Hugging Face's logo

InternRobotics
/
VLAC-Cut

Image-Text-to-Text
Transformers
Safetensors
qwen3_vl_moe
robotics
embodied-ai
video-understanding
progress-estimation
reward-modeling
qwen3-vl
conversational
Model card Files
xet
 Community
futurefantasy commited on
Commit
44b7d8e
·
verified ·
1 Parent(s): feac113

Upload examples/example_01/metadata.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. examples/example_01/metadata.json +305 -9
examples/example_01/metadata.json CHANGED
@@ -1,20 +1,316 @@
1
  {
2
  "example_id": "example_01",
3
  "bucket": "test_nonexpert_seen",
4
- "global_episode_id": "ARX-data/human-data-1128/20251126-150505-good-cocket-cylindrical-container/videos/chunk-000/observation.images.front/episode_000007",
5
- "main_path": "ARX-data/human-data-1128/20251126-150505-good-cocket-cylindrical-container/videos/chunk-000/observation.images.front/episode_000007",
6
- "task_instruction": "抓取插座放入电饭锅的内锅中",
7
- "task_description": "抓取插座放入电饭锅的内锅中:\n开始移动:0%\n爪夹靠近插座:20%\n爪夹抓取插座:40%\n爪夹夹紧插座靠近内锅:60%\n爪夹夹紧插座落下放到内锅:80%\n爪夹松开，插座落入内锅中:100%",
8
  "fps": 30.0,
9
  "is_expert": false,
10
- "num_frames": 338,
11
- "final_gt_progress": 100.0,
12
  "video_path": "examples/example_01/episode.mp4",
13
  "benchmark_progress_type": "dense_kinematic_progress",
14
- "benchmark_dense_progress": [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, -0.2, -0.4, -0.5, -0.7, -0.9, -1.1, -1.2, -1.4, -1.6, -1.8, -2.0, -2.1, -2.3, -2.5, -2.7, -2.9, -3.0, -3.2, -3.4, -3.6, -3.8, -3.9, -4.1, -4.3, -4.5, -4.6, -4.8, -5.0, -5.1, -5.2, -5.3, -5.5, -5.6, -5.7, -5.8, -5.9, -6.0, -6.2, -6.3, -6.4, -6.5, -6.6, -6.7, -6.8, -7.0, -7.1, -7.2, -7.3, -7.4, -7.5, -7.7, -7.8, -7.9, -8.0, -7.2, -6.4, -5.7, -4.9, -4.1, -3.3, -2.5, -1.7, -1.0, -0.2, 0.6, 1.4, 2.2, 3.0, 3.7, 4.5, 5.3, 6.1, 6.9, 7.7, 8.4, 9.2, 10.0, 10.3, 10.5, 10.8, 11.1, 11.3, 11.6, 11.8, 12.1, 12.4, 12.6, 12.9, 13.2, 13.4, 13.7, 13.9, 14.2, 14.5, 14.7, 15.0, 15.2, 15.3, 15.5, 15.7, 15.8, 16.0, 16.2, 16.3, 16.5, 16.7, 16.8, 17.0, 17.2, 17.3, 17.5, 17.7, 17.8, 18.0, 18.2, 18.3, 18.5, 18.7, 18.8, 19.0, 19.2, 19.3, 19.5, 19.7, 19.8, 20.0, 20.6, 21.3, 21.9, 22.6, 23.2, 23.9, 24.5, 25.2, 25.8, 26.5, 27.1, 27.7, 28.4, 29.0, 29.7, 30.3, 31.0, 31.6, 32.3, 32.9, 33.5, 34.2, 34.8, 35.5, 36.1, 36.8, 37.4, 38.1, 38.7, 39.4, 40.0, 40.3, 40.7, 41.0, 41.4, 41.7, 42.1, 42.4, 42.8, 43.1, 43.4, 43.8, 44.1, 44.5, 44.8, 45.2, 45.5, 45.9, 46.2, 46.6, 46.9, 47.2, 47.6, 47.9, 48.3, 48.6, 49.0, 49.3, 49.7, 50.0, 50.3, 50.6, 50.9, 51.1, 51.4, 51.7, 52.0, 52.3, 52.6, 52.9, 53.1, 53.4, 53.7, 54.0, 54.3, 54.6, 54.9, 55.1, 55.4, 55.7, 56.0, 56.3, 56.6, 56.9, 57.1, 57.4, 57.7, 58.0, 58.3, 58.6, 58.9, 59.1, 59.4, 59.7, 60.0, 61.0, 62.0, 63.0, 64.0, 65.0, 66.0, 67.0, 68.0, 69.0, 70.0, 71.0, 72.0, 73.0, 74.0, 75.0, 76.0, 77.0, 78.0, 79.0, 80.0, 80.9, 81.7, 82.6, 83.5, 84.3, 85.2, 86.1, 87.0, 87.8, 88.7, 89.6, 90.4, 91.3, 92.2, 93.0, 93.9, 94.8, 95.7, 96.5, 97.4, 98.3, 99.1, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0],
15
- "benchmark_semantic_anchors": [{"frame": 42, "progress": 0.0}, {"frame": 70, "progress": -5.0}, {"frame": 96, "progress": -8.0}, {"frame": 119, "progress": 10.0}, {"frame": 138, "progress": 15.0}, {"frame": 168, "progress": 20.0}, {"frame": 199, "progress": 40.0}, {"frame": 228, "progress": 50.0}, {"frame": 263, "progress": 60.0}, {"frame": 283, "progress": 80.0}, {"frame": 306, "progress": 100.0}],
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  "benchmark_progress_source": {
17
  "bucket": "test_nonexpert_seen",
18
- "global_episode_id": "ARX-data/human-data-1128/20251126-150505-good-cocket-cylindrical-container/videos/chunk-000/observation.images.front/episode_000007"
19
  }
20
  }
 
1
  {
2
  "example_id": "example_01",
3
  "bucket": "test_nonexpert_seen",
4
+ "global_episode_id": "ARX-data/human-data-1204-2cameras/20251130-162433-bad-purple cube-shaped building block/videos/chunk-000/observation.images.front/episode_000007",
5
+ "main_path": "ARX-data/human-data-1204-2cameras/20251130-162433-bad-purple cube-shaped building block/videos/chunk-000/observation.images.front/episode_000007",
6
+ "task_instruction": "抓取紫色方块使其从方形洞口落入积木桶中",
7
+ "task_description": "抓取紫色方块使其从方形洞口落入积木桶中：\n爪夹开始移动：0%\n爪夹接近紫色方块：20%\n爪夹抓紧紫色方块：40%\n爪夹抓紧紫色方块接近方形洞口：60%\n爪夹将紫色方块对准方形洞口：80%\n紫色方块从方形洞口落入积木桶中，爪夹移开：100%",
8
  "fps": 30.0,
9
  "is_expert": false,
10
+ "num_frames": 258,
11
+ "final_gt_progress": -10.0,
12
  "video_path": "examples/example_01/episode.mp4",
13
  "benchmark_progress_type": "dense_kinematic_progress",
14
+ "benchmark_dense_progress": [
15
+ 0.0,
16
+ 0.0,
17
+ 0.0,
18
+ 0.0,
19
+ 0.0,
20
+ 0.0,
21
+ 0.0,
22
+ 0.0,
23
+ 0.0,
24
+ 0.0,
25
+ 0.0,
26
+ 0.0,
27
+ 0.0,
28
+ 0.0,
29
+ 0.0,
30
+ 0.0,
31
+ 0.0,
32
+ 0.0,
33
+ 0.0,
34
+ 0.0,
35
+ 0.0,
36
+ 0.0,
37
+ 0.0,
38
+ 0.0,
39
+ 0.0,
40
+ 0.0,
41
+ 0.0,
42
+ 0.0,
43
+ 0.0,
44
+ 0.0,
45
+ 0.0,
46
+ 0.0,
47
+ 0.0,
48
+ 0.0,
49
+ 0.0,
50
+ 0.0,
51
+ 0.0,
52
+ 0.0,
53
+ 0.0,
54
+ 0.0,
55
+ 0.0,
56
+ 0.0,
57
+ 0.0,
58
+ 0.0,
59
+ 0.0,
60
+ 0.0,
61
+ 0.0,
62
+ 0.0,
63
+ 0.0,
64
+ 0.0,
65
+ 0.2,
66
+ 0.4,
67
+ 0.6,
68
+ 0.8,
69
+ 1.0,
70
+ 1.2,
71
+ 1.4,
72
+ 1.6,
73
+ 1.8,
74
+ 2.0,
75
+ 2.2,
76
+ 2.4,
77
+ 2.6,
78
+ 2.8,
79
+ 3.0,
80
+ 3.2,
81
+ 3.4,
82
+ 3.6,
83
+ 3.8,
84
+ 4.0,
85
+ 4.2,
86
+ 4.4,
87
+ 4.6,
88
+ 4.8,
89
+ 5.0,
90
+ 5.4,
91
+ 5.8,
92
+ 6.2,
93
+ 6.5,
94
+ 6.9,
95
+ 7.3,
96
+ 7.7,
97
+ 8.1,
98
+ 8.5,
99
+ 8.8,
100
+ 9.2,
101
+ 9.6,
102
+ 10.0,
103
+ 10.1,
104
+ 10.2,
105
+ 10.3,
106
+ 10.4,
107
+ 10.5,
108
+ 10.6,
109
+ 10.7,
110
+ 10.9,
111
+ 11.0,
112
+ 11.1,
113
+ 11.2,
114
+ 11.3,
115
+ 11.4,
116
+ 11.5,
117
+ 11.6,
118
+ 11.7,
119
+ 11.8,
120
+ 11.9,
121
+ 12.0,
122
+ 12.1,
123
+ 12.2,
124
+ 12.3,
125
+ 12.4,
126
+ 12.6,
127
+ 12.7,
128
+ 12.8,
129
+ 12.9,
130
+ 13.0,
131
+ 13.1,
132
+ 13.2,
133
+ 13.3,
134
+ 13.4,
135
+ 13.5,
136
+ 13.6,
137
+ 13.7,
138
+ 13.8,
139
+ 13.9,
140
+ 14.0,
141
+ 14.1,
142
+ 14.3,
143
+ 14.4,
144
+ 14.5,
145
+ 14.6,
146
+ 14.7,
147
+ 14.8,
148
+ 14.9,
149
+ 15.0,
150
+ 14.9,
151
+ 14.9,
152
+ 14.8,
153
+ 14.8,
154
+ 14.7,
155
+ 14.7,
156
+ 14.6,
157
+ 14.6,
158
+ 14.5,
159
+ 14.5,
160
+ 14.4,
161
+ 14.4,
162
+ 14.3,
163
+ 14.3,
164
+ 14.2,
165
+ 14.2,
166
+ 14.1,
167
+ 14.1,
168
+ 14.0,
169
+ 14.0,
170
+ 13.9,
171
+ 13.9,
172
+ 13.8,
173
+ 13.8,
174
+ 13.7,
175
+ 13.7,
176
+ 13.6,
177
+ 13.6,
178
+ 13.5,
179
+ 13.5,
180
+ 13.4,
181
+ 13.4,
182
+ 13.3,
183
+ 13.3,
184
+ 13.2,
185
+ 13.2,
186
+ 13.1,
187
+ 13.1,
188
+ 13.0,
189
+ 10.8,
190
+ 8.7,
191
+ 6.5,
192
+ 4.4,
193
+ 2.2,
194
+ 0.1,
195
+ -2.1,
196
+ -4.2,
197
+ -6.4,
198
+ -8.5,
199
+ -10.7,
200
+ -12.8,
201
+ -15.0,
202
+ -15.4,
203
+ -15.9,
204
+ -16.3,
205
+ -16.8,
206
+ -17.2,
207
+ -17.6,
208
+ -18.1,
209
+ -18.5,
210
+ -19.0,
211
+ -19.4,
212
+ -19.9,
213
+ -20.3,
214
+ -20.7,
215
+ -21.2,
216
+ -21.6,
217
+ -22.1,
218
+ -22.5,
219
+ -22.9,
220
+ -23.4,
221
+ -23.8,
222
+ -24.3,
223
+ -24.7,
224
+ -25.1,
225
+ -25.6,
226
+ -26.0,
227
+ -26.5,
228
+ -26.9,
229
+ -27.4,
230
+ -27.8,
231
+ -28.2,
232
+ -28.7,
233
+ -29.1,
234
+ -29.6,
235
+ -30.0,
236
+ -29.5,
237
+ -28.9,
238
+ -28.4,
239
+ -27.9,
240
+ -27.4,
241
+ -26.8,
242
+ -26.3,
243
+ -25.8,
244
+ -25.3,
245
+ -24.7,
246
+ -24.2,
247
+ -23.7,
248
+ -23.2,
249
+ -22.6,
250
+ -22.1,
251
+ -21.6,
252
+ -21.1,
253
+ -20.5,
254
+ -20.0,
255
+ -19.4,
256
+ -18.9,
257
+ -18.3,
258
+ -17.8,
259
+ -17.2,
260
+ -16.7,
261
+ -16.1,
262
+ -15.6,
263
+ -15.0,
264
+ -14.4,
265
+ -13.9,
266
+ -13.3,
267
+ -12.8,
268
+ -12.2,
269
+ -11.7,
270
+ -11.1,
271
+ -10.6,
272
+ -10.0
273
+ ],
274
+ "benchmark_semantic_anchors": [
275
+ {
276
+ "frame": 49,
277
+ "progress": 0.0
278
+ },
279
+ {
280
+ "frame": 74,
281
+ "progress": 5.0
282
+ },
283
+ {
284
+ "frame": 87,
285
+ "progress": 10.0
286
+ },
287
+ {
288
+ "frame": 134,
289
+ "progress": 15.0
290
+ },
291
+ {
292
+ "frame": 173,
293
+ "progress": 13.0
294
+ },
295
+ {
296
+ "frame": 186,
297
+ "progress": -15.0
298
+ },
299
+ {
300
+ "frame": 220,
301
+ "progress": -30.0
302
+ },
303
+ {
304
+ "frame": 239,
305
+ "progress": -20.0
306
+ },
307
+ {
308
+ "frame": 257,
309
+ "progress": -10.0
310
+ }
311
+ ],
312
  "benchmark_progress_source": {
313
  "bucket": "test_nonexpert_seen",
314
+ "global_episode_id": "ARX-data/human-data-1204-2cameras/20251130-162433-bad-purple cube-shaped building block/videos/chunk-000/observation.images.front/episode_000007"
315
  }
316
  }