkpoint_dir, zero_stage):
|
| 69 |
+
if not os.path.isdir(checkpoint_dir):
|
| 70 |
+
raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
|
| 71 |
+
|
| 72 |
+
# there should be only one file
|
| 73 |
+
if zero_stage <= 2:
|
| 74 |
+
file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
|
| 75 |
+
elif zero_stage == 3:
|
| 76 |
+
file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
|
| 77 |
+
|
| 78 |
+
if not os.path.exists(file):
|
| 79 |
+
raise FileNotFoundError(f"can't find model states file at '{file}'")
|
| 80 |
+
|
| 81 |
+
return file
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
def get_checkpoint_files(checkpoint_dir, glob_pattern):
|
| 85 |
+
# XXX: need to test that this simple glob rule works for multi-node setup too
|
| 86 |
+
ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
|
| 87 |
+
|
| 88 |
+
if len(ckpt_files) == 0:
|
| 89 |
+
raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
|
| 90 |
+
|
| 91 |
+
return ckpt_files
|
| 92 |
+
|
| 93 |
+
|
| 94 |
+
def get_optim_files(checkpoint_dir):
|
| 95 |
+
return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
|
| 96 |
+
|
| 97 |
+
|
| 98 |
+
def get_model_state_files(checkpoint_dir):
|
| 99 |
+
return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
def parse_model_states(files):
|
| 103 |
+
zero_model_states = []
|
| 104 |
+
for file in files:
|
| 105 |
+
state_dict = torch.load(file, map_location=device, weights_only=False)
|
| 106 |
+
|
| 107 |
+
if BUFFER_NAMES not in state_dict:
|
| 108 |
+
raise ValueError(f"{file} is not a model state checkpoint")
|
| 109 |
+
buffer_names = state_dict[BUFFER_NAMES]
|
| 110 |
+
if debug:
|
| 111 |
+
print("Found buffers:", buffer_names)
|
| 112 |
+
|
| 113 |
+
# recover just the buffers while restoring them to fp32 if they were saved in fp16
|
| 114 |
+
buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
|
| 115 |
+
param_shapes = state_dict[PARAM_SHAPES]
|
| 116 |
+
|
| 117 |
+
# collect parameters that are included in param_shapes
|
| 118 |
+
param_names = []
|
| 119 |
+
for s in param_shapes:
|
| 120 |
+
for name in s.keys():
|
| 121 |
+
param_names.append(name)
|
| 122 |
+
|
| 123 |
+
# update with frozen parameters
|
| 124 |
+
frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
|
| 125 |
+
if frozen_param_shapes is not None:
|
| 126 |
+
if debug:
|
| 127 |
+
print(f"Found frozen_param_shapes: {frozen_param_shapes}")
|
| 128 |
+
param_names += list(frozen_param_shapes.keys())
|
| 129 |
+
|
| 130 |
+
# handle shared params
|
| 131 |
+
shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
|
| 132 |
+
|
| 133 |
+
ds_version = state_dict.get(DS_VERSION, None)
|
| 134 |
+
|
| 135 |
+
frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
|
| 136 |
+
|
| 137 |
+
z_model_state = zero_model_state(buffers=buffers,
|
| 138 |
+
param_shapes=param_shapes,
|
| 139 |
+
shared_params=shared_params,
|
| 140 |
+
ds_version=ds_version,
|
| 141 |
+
frozen_param_shapes=frozen_param_shapes,
|
| 142 |
+
frozen_param_fragments=frozen_param_fragments)
|
| 143 |
+
zero_model_states.append(z_model_state)
|
| 144 |
+
|
| 145 |
+
return zero_model_states
|
| 146 |
+
|
| 147 |
+
|
| 148 |
+
def parse_optim_states(files, ds_checkpoint_dir):
|
| 149 |
+
total_files = len(files)
|
| 150 |
+
state_dicts = []
|
| 151 |
+
for f in tqdm(files, desc='Loading checkpoint shards'):
|
| 152 |
+
state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
|
| 153 |
+
# immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
|
| 154 |
+
# and also handle the case where it was already removed by another helper script
|
| 155 |
+
state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
|
| 156 |
+
state_dicts.append(state_dict)
|
| 157 |
+
|
| 158 |
+
if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]:
|
| 159 |
+
raise ValueError(f"{files[0]} is not a zero checkpoint")
|
| 160 |
+
zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
|
| 161 |
+
world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
|
| 162 |
+
|
| 163 |
+
# For ZeRO-2 each param group can have different partition_count as data parallelism for expert
|
| 164 |
+
# parameters can be different from data parallelism for non-expert parameters. So we can just
|
| 165 |
+
# use the max of the partition_count to get the dp world_size.
|
| 166 |
+
|
| 167 |
+
if type(world_size) is list:
|
| 168 |
+
world_size = max(world_size)
|
| 169 |
+
|
| 170 |
+
if world_size != total_files:
|
| 171 |
+
raise ValueError(
|
| 172 |
+
f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
|
| 173 |
+
"Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
|
| 174 |
+
)
|
| 175 |
+
|
| 176 |
+
# the groups are named differently in each stage
|
| 177 |
+
if zero_stage <= 2:
|
| 178 |
+
fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
|
| 179 |
+
elif zero_stage == 3:
|
| 180 |
+
fp32_groups_key = FP32_FLAT_GROUPS
|
| 181 |
+
else:
|
| 182 |
+
raise ValueError(f"unknown zero stage {zero_stage}")
|
| 183 |
+
|
| 184 |
+
fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
|
| 185 |
+
return zero_stage, world_size, fp32_flat_groups
|
| 186 |
+
|
| 187 |
+
|
| 188 |
+
def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
|
| 189 |
+
"""
|
| 190 |
+
Returns fp32 state_dict reconstructed from ds checkpoint
|
| 191 |
+
|
| 192 |
+
Args:
|
| 193 |
+
- ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
|
| 194 |
+
|
| 195 |
+
"""
|
| 196 |
+
print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
|
| 197 |
+
|
| 198 |
+
optim_files = get_optim_files(ds_checkpoint_dir)
|
| 199 |
+
zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
|
| 200 |
+
print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
|
| 201 |
+
|
| 202 |
+
model_files = get_model_state_files(ds_checkpoint_dir)
|
| 203 |
+
|
| 204 |
+
zero_model_states = parse_model_states(model_files)
|
| 205 |
+
print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
|
| 206 |
+
|
| 207 |
+
if zero_stage <= 2:
|
| 208 |
+
return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 209 |
+
exclude_frozen_parameters)
|
| 210 |
+
elif zero_stage == 3:
|
| 211 |
+
return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 212 |
+
exclude_frozen_parameters)
|
| 213 |
+
|
| 214 |
+
|
| 215 |
+
def _zero2_merge_frozen_params(state_dict, zero_model_states):
|
| 216 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 217 |
+
return
|
| 218 |
+
|
| 219 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 220 |
+
frozen_param_fragments = zero_model_states[0].frozen_param_fragments
|
| 221 |
+
|
| 222 |
+
if debug:
|
| 223 |
+
num_elem = sum(s.numel() for s in frozen_param_shapes.values())
|
| 224 |
+
print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 225 |
+
|
| 226 |
+
wanted_params = len(frozen_param_shapes)
|
| 227 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 228 |
+
avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
|
| 229 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 230 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 231 |
+
|
| 232 |
+
total_params = 0
|
| 233 |
+
total_numel = 0
|
| 234 |
+
for name, shape in frozen_param_shapes.items():
|
| 235 |
+
total_params += 1
|
| 236 |
+
unpartitioned_numel = shape.numel()
|
| 237 |
+
total_numel += unpartitioned_numel
|
| 238 |
+
|
| 239 |
+
state_dict[name] = frozen_param_fragments[name]
|
| 240 |
+
|
| 241 |
+
if debug:
|
| 242 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 243 |
+
|
| 244 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 245 |
+
|
| 246 |
+
|
| 247 |
+
def _has_callable(obj, fn):
|
| 248 |
+
attr = getattr(obj, fn, None)
|
| 249 |
+
return callable(attr)
|
| 250 |
+
|
| 251 |
+
|
| 252 |
+
def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 253 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 254 |
+
|
| 255 |
+
# Reconstruction protocol:
|
| 256 |
+
#
|
| 257 |
+
# XXX: document this
|
| 258 |
+
|
| 259 |
+
if debug:
|
| 260 |
+
for i in range(world_size):
|
| 261 |
+
for j in range(len(fp32_flat_groups[0])):
|
| 262 |
+
print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
|
| 263 |
+
|
| 264 |
+
# XXX: memory usage doubles here (zero2)
|
| 265 |
+
num_param_groups = len(fp32_flat_groups[0])
|
| 266 |
+
merged_single_partition_of_fp32_groups = []
|
| 267 |
+
for i in range(num_param_groups):
|
| 268 |
+
merged_partitions = [sd[i] for sd in fp32_flat_groups]
|
| 269 |
+
full_single_fp32_vector = torch.cat(merged_partitions, 0)
|
| 270 |
+
merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
|
| 271 |
+
avail_numel = sum(
|
| 272 |
+
[full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
|
| 273 |
+
|
| 274 |
+
if debug:
|
| 275 |
+
wanted_params = sum([len(shapes) for shapes in param_shapes])
|
| 276 |
+
wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
|
| 277 |
+
# not asserting if there is a mismatch due to possible padding
|
| 278 |
+
print(f"Have {avail_numel} numels to process.")
|
| 279 |
+
print(f"Need {wanted_numel} numels in {wanted_params} params.")
|
| 280 |
+
|
| 281 |
+
# params
|
| 282 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 283 |
+
# out-of-core computing solution
|
| 284 |
+
total_numel = 0
|
| 285 |
+
total_params = 0
|
| 286 |
+
for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
|
| 287 |
+
offset = 0
|
| 288 |
+
avail_numel = full_single_fp32_vector.numel()
|
| 289 |
+
for name, shape in shapes.items():
|
| 290 |
+
|
| 291 |
+
unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
|
| 292 |
+
total_numel += unpartitioned_numel
|
| 293 |
+
total_params += 1
|
| 294 |
+
|
| 295 |
+
if debug:
|
| 296 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 297 |
+
state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
|
| 298 |
+
offset += unpartitioned_numel
|
| 299 |
+
|
| 300 |
+
# Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
|
| 301 |
+
# avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
|
| 302 |
+
# paddings performed in the code it's almost impossible to predict the exact numbers w/o the
|
| 303 |
+
# live optimizer object, so we are checking that the numbers are within the right range
|
| 304 |
+
align_to = 2 * world_size
|
| 305 |
+
|
| 306 |
+
def zero2_align(x):
|
| 307 |
+
return align_to * math.ceil(x / align_to)
|
| 308 |
+
|
| 309 |
+
if debug:
|
| 310 |
+
print(f"original offset={offset}, avail_numel={avail_numel}")
|
| 311 |
+
|
| 312 |
+
offset = zero2_align(offset)
|
| 313 |
+
avail_numel = zero2_align(avail_numel)
|
| 314 |
+
|
| 315 |
+
if debug:
|
| 316 |
+
print(f"aligned offset={offset}, avail_numel={avail_numel}")
|
| 317 |
+
|
| 318 |
+
# Sanity check
|
| 319 |
+
if offset != avail_numel:
|
| 320 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 321 |
+
|
| 322 |
+
print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
|
| 323 |
+
|
| 324 |
+
|
| 325 |
+
def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 326 |
+
exclude_frozen_parameters):
|
| 327 |
+
state_dict = OrderedDict()
|
| 328 |
+
|
| 329 |
+
# buffers
|
| 330 |
+
buffers = zero_model_states[0].buffers
|
| 331 |
+
state_dict.update(buffers)
|
| 332 |
+
if debug:
|
| 333 |
+
print(f"added {len(buffers)} buffers")
|
| 334 |
+
|
| 335 |
+
if not exclude_frozen_parameters:
|
| 336 |
+
_zero2_merge_frozen_params(state_dict, zero_model_states)
|
| 337 |
+
|
| 338 |
+
_zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 339 |
+
|
| 340 |
+
# recover shared parameters
|
| 341 |
+
for pair in zero_model_states[0].shared_params:
|
| 342 |
+
if pair[1] in state_dict:
|
| 343 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 344 |
+
|
| 345 |
+
return state_dict
|
| 346 |
+
|
| 347 |
+
|
| 348 |
+
def zero3_partitioned_param_info(unpartitioned_numel, world_size):
|
| 349 |
+
remainder = unpartitioned_numel % world_size
|
| 350 |
+
padding_numel = (world_size - remainder) if remainder else 0
|
| 351 |
+
partitioned_numel = math.ceil(unpartitioned_numel / world_size)
|
| 352 |
+
return partitioned_numel, padding_numel
|
| 353 |
+
|
| 354 |
+
|
| 355 |
+
def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
|
| 356 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 357 |
+
return
|
| 358 |
+
|
| 359 |
+
if debug:
|
| 360 |
+
for i in range(world_size):
|
| 361 |
+
num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
|
| 362 |
+
print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 363 |
+
|
| 364 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 365 |
+
wanted_params = len(frozen_param_shapes)
|
| 366 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 367 |
+
avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
|
| 368 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 369 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 370 |
+
|
| 371 |
+
total_params = 0
|
| 372 |
+
total_numel = 0
|
| 373 |
+
for name, shape in zero_model_states[0].frozen_param_shapes.items():
|
| 374 |
+
total_params += 1
|
| 375 |
+
unpartitioned_numel = shape.numel()
|
| 376 |
+
total_numel += unpartitioned_numel
|
| 377 |
+
|
| 378 |
+
param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
|
| 379 |
+
state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
|
| 380 |
+
|
| 381 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 382 |
+
|
| 383 |
+
if debug:
|
| 384 |
+
print(
|
| 385 |
+
f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 386 |
+
)
|
| 387 |
+
|
| 388 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 389 |
+
|
| 390 |
+
|
| 391 |
+
class GatheredTensor:
|
| 392 |
+
"""
|
| 393 |
+
A pseudo tensor that collects partitioned weights.
|
| 394 |
+
It is more memory efficient when there are multiple groups.
|
| 395 |
+
"""
|
| 396 |
+
|
| 397 |
+
def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
|
| 398 |
+
self.flat_groups = flat_groups
|
| 399 |
+
self.flat_groups_offset = flat_groups_offset
|
| 400 |
+
self.offset = offset
|
| 401 |
+
self.partitioned_numel = partitioned_numel
|
| 402 |
+
self.shape = shape
|
| 403 |
+
self.dtype = self.flat_groups[0][0].dtype
|
| 404 |
+
|
| 405 |
+
def contiguous(self):
|
| 406 |
+
"""
|
| 407 |
+
Merge partitioned weights from flat_groups into a single tensor.
|
| 408 |
+
"""
|
| 409 |
+
end_idx = self.offset + self.partitioned_numel
|
| 410 |
+
world_size = len(self.flat_groups)
|
| 411 |
+
pad_flat_param_chunks = []
|
| 412 |
+
|
| 413 |
+
for rank_i in range(world_size):
|
| 414 |
+
# for each rank, we need to collect weights from related group/groups
|
| 415 |
+
flat_groups_at_rank_i = self.flat_groups[rank_i]
|
| 416 |
+
start_group_id = None
|
| 417 |
+
end_group_id = None
|
| 418 |
+
for group_id in range(len(self.flat_groups_offset)):
|
| 419 |
+
if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
|
| 420 |
+
start_group_id = group_id
|
| 421 |
+
if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
|
| 422 |
+
end_group_id = group_id
|
| 423 |
+
break
|
| 424 |
+
# collect weights from related group/groups
|
| 425 |
+
for group_id in range(start_group_id, end_group_id + 1):
|
| 426 |
+
flat_tensor = flat_groups_at_rank_i[group_id]
|
| 427 |
+
start_offset = self.offset - self.flat_groups_offset[group_id]
|
| 428 |
+
end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
|
| 429 |
+
pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
|
| 430 |
+
|
| 431 |
+
# collect weights from all ranks
|
| 432 |
+
pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
|
| 433 |
+
param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
|
| 434 |
+
return param
|
| 435 |
+
|
| 436 |
+
|
| 437 |
+
def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 438 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 439 |
+
avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
|
| 440 |
+
|
| 441 |
+
# Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
|
| 442 |
+
# param, re-consolidating each param, while dealing with padding if any
|
| 443 |
+
|
| 444 |
+
# merge list of dicts, preserving order
|
| 445 |
+
param_shapes = {k: v for d in param_shapes for k, v in d.items()}
|
| 446 |
+
|
| 447 |
+
if debug:
|
| 448 |
+
for i in range(world_size):
|
| 449 |
+
print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
|
| 450 |
+
|
| 451 |
+
wanted_params = len(param_shapes)
|
| 452 |
+
wanted_numel = sum(shape.numel() for shape in param_shapes.values())
|
| 453 |
+
# not asserting if there is a mismatch due to possible padding
|
| 454 |
+
avail_numel = fp32_flat_groups[0].numel() * world_size
|
| 455 |
+
print(f"Trainable params: Have {avail_numel} numels to process.")
|
| 456 |
+
print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
|
| 457 |
+
|
| 458 |
+
# params
|
| 459 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 460 |
+
# out-of-core computing solution
|
| 461 |
+
offset = 0
|
| 462 |
+
total_numel = 0
|
| 463 |
+
total_params = 0
|
| 464 |
+
flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
|
| 465 |
+
for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
|
| 466 |
+
unpartitioned_numel = shape.numel()
|
| 467 |
+
total_numel += unpartitioned_numel
|
| 468 |
+
total_params += 1
|
| 469 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 470 |
+
|
| 471 |
+
if debug:
|
| 472 |
+
print(
|
| 473 |
+
f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 474 |
+
)
|
| 475 |
+
|
| 476 |
+
# memory efficient tensor
|
| 477 |
+
tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
|
| 478 |
+
state_dict[name] = tensor
|
| 479 |
+
offset += partitioned_numel
|
| 480 |
+
|
| 481 |
+
offset *= world_size
|
| 482 |
+
|
| 483 |
+
# Sanity check
|
| 484 |
+
if offset != avail_numel:
|
| 485 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 486 |
+
|
| 487 |
+
print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
|
| 488 |
+
|
| 489 |
+
|
| 490 |
+
def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 491 |
+
exclude_frozen_parameters):
|
| 492 |
+
state_dict = OrderedDict()
|
| 493 |
+
|
| 494 |
+
# buffers
|
| 495 |
+
buffers = zero_model_states[0].buffers
|
| 496 |
+
state_dict.update(buffers)
|
| 497 |
+
if debug:
|
| 498 |
+
print(f"added {len(buffers)} buffers")
|
| 499 |
+
|
| 500 |
+
if not exclude_frozen_parameters:
|
| 501 |
+
_zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
|
| 502 |
+
|
| 503 |
+
_zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 504 |
+
|
| 505 |
+
# recover shared parameters
|
| 506 |
+
for pair in zero_model_states[0].shared_params:
|
| 507 |
+
if pair[1] in state_dict:
|
| 508 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 509 |
+
|
| 510 |
+
return state_dict
|
| 511 |
+
|
| 512 |
+
|
| 513 |
+
def to_torch_tensor(state_dict, return_empty_tensor=False):
|
| 514 |
+
"""
|
| 515 |
+
Convert state_dict of GatheredTensor to torch tensor
|
| 516 |
+
"""
|
| 517 |
+
torch_state_dict = {}
|
| 518 |
+
converted_tensors = {}
|
| 519 |
+
for name, tensor in state_dict.items():
|
| 520 |
+
tensor_id = id(tensor)
|
| 521 |
+
if tensor_id in converted_tensors: # shared tensors
|
| 522 |
+
shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
|
| 523 |
+
torch_state_dict[name] = shared_tensor
|
| 524 |
+
else:
|
| 525 |
+
converted_tensors[tensor_id] = name
|
| 526 |
+
if return_empty_tensor:
|
| 527 |
+
torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
|
| 528 |
+
else:
|
| 529 |
+
torch_state_dict[name] = tensor.contiguous()
|
| 530 |
+
return torch_state_dict
|
| 531 |
+
|
| 532 |
+
|
| 533 |
+
def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 534 |
+
tag=None,
|
| 535 |
+
exclude_frozen_parameters=False,
|
| 536 |
+
lazy_mode=False):
|
| 537 |
+
"""
|
| 538 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
|
| 539 |
+
``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
|
| 540 |
+
via a model hub.
|
| 541 |
+
|
| 542 |
+
Args:
|
| 543 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder
|
| 544 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
|
| 545 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 546 |
+
- ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
|
| 547 |
+
Convert the pesduo tensor to torch tensor by ``.contiguous()``
|
| 548 |
+
|
| 549 |
+
Returns:
|
| 550 |
+
- pytorch ``state_dict``
|
| 551 |
+
|
| 552 |
+
A typical usage might be ::
|
| 553 |
+
|
| 554 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 555 |
+
# do the training and checkpoint saving
|
| 556 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
|
| 557 |
+
model = model.cpu() # move to cpu
|
| 558 |
+
model.load_state_dict(state_dict)
|
| 559 |
+
# submit to model hub or save the model to share with others
|
| 560 |
+
|
| 561 |
+
In this example the ``model`` will no longer be usable in the deepspeed context of the same
|
| 562 |
+
application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 563 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 564 |
+
|
| 565 |
+
If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
|
| 566 |
+
|
| 567 |
+
Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
|
| 568 |
+
You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
|
| 569 |
+
the checkpoint. Or you can load state_dict in lazy mode ::
|
| 570 |
+
|
| 571 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 572 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
|
| 573 |
+
for name, lazy_tensor in state_dict.item():
|
| 574 |
+
tensor = lazy_tensor.contiguous() # to cpu
|
| 575 |
+
print(name, tensor)
|
| 576 |
+
# del tensor to release memory if it no longer in use
|
| 577 |
+
"""
|
| 578 |
+
if tag is None:
|
| 579 |
+
latest_path = os.path.join(checkpoint_dir, 'latest')
|
| 580 |
+
if os.path.isfile(latest_path):
|
| 581 |
+
with open(latest_path, 'r') as fd:
|
| 582 |
+
tag = fd.read().strip()
|
| 583 |
+
else:
|
| 584 |
+
raise ValueError(f"Unable to find 'latest' file at {latest_path}")
|
| 585 |
+
|
| 586 |
+
ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
|
| 587 |
+
|
| 588 |
+
if not os.path.isdir(ds_checkpoint_dir):
|
| 589 |
+
raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
|
| 590 |
+
|
| 591 |
+
state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
|
| 592 |
+
if lazy_mode:
|
| 593 |
+
return state_dict
|
| 594 |
+
else:
|
| 595 |
+
return to_torch_tensor(state_dict)
|
| 596 |
+
|
| 597 |
+
|
| 598 |
+
def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
|
| 599 |
+
output_dir,
|
| 600 |
+
max_shard_size="5GB",
|
| 601 |
+
safe_serialization=False,
|
| 602 |
+
tag=None,
|
| 603 |
+
exclude_frozen_parameters=False):
|
| 604 |
+
"""
|
| 605 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
|
| 606 |
+
loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
|
| 607 |
+
|
| 608 |
+
Args:
|
| 609 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 610 |
+
- ``output_dir``: directory to the pytorch fp32 state_dict output files
|
| 611 |
+
- ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
|
| 612 |
+
- ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
|
| 613 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 614 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 615 |
+
"""
|
| 616 |
+
|
| 617 |
+
# Dependency pre-check
|
| 618 |
+
if safe_serialization:
|
| 619 |
+
try:
|
| 620 |
+
from safetensors.torch import save_file
|
| 621 |
+
except ImportError:
|
| 622 |
+
print('If you want to use `safe_serialization`, please `pip install safetensors`')
|
| 623 |
+
raise
|
| 624 |
+
if max_shard_size is not None:
|
| 625 |
+
try:
|
| 626 |
+
from huggingface_hub import split_torch_state_dict_into_shards
|
| 627 |
+
except ImportError:
|
| 628 |
+
print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
|
| 629 |
+
raise
|
| 630 |
+
|
| 631 |
+
# Convert zero checkpoint to state_dict
|
| 632 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 633 |
+
tag,
|
| 634 |
+
exclude_frozen_parameters,
|
| 635 |
+
lazy_mode=True)
|
| 636 |
+
|
| 637 |
+
# Shard the model if it is too big.
|
| 638 |
+
weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
|
| 639 |
+
if max_shard_size is not None:
|
| 640 |
+
filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
|
| 641 |
+
# an memory-efficient approach for sharding
|
| 642 |
+
empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
|
| 643 |
+
state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
|
| 644 |
+
filename_pattern=filename_pattern,
|
| 645 |
+
max_shard_size=max_shard_size)
|
| 646 |
+
else:
|
| 647 |
+
from collections import namedtuple
|
| 648 |
+
StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
|
| 649 |
+
state_dict_split = StateDictSplit(is_sharded=False,
|
| 650 |
+
filename_to_tensors={weights_name: list(state_dict.keys())})
|
| 651 |
+
|
| 652 |
+
# Save the model by shard
|
| 653 |
+
os.makedirs(output_dir, exist_ok=True)
|
| 654 |
+
filename_to_tensors = state_dict_split.filename_to_tensors.items()
|
| 655 |
+
for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
|
| 656 |
+
shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
|
| 657 |
+
shard_state_dict = to_torch_tensor(shard_state_dict)
|
| 658 |
+
output_path = os.path.join(output_dir, shard_file)
|
| 659 |
+
if safe_serialization:
|
| 660 |
+
save_file(shard_state_dict, output_path, metadata={"format": "pt"})
|
| 661 |
+
else:
|
| 662 |
+
torch.save(shard_state_dict, output_path)
|
| 663 |
+
# release the memory of current shard
|
| 664 |
+
for tensor_name in list(shard_state_dict.keys()):
|
| 665 |
+
del state_dict[tensor_name]
|
| 666 |
+
del shard_state_dict[tensor_name]
|
| 667 |
+
del shard_state_dict
|
| 668 |
+
gc.collect()
|
| 669 |
+
|
| 670 |
+
# Save index if sharded
|
| 671 |
+
if state_dict_split.is_sharded:
|
| 672 |
+
index = {
|
| 673 |
+
"metadata": state_dict_split.metadata,
|
| 674 |
+
"weight_map": state_dict_split.tensor_to_filename,
|
| 675 |
+
}
|
| 676 |
+
save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
|
| 677 |
+
save_index_file = os.path.join(output_dir, save_index_file)
|
| 678 |
+
with open(save_index_file, "w", encoding="utf-8") as f:
|
| 679 |
+
content = json.dumps(index, indent=2, sort_keys=True) + "\n"
|
| 680 |
+
f.write(content)
|
| 681 |
+
|
| 682 |
+
|
| 683 |
+
def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
|
| 684 |
+
"""
|
| 685 |
+
1. Put the provided model to cpu
|
| 686 |
+
2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
|
| 687 |
+
3. Load it into the provided model
|
| 688 |
+
|
| 689 |
+
Args:
|
| 690 |
+
- ``model``: the model object to update
|
| 691 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 692 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 693 |
+
|
| 694 |
+
Returns:
|
| 695 |
+
- ``model`: modified model
|
| 696 |
+
|
| 697 |
+
Make sure you have plenty of CPU memory available before you call this function. If you don't
|
| 698 |
+
have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
|
| 699 |
+
conveniently placed for you in the checkpoint folder.
|
| 700 |
+
|
| 701 |
+
A typical usage might be ::
|
| 702 |
+
|
| 703 |
+
from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
|
| 704 |
+
model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
|
| 705 |
+
# submit to model hub or save the model to share with others
|
| 706 |
+
|
| 707 |
+
Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
|
| 708 |
+
of the same application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 709 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 710 |
+
|
| 711 |
+
"""
|
| 712 |
+
logger.info("Extracting fp32 weights")
|
| 713 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
| 714 |
+
|
| 715 |
+
logger.info("Overwriting model with fp32 weights")
|
| 716 |
+
model = model.cpu()
|
| 717 |
+
model.load_state_dict(state_dict, strict=False)
|
| 718 |
+
|
| 719 |
+
return model
|
| 720 |
+
|
| 721 |
+
|
| 722 |
+
if __name__ == "__main__":
|
| 723 |
+
parser = argparse.ArgumentParser()
|
| 724 |
+
parser.add_argument("checkpoint_dir",
|
| 725 |
+
type=str,
|
| 726 |
+
help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
|
| 727 |
+
parser.add_argument("output_dir",
|
| 728 |
+
type=str,
|
| 729 |
+
help="directory to the pytorch fp32 state_dict output files"
|
| 730 |
+
"(e.g. path/checkpoint-12-output/)")
|
| 731 |
+
parser.add_argument(
|
| 732 |
+
"--max_shard_size",
|
| 733 |
+
type=str,
|
| 734 |
+
default="5GB",
|
| 735 |
+
help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
|
| 736 |
+
"lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
|
| 737 |
+
"We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
|
| 738 |
+
"without CPU OOM issues.")
|
| 739 |
+
parser.add_argument(
|
| 740 |
+
"--safe_serialization",
|
| 741 |
+
default=False,
|
| 742 |
+
action='store_true',
|
| 743 |
+
help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
|
| 744 |
+
parser.add_argument("-t",
|
| 745 |
+
"--tag",
|
| 746 |
+
type=str,
|
| 747 |
+
default=None,
|
| 748 |
+
help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
|
| 749 |
+
parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
|
| 750 |
+
parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
|
| 751 |
+
args = parser.parse_args()
|
| 752 |
+
|
| 753 |
+
debug = args.debug
|
| 754 |
+
|
| 755 |
+
convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
|
| 756 |
+
args.output_dir,
|
| 757 |
+
max_shard_size=args.max_shard_size,
|
| 758 |
+
safe_serialization=args.safe_serialization,
|
| 759 |
+
tag=args.tag,
|
| 760 |
+
exclude_frozen_parameters=args.exclude_frozen_parameters)
|