Upload folder using huggingface_hub
Browse files- config.json +64 -0
- experiment_cfg/metadata.json +307 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +0 -0
- trainer_state.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"action_dim": 32,
|
| 3 |
+
"action_head_cfg": {
|
| 4 |
+
"action_dim": 32,
|
| 5 |
+
"action_horizon": 16,
|
| 6 |
+
"add_pos_embed": true,
|
| 7 |
+
"backbone_embedding_dim": 2048,
|
| 8 |
+
"diffusion_model_cfg": {
|
| 9 |
+
"attention_head_dim": 48,
|
| 10 |
+
"cross_attention_dim": 2048,
|
| 11 |
+
"dropout": 0.2,
|
| 12 |
+
"final_dropout": true,
|
| 13 |
+
"interleave_self_attention": true,
|
| 14 |
+
"norm_type": "ada_norm",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_layers": 16,
|
| 17 |
+
"output_dim": 1024,
|
| 18 |
+
"positional_embeddings": null
|
| 19 |
+
},
|
| 20 |
+
"hidden_size": 1024,
|
| 21 |
+
"input_embedding_dim": 1536,
|
| 22 |
+
"max_action_dim": 32,
|
| 23 |
+
"max_state_dim": 64,
|
| 24 |
+
"model_dtype": "float32",
|
| 25 |
+
"noise_beta_alpha": 1.5,
|
| 26 |
+
"noise_beta_beta": 1.0,
|
| 27 |
+
"noise_s": 0.999,
|
| 28 |
+
"num_inference_timesteps": 4,
|
| 29 |
+
"num_target_vision_tokens": 32,
|
| 30 |
+
"num_timestep_buckets": 1000,
|
| 31 |
+
"tune_diffusion_model": true,
|
| 32 |
+
"tune_projector": true,
|
| 33 |
+
"use_vlln": true,
|
| 34 |
+
"vl_self_attention_cfg": {
|
| 35 |
+
"attention_head_dim": 64,
|
| 36 |
+
"dropout": 0.2,
|
| 37 |
+
"final_dropout": true,
|
| 38 |
+
"num_attention_heads": 32,
|
| 39 |
+
"num_layers": 4,
|
| 40 |
+
"positional_embeddings": null
|
| 41 |
+
}
|
| 42 |
+
},
|
| 43 |
+
"action_horizon": 16,
|
| 44 |
+
"architectures": [
|
| 45 |
+
"GR00T_N1_5"
|
| 46 |
+
],
|
| 47 |
+
"attn_implementation": null,
|
| 48 |
+
"backbone_cfg": {
|
| 49 |
+
"eagle_path": "NVEagle/eagle_er-qwen3_1_7B-Siglip2_400M_stage1_5_128gpu_er_v7_1mlp_nops",
|
| 50 |
+
"load_bf16": false,
|
| 51 |
+
"project_to_dim": null,
|
| 52 |
+
"reproject_vision": false,
|
| 53 |
+
"select_layer": 12,
|
| 54 |
+
"tune_llm": false,
|
| 55 |
+
"tune_visual": true,
|
| 56 |
+
"use_flash_attention": true
|
| 57 |
+
},
|
| 58 |
+
"compute_dtype": "bfloat16",
|
| 59 |
+
"hidden_size": 2048,
|
| 60 |
+
"model_dtype": "float32",
|
| 61 |
+
"model_type": "gr00t_n1_5",
|
| 62 |
+
"torch_dtype": "bfloat16",
|
| 63 |
+
"transformers_version": "4.51.3"
|
| 64 |
+
}
|
experiment_cfg/metadata.json
ADDED
|
@@ -0,0 +1,307 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"new_embodiment": {
|
| 3 |
+
"statistics": {
|
| 4 |
+
"state": {
|
| 5 |
+
"left_arm": {
|
| 6 |
+
"max": [
|
| 7 |
+
0.6258881092071533,
|
| 8 |
+
0.5225961208343506,
|
| 9 |
+
0.4540702998638153,
|
| 10 |
+
-0.4083384871482849,
|
| 11 |
+
1.9848512411117554,
|
| 12 |
+
1.3622229099273682,
|
| 13 |
+
0.32285746932029724,
|
| 14 |
+
1.1980390548706055
|
| 15 |
+
],
|
| 16 |
+
"min": [
|
| 17 |
+
-1.0356887578964233,
|
| 18 |
+
-0.00011984224693151191,
|
| 19 |
+
-0.8163893818855286,
|
| 20 |
+
-2.1820757389068604,
|
| 21 |
+
-0.1257864236831665,
|
| 22 |
+
-0.619260847568512,
|
| 23 |
+
-1.2150284051895142,
|
| 24 |
+
0.0
|
| 25 |
+
],
|
| 26 |
+
"mean": [
|
| 27 |
+
-0.030505895614624023,
|
| 28 |
+
0.17493046820163727,
|
| 29 |
+
-0.1724156141281128,
|
| 30 |
+
-1.5434216260910034,
|
| 31 |
+
0.41967833042144775,
|
| 32 |
+
0.09581489861011505,
|
| 33 |
+
-0.37070271372795105,
|
| 34 |
+
0.3712228238582611
|
| 35 |
+
],
|
| 36 |
+
"std": [
|
| 37 |
+
0.35358697175979614,
|
| 38 |
+
0.08640537410974503,
|
| 39 |
+
0.1647699922323227,
|
| 40 |
+
0.2742176055908203,
|
| 41 |
+
0.4536644518375397,
|
| 42 |
+
0.37965625524520874,
|
| 43 |
+
0.3246261775493622,
|
| 44 |
+
0.39878013730049133
|
| 45 |
+
],
|
| 46 |
+
"q01": [
|
| 47 |
+
-0.8253655433654785,
|
| 48 |
+
0.0005007008908432905,
|
| 49 |
+
-0.6553141909837723,
|
| 50 |
+
-2.006601107120514,
|
| 51 |
+
-0.05368932709097862,
|
| 52 |
+
-0.40650489926338196,
|
| 53 |
+
-1.0646234941482544,
|
| 54 |
+
0.01840776950120926
|
| 55 |
+
],
|
| 56 |
+
"q99": [
|
| 57 |
+
0.46977610290050514,
|
| 58 |
+
0.42484736740589146,
|
| 59 |
+
0.20448683202266693,
|
| 60 |
+
-0.6904751861095428,
|
| 61 |
+
1.8422895801067352,
|
| 62 |
+
1.0802250385284429,
|
| 63 |
+
0.2055506557226181,
|
| 64 |
+
1.1320778131484985
|
| 65 |
+
]
|
| 66 |
+
},
|
| 67 |
+
"right_arm": {
|
| 68 |
+
"max": [
|
| 69 |
+
0.6432053446769714,
|
| 70 |
+
7.190534961409867e-05,
|
| 71 |
+
0.37275734543800354,
|
| 72 |
+
-0.5527963638305664,
|
| 73 |
+
0.0950828418135643,
|
| 74 |
+
0.7803407907485962,
|
| 75 |
+
1.1357827186584473,
|
| 76 |
+
1.136679768562317
|
| 77 |
+
],
|
| 78 |
+
"min": [
|
| 79 |
+
-0.9628486037254333,
|
| 80 |
+
-0.5433168411254883,
|
| 81 |
+
-0.5081910490989685,
|
| 82 |
+
-2.1780848503112793,
|
| 83 |
+
-0.8541157245635986,
|
| 84 |
+
-0.7917377948760986,
|
| 85 |
+
-0.3289906978607178,
|
| 86 |
+
0.0
|
| 87 |
+
],
|
| 88 |
+
"mean": [
|
| 89 |
+
0.19008710980415344,
|
| 90 |
+
-0.10576711595058441,
|
| 91 |
+
0.09301091730594635,
|
| 92 |
+
-1.633500576019287,
|
| 93 |
+
-0.17586009204387665,
|
| 94 |
+
-0.15116524696350098,
|
| 95 |
+
0.17336353659629822,
|
| 96 |
+
0.29084932804107666
|
| 97 |
+
],
|
| 98 |
+
"std": [
|
| 99 |
+
0.32597336173057556,
|
| 100 |
+
0.09622962027788162,
|
| 101 |
+
0.1411556452512741,
|
| 102 |
+
0.20505863428115845,
|
| 103 |
+
0.13772115111351013,
|
| 104 |
+
0.3104083836078644,
|
| 105 |
+
0.3759829103946686,
|
| 106 |
+
0.3420809507369995
|
| 107 |
+
],
|
| 108 |
+
"q01": [
|
| 109 |
+
-0.7115159922838211,
|
| 110 |
+
-0.43976113200187683,
|
| 111 |
+
-0.3370482853055,
|
| 112 |
+
-2.006547954082489,
|
| 113 |
+
-0.5956598317623139,
|
| 114 |
+
-0.7027443116903305,
|
| 115 |
+
-0.28981370240449905,
|
| 116 |
+
0.02147573046386242
|
| 117 |
+
],
|
| 118 |
+
"q99": [
|
| 119 |
+
0.5825891494750977,
|
| 120 |
+
0.0,
|
| 121 |
+
0.32810542643070223,
|
| 122 |
+
-0.9048770409822463,
|
| 123 |
+
0.06654839962720871,
|
| 124 |
+
0.568606151342392,
|
| 125 |
+
0.9971432673931123,
|
| 126 |
+
1.1182719469070435
|
| 127 |
+
]
|
| 128 |
+
}
|
| 129 |
+
},
|
| 130 |
+
"action": {
|
| 131 |
+
"left_arm": {
|
| 132 |
+
"max": [
|
| 133 |
+
0.6273981332778931,
|
| 134 |
+
0.5230874419212341,
|
| 135 |
+
0.4555923044681549,
|
| 136 |
+
-0.4080388844013214,
|
| 137 |
+
1.9849711656570435,
|
| 138 |
+
1.3621749877929688,
|
| 139 |
+
0.32213595509529114,
|
| 140 |
+
1.2065826654434204
|
| 141 |
+
],
|
| 142 |
+
"min": [
|
| 143 |
+
-1.0369709730148315,
|
| 144 |
+
-0.11504855751991272,
|
| 145 |
+
-0.8191457390785217,
|
| 146 |
+
-2.182854652404785,
|
| 147 |
+
-0.1257864236831665,
|
| 148 |
+
-0.6212621927261353,
|
| 149 |
+
-1.21491277217865,
|
| 150 |
+
-0.3212622106075287
|
| 151 |
+
],
|
| 152 |
+
"mean": [
|
| 153 |
+
-0.030559606850147247,
|
| 154 |
+
0.174823597073555,
|
| 155 |
+
-0.17283093929290771,
|
| 156 |
+
-1.543239712715149,
|
| 157 |
+
0.4216526448726654,
|
| 158 |
+
0.09620700031518936,
|
| 159 |
+
-0.3706417381763458,
|
| 160 |
+
0.3887162208557129
|
| 161 |
+
],
|
| 162 |
+
"std": [
|
| 163 |
+
0.3538479208946228,
|
| 164 |
+
0.08742091804742813,
|
| 165 |
+
0.1650300770998001,
|
| 166 |
+
0.27444228529930115,
|
| 167 |
+
0.4572615325450897,
|
| 168 |
+
0.3797238767147064,
|
| 169 |
+
0.3246603012084961,
|
| 170 |
+
0.44660818576812744
|
| 171 |
+
],
|
| 172 |
+
"q01": [
|
| 173 |
+
-0.8252816796302795,
|
| 174 |
+
-0.007669903803616762,
|
| 175 |
+
-0.6536445593833924,
|
| 176 |
+
-2.0079808235168457,
|
| 177 |
+
-0.05368932709097862,
|
| 178 |
+
-0.40650489926338196,
|
| 179 |
+
-1.0645827054977417,
|
| 180 |
+
-0.08502917736768723
|
| 181 |
+
],
|
| 182 |
+
"q99": [
|
| 183 |
+
0.4724660813808441,
|
| 184 |
+
0.4250814262032518,
|
| 185 |
+
0.2040194422006607,
|
| 186 |
+
-0.6902913451194763,
|
| 187 |
+
1.8532175397872934,
|
| 188 |
+
1.0722525119781494,
|
| 189 |
+
0.20555342733860016,
|
| 190 |
+
1.2004467248916626
|
| 191 |
+
]
|
| 192 |
+
},
|
| 193 |
+
"right_arm": {
|
| 194 |
+
"max": [
|
| 195 |
+
0.6442719101905823,
|
| 196 |
+
0.14419420063495636,
|
| 197 |
+
0.37275734543800354,
|
| 198 |
+
-0.552233099937439,
|
| 199 |
+
0.0951068103313446,
|
| 200 |
+
0.7823302149772644,
|
| 201 |
+
1.136679768562317,
|
| 202 |
+
1.2909516096115112
|
| 203 |
+
],
|
| 204 |
+
"min": [
|
| 205 |
+
-0.9633399248123169,
|
| 206 |
+
-0.5430291891098022,
|
| 207 |
+
-0.5092816352844238,
|
| 208 |
+
-2.1797866821289062,
|
| 209 |
+
-0.855961263179779,
|
| 210 |
+
-0.7930680513381958,
|
| 211 |
+
-0.33133983612060547,
|
| 212 |
+
-0.256835013628006
|
| 213 |
+
],
|
| 214 |
+
"mean": [
|
| 215 |
+
0.19033801555633545,
|
| 216 |
+
-0.09965736418962479,
|
| 217 |
+
0.09292927384376526,
|
| 218 |
+
-1.633928894996643,
|
| 219 |
+
-0.17576472461223602,
|
| 220 |
+
-0.1508656144142151,
|
| 221 |
+
0.17342375218868256,
|
| 222 |
+
0.3023458421230316
|
| 223 |
+
],
|
| 224 |
+
"std": [
|
| 225 |
+
0.32630667090415955,
|
| 226 |
+
0.1058855652809143,
|
| 227 |
+
0.14127664268016815,
|
| 228 |
+
0.20565293729305267,
|
| 229 |
+
0.13783322274684906,
|
| 230 |
+
0.3103736639022827,
|
| 231 |
+
0.3760031461715698,
|
| 232 |
+
0.3822594881057739
|
| 233 |
+
],
|
| 234 |
+
"q01": [
|
| 235 |
+
-0.7133010625839233,
|
| 236 |
+
-0.44025248289108276,
|
| 237 |
+
-0.3374757766723633,
|
| 238 |
+
-2.0079808235168457,
|
| 239 |
+
-0.5967185497283936,
|
| 240 |
+
-0.7025632262229919,
|
| 241 |
+
-0.2899223566055298,
|
| 242 |
+
0.017747534438967705
|
| 243 |
+
],
|
| 244 |
+
"q99": [
|
| 245 |
+
0.5844466686248779,
|
| 246 |
+
0.13192234933376312,
|
| 247 |
+
0.32827189564704895,
|
| 248 |
+
-0.9035146832466125,
|
| 249 |
+
0.06596117466688156,
|
| 250 |
+
0.5675728917121887,
|
| 251 |
+
0.9970875382423401,
|
| 252 |
+
1.2710098028182983
|
| 253 |
+
]
|
| 254 |
+
}
|
| 255 |
+
}
|
| 256 |
+
},
|
| 257 |
+
"modalities": {
|
| 258 |
+
"video": {
|
| 259 |
+
"cam_head": {
|
| 260 |
+
"resolution": [
|
| 261 |
+
224,
|
| 262 |
+
224
|
| 263 |
+
],
|
| 264 |
+
"channels": 3,
|
| 265 |
+
"fps": 10.0
|
| 266 |
+
}
|
| 267 |
+
},
|
| 268 |
+
"state": {
|
| 269 |
+
"left_arm": {
|
| 270 |
+
"absolute": true,
|
| 271 |
+
"rotation_type": null,
|
| 272 |
+
"shape": [
|
| 273 |
+
8
|
| 274 |
+
],
|
| 275 |
+
"continuous": true
|
| 276 |
+
},
|
| 277 |
+
"right_arm": {
|
| 278 |
+
"absolute": true,
|
| 279 |
+
"rotation_type": null,
|
| 280 |
+
"shape": [
|
| 281 |
+
8
|
| 282 |
+
],
|
| 283 |
+
"continuous": true
|
| 284 |
+
}
|
| 285 |
+
},
|
| 286 |
+
"action": {
|
| 287 |
+
"left_arm": {
|
| 288 |
+
"absolute": true,
|
| 289 |
+
"rotation_type": null,
|
| 290 |
+
"shape": [
|
| 291 |
+
8
|
| 292 |
+
],
|
| 293 |
+
"continuous": true
|
| 294 |
+
},
|
| 295 |
+
"right_arm": {
|
| 296 |
+
"absolute": true,
|
| 297 |
+
"rotation_type": null,
|
| 298 |
+
"shape": [
|
| 299 |
+
8
|
| 300 |
+
],
|
| 301 |
+
"continuous": true
|
| 302 |
+
}
|
| 303 |
+
}
|
| 304 |
+
},
|
| 305 |
+
"embodiment_tag": "new_embodiment"
|
| 306 |
+
}
|
| 307 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b91fe2c71125a077ea98496a0fe46253242ab5662a84cc4b5fbaa21f5ba614e
|
| 3 |
+
size 4999367032
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f9948be31cd0c18faa3c18a9862e383301fdef85a6fb52cdf7e4951ac48b8f0
|
| 3 |
+
size 2586705312
|
model.safetensors.index.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
|
|
|