Upload config.json with huggingface_hub
Browse files- config.json +165 -0
config.json
ADDED
|
@@ -0,0 +1,165 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/NHNHOME/WORKSPACE/0526040036_A/BASE/user/01/youngjoonjeong/outputs/hf_univla_bridge_lam_stage2_b200_tf21_vla50k_cleanenv_step050000",
|
| 3 |
+
"arch_specifier": "no-align+fused-gelu-mlp",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"OpenVLAForActionPrediction"
|
| 6 |
+
],
|
| 7 |
+
"hf_llm_id": "meta-llama/Llama-2-7b-hf",
|
| 8 |
+
"image_resize_strategy": "resize-naive",
|
| 9 |
+
"image_sizes": [
|
| 10 |
+
224,
|
| 11 |
+
224
|
| 12 |
+
],
|
| 13 |
+
"llm_backbone_id": "llama2-7b-pure",
|
| 14 |
+
"llm_max_length": 2048,
|
| 15 |
+
"model_type": "openvla",
|
| 16 |
+
"n_action_bins": 256,
|
| 17 |
+
"norm_stats": {
|
| 18 |
+
"bridge_dataset": {
|
| 19 |
+
"action": {
|
| 20 |
+
"mask": [
|
| 21 |
+
true,
|
| 22 |
+
true,
|
| 23 |
+
true,
|
| 24 |
+
true,
|
| 25 |
+
true,
|
| 26 |
+
true,
|
| 27 |
+
false
|
| 28 |
+
],
|
| 29 |
+
"max": [
|
| 30 |
+
0.41691166162490845,
|
| 31 |
+
0.25864794850349426,
|
| 32 |
+
0.21218234300613403,
|
| 33 |
+
3.122201919555664,
|
| 34 |
+
1.8618112802505493,
|
| 35 |
+
6.280478477478027,
|
| 36 |
+
1.0
|
| 37 |
+
],
|
| 38 |
+
"mean": [
|
| 39 |
+
0.00023341922496911138,
|
| 40 |
+
0.0001300483418162912,
|
| 41 |
+
-0.00012762336700689048,
|
| 42 |
+
-0.00015565520152449608,
|
| 43 |
+
-0.0004039322375319898,
|
| 44 |
+
0.00023557532404083759,
|
| 45 |
+
0.5764579176902771
|
| 46 |
+
],
|
| 47 |
+
"min": [
|
| 48 |
+
-0.4007510244846344,
|
| 49 |
+
-0.13874775171279907,
|
| 50 |
+
-0.22553899884223938,
|
| 51 |
+
-3.2010786533355713,
|
| 52 |
+
-1.8618112802505493,
|
| 53 |
+
-6.279075622558594,
|
| 54 |
+
0.0
|
| 55 |
+
],
|
| 56 |
+
"q01": [
|
| 57 |
+
-0.028727253898978233,
|
| 58 |
+
-0.04170349985361099,
|
| 59 |
+
-0.02609385922551155,
|
| 60 |
+
-0.08092105388641357,
|
| 61 |
+
-0.09288699924945831,
|
| 62 |
+
-0.20718276500701904,
|
| 63 |
+
0.0
|
| 64 |
+
],
|
| 65 |
+
"q99": [
|
| 66 |
+
0.028309717774391174,
|
| 67 |
+
0.040855348110198975,
|
| 68 |
+
0.040161602199077606,
|
| 69 |
+
0.08192078769207001,
|
| 70 |
+
0.0779285803437233,
|
| 71 |
+
0.203825905919075,
|
| 72 |
+
1.0
|
| 73 |
+
],
|
| 74 |
+
"std": [
|
| 75 |
+
0.009765924885869026,
|
| 76 |
+
0.013689165003597736,
|
| 77 |
+
0.012667360715568066,
|
| 78 |
+
0.02853407710790634,
|
| 79 |
+
0.03063797764480114,
|
| 80 |
+
0.07691451907157898,
|
| 81 |
+
0.4973701238632202
|
| 82 |
+
]
|
| 83 |
+
},
|
| 84 |
+
"num_trajectories": 60064,
|
| 85 |
+
"num_transitions": 2135463,
|
| 86 |
+
"proprio": {
|
| 87 |
+
"max": [
|
| 88 |
+
0.0,
|
| 89 |
+
0.0,
|
| 90 |
+
0.0,
|
| 91 |
+
0.0,
|
| 92 |
+
0.0,
|
| 93 |
+
0.0,
|
| 94 |
+
0.0
|
| 95 |
+
],
|
| 96 |
+
"mean": [
|
| 97 |
+
0.0,
|
| 98 |
+
0.0,
|
| 99 |
+
0.0,
|
| 100 |
+
0.0,
|
| 101 |
+
0.0,
|
| 102 |
+
0.0,
|
| 103 |
+
0.0
|
| 104 |
+
],
|
| 105 |
+
"min": [
|
| 106 |
+
0.0,
|
| 107 |
+
0.0,
|
| 108 |
+
0.0,
|
| 109 |
+
0.0,
|
| 110 |
+
0.0,
|
| 111 |
+
0.0,
|
| 112 |
+
0.0
|
| 113 |
+
],
|
| 114 |
+
"q01": [
|
| 115 |
+
0.0,
|
| 116 |
+
0.0,
|
| 117 |
+
0.0,
|
| 118 |
+
0.0,
|
| 119 |
+
0.0,
|
| 120 |
+
0.0,
|
| 121 |
+
0.0
|
| 122 |
+
],
|
| 123 |
+
"q99": [
|
| 124 |
+
0.0,
|
| 125 |
+
0.0,
|
| 126 |
+
0.0,
|
| 127 |
+
0.0,
|
| 128 |
+
0.0,
|
| 129 |
+
0.0,
|
| 130 |
+
0.0
|
| 131 |
+
],
|
| 132 |
+
"std": [
|
| 133 |
+
0.0,
|
| 134 |
+
0.0,
|
| 135 |
+
0.0,
|
| 136 |
+
0.0,
|
| 137 |
+
0.0,
|
| 138 |
+
0.0,
|
| 139 |
+
0.0
|
| 140 |
+
]
|
| 141 |
+
}
|
| 142 |
+
}
|
| 143 |
+
},
|
| 144 |
+
"output_projector_states": false,
|
| 145 |
+
"pad_to_multiple_of": 64,
|
| 146 |
+
"pad_token_id": 32000,
|
| 147 |
+
"text_config": {
|
| 148 |
+
"model_type": "llama",
|
| 149 |
+
"pad_token_id": 32000,
|
| 150 |
+
"torch_dtype": "bfloat16",
|
| 151 |
+
"vocab_size": 32064
|
| 152 |
+
},
|
| 153 |
+
"timm_model_ids": [
|
| 154 |
+
"vit_large_patch14_reg4_dinov2.lvd142m",
|
| 155 |
+
"vit_so400m_patch14_siglip_224"
|
| 156 |
+
],
|
| 157 |
+
"timm_override_act_layers": [
|
| 158 |
+
null,
|
| 159 |
+
null
|
| 160 |
+
],
|
| 161 |
+
"torch_dtype": "bfloat16",
|
| 162 |
+
"transformers_version": "4.40.1",
|
| 163 |
+
"use_fused_vision_backbone": true,
|
| 164 |
+
"vision_backbone_id": "dinosiglip-vit-so-224px"
|
| 165 |
+
}
|