anhtld commited on
Commit
f4cace9
·
verified ·
1 Parent(s): ca2b39e

auto-sync 2026-07-03T15:40:37Z workspace (part 3)

Browse files
workspace/runs/chart_feature_audit_rgb_refs/data_hash.txt CHANGED
@@ -1,14 +1,14 @@
1
  {
2
  "data/cil_charts_rgb_refs/test/index.json": {
3
- "content_hash": "a249a4d627b8113988660be3aecd867ba1b7b8b8d00f66eb0e3070439a395554",
4
  "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
5
  },
6
  "data/cil_charts_rgb_refs/train/index.json": {
7
- "content_hash": "2cad4dbf330d10ecaeb4ed4e86ceb162850223d2d9618170b6800b1c73e3d531",
8
  "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
9
  },
10
  "data/cil_charts_rgb_refs/val/index.json": {
11
- "content_hash": "bad9b7241350f81c31e394e231d066a14138a61bb8e23a68e13e15eb5ac65d64",
12
  "split_hash": "9d1dc1c10868f76fa080895df0f2730c173f12b6db14932599d55c2b60859381"
13
  }
14
  }
 
1
  {
2
  "data/cil_charts_rgb_refs/test/index.json": {
3
+ "content_hash": "d258ed27d587e6f41ef6ffd520832087fe7626b5e5cba2bd6352f50a1f45125a",
4
  "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
5
  },
6
  "data/cil_charts_rgb_refs/train/index.json": {
7
+ "content_hash": "1d15143588697e89f7fc2f6375b2745e4d479c726c9a947ff7eb7b2705280e1a",
8
  "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
9
  },
10
  "data/cil_charts_rgb_refs/val/index.json": {
11
+ "content_hash": "577761809f93f760fc7de6558a25d8085b2a0e51ca5d969680e99bcfa8586850",
12
  "split_hash": "9d1dc1c10868f76fa080895df0f2730c173f12b6db14932599d55c2b60859381"
13
  }
14
  }
workspace/runs/chart_feature_audit_rgb_refs/metrics.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "conclusion": "Current chart exports do not contain observation embeddings or raw observation refs; visual/object-centric chart tokens require a new export or embedding pass.",
3
  "indexes": [
4
  "data/cil_charts_rgb_refs/train/index.json",
5
  "data/cil_charts_rgb_refs/val/index.json",
@@ -10,7 +10,7 @@
10
  {
11
  "chart_id": "ms-52ea0261959dfacdbcac",
12
  "has_instruction": true,
13
- "has_observation_embedding_path": false,
14
  "has_observation_ref": true,
15
  "has_scene_id": true,
16
  "split": "train",
@@ -19,7 +19,7 @@
19
  {
20
  "chart_id": "ms-52ea0261959dfacdbcac",
21
  "has_instruction": true,
22
- "has_observation_embedding_path": false,
23
  "has_observation_ref": true,
24
  "has_scene_id": true,
25
  "split": "train",
@@ -28,7 +28,7 @@
28
  {
29
  "chart_id": "ms-52ea0261959dfacdbcac",
30
  "has_instruction": true,
31
- "has_observation_embedding_path": false,
32
  "has_observation_ref": true,
33
  "has_scene_id": true,
34
  "split": "train",
@@ -37,7 +37,7 @@
37
  {
38
  "chart_id": "ms-52ea0261959dfacdbcac",
39
  "has_instruction": true,
40
- "has_observation_embedding_path": false,
41
  "has_observation_ref": true,
42
  "has_scene_id": true,
43
  "split": "train",
@@ -46,7 +46,7 @@
46
  {
47
  "chart_id": "ms-52ea0261959dfacdbcac",
48
  "has_instruction": true,
49
- "has_observation_embedding_path": false,
50
  "has_observation_ref": true,
51
  "has_scene_id": true,
52
  "split": "train",
@@ -55,7 +55,7 @@
55
  {
56
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
57
  "has_instruction": true,
58
- "has_observation_embedding_path": false,
59
  "has_observation_ref": true,
60
  "has_scene_id": true,
61
  "split": "val",
@@ -64,7 +64,7 @@
64
  {
65
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
66
  "has_instruction": true,
67
- "has_observation_embedding_path": false,
68
  "has_observation_ref": true,
69
  "has_scene_id": true,
70
  "split": "val",
@@ -73,7 +73,7 @@
73
  {
74
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
75
  "has_instruction": true,
76
- "has_observation_embedding_path": false,
77
  "has_observation_ref": true,
78
  "has_scene_id": true,
79
  "split": "val",
@@ -82,7 +82,7 @@
82
  {
83
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
84
  "has_instruction": true,
85
- "has_observation_embedding_path": false,
86
  "has_observation_ref": true,
87
  "has_scene_id": true,
88
  "split": "val",
@@ -91,7 +91,7 @@
91
  {
92
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
93
  "has_instruction": true,
94
- "has_observation_embedding_path": false,
95
  "has_observation_ref": true,
96
  "has_scene_id": true,
97
  "split": "val",
@@ -100,7 +100,7 @@
100
  {
101
  "chart_id": "ms-4b3a296acc66a14476c3",
102
  "has_instruction": true,
103
- "has_observation_embedding_path": false,
104
  "has_observation_ref": true,
105
  "has_scene_id": true,
106
  "split": "test",
@@ -109,7 +109,7 @@
109
  {
110
  "chart_id": "ms-4b3a296acc66a14476c3",
111
  "has_instruction": true,
112
- "has_observation_embedding_path": false,
113
  "has_observation_ref": true,
114
  "has_scene_id": true,
115
  "split": "test",
@@ -118,7 +118,7 @@
118
  {
119
  "chart_id": "ms-4b3a296acc66a14476c3",
120
  "has_instruction": true,
121
- "has_observation_embedding_path": false,
122
  "has_observation_ref": true,
123
  "has_scene_id": true,
124
  "split": "test",
@@ -127,7 +127,7 @@
127
  {
128
  "chart_id": "ms-4b3a296acc66a14476c3",
129
  "has_instruction": true,
130
- "has_observation_embedding_path": false,
131
  "has_observation_ref": true,
132
  "has_scene_id": true,
133
  "split": "test",
@@ -136,7 +136,7 @@
136
  {
137
  "chart_id": "ms-4b3a296acc66a14476c3",
138
  "has_instruction": true,
139
- "has_observation_embedding_path": false,
140
  "has_observation_ref": true,
141
  "has_scene_id": true,
142
  "split": "test",
@@ -149,14 +149,15 @@
149
  "charts": 2044,
150
  "feature_dims": {
151
  "base": 112,
152
- "base_context": 130
 
153
  },
154
  "include_outcomes": true,
155
  "index": "data/cil_charts_rgb_refs/train/index.json",
156
  "instruction_present": 32704,
157
  "instruction_rate": 1.0,
158
- "observation_embedding_path_present": 0,
159
- "observation_embedding_path_rate": 0.0,
160
  "observation_ref_present": 32704,
161
  "observation_ref_rate": 1.0,
162
  "retrieval_index_allowed": true,
@@ -178,14 +179,15 @@
178
  "charts": 419,
179
  "feature_dims": {
180
  "base": 112,
181
- "base_context": 130
 
182
  },
183
  "include_outcomes": true,
184
  "index": "data/cil_charts_rgb_refs/val/index.json",
185
  "instruction_present": 6704,
186
  "instruction_rate": 1.0,
187
- "observation_embedding_path_present": 0,
188
- "observation_embedding_path_rate": 0.0,
189
  "observation_ref_present": 6704,
190
  "observation_ref_rate": 1.0,
191
  "retrieval_index_allowed": false,
@@ -207,14 +209,15 @@
207
  "charts": 410,
208
  "feature_dims": {
209
  "base": 112,
210
- "base_context": 130
 
211
  },
212
  "include_outcomes": true,
213
  "index": "data/cil_charts_rgb_refs/test/index.json",
214
  "instruction_present": 6560,
215
  "instruction_rate": 1.0,
216
- "observation_embedding_path_present": 0,
217
- "observation_embedding_path_rate": 0.0,
218
  "observation_ref_present": 6560,
219
  "observation_ref_rate": 1.0,
220
  "retrieval_index_allowed": false,
 
1
  {
2
+ "conclusion": "Current chart exports expose observation embeddings in every split; visual chart features can be evaluated with leakage-audited indexes.",
3
  "indexes": [
4
  "data/cil_charts_rgb_refs/train/index.json",
5
  "data/cil_charts_rgb_refs/val/index.json",
 
10
  {
11
  "chart_id": "ms-52ea0261959dfacdbcac",
12
  "has_instruction": true,
13
+ "has_observation_embedding_path": true,
14
  "has_observation_ref": true,
15
  "has_scene_id": true,
16
  "split": "train",
 
19
  {
20
  "chart_id": "ms-52ea0261959dfacdbcac",
21
  "has_instruction": true,
22
+ "has_observation_embedding_path": true,
23
  "has_observation_ref": true,
24
  "has_scene_id": true,
25
  "split": "train",
 
28
  {
29
  "chart_id": "ms-52ea0261959dfacdbcac",
30
  "has_instruction": true,
31
+ "has_observation_embedding_path": true,
32
  "has_observation_ref": true,
33
  "has_scene_id": true,
34
  "split": "train",
 
37
  {
38
  "chart_id": "ms-52ea0261959dfacdbcac",
39
  "has_instruction": true,
40
+ "has_observation_embedding_path": true,
41
  "has_observation_ref": true,
42
  "has_scene_id": true,
43
  "split": "train",
 
46
  {
47
  "chart_id": "ms-52ea0261959dfacdbcac",
48
  "has_instruction": true,
49
+ "has_observation_embedding_path": true,
50
  "has_observation_ref": true,
51
  "has_scene_id": true,
52
  "split": "train",
 
55
  {
56
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
57
  "has_instruction": true,
58
+ "has_observation_embedding_path": true,
59
  "has_observation_ref": true,
60
  "has_scene_id": true,
61
  "split": "val",
 
64
  {
65
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
66
  "has_instruction": true,
67
+ "has_observation_embedding_path": true,
68
  "has_observation_ref": true,
69
  "has_scene_id": true,
70
  "split": "val",
 
73
  {
74
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
75
  "has_instruction": true,
76
+ "has_observation_embedding_path": true,
77
  "has_observation_ref": true,
78
  "has_scene_id": true,
79
  "split": "val",
 
82
  {
83
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
84
  "has_instruction": true,
85
+ "has_observation_embedding_path": true,
86
  "has_observation_ref": true,
87
  "has_scene_id": true,
88
  "split": "val",
 
91
  {
92
  "chart_id": "ms-0a8109ce7ab2ddc1bb1e",
93
  "has_instruction": true,
94
+ "has_observation_embedding_path": true,
95
  "has_observation_ref": true,
96
  "has_scene_id": true,
97
  "split": "val",
 
100
  {
101
  "chart_id": "ms-4b3a296acc66a14476c3",
102
  "has_instruction": true,
103
+ "has_observation_embedding_path": true,
104
  "has_observation_ref": true,
105
  "has_scene_id": true,
106
  "split": "test",
 
109
  {
110
  "chart_id": "ms-4b3a296acc66a14476c3",
111
  "has_instruction": true,
112
+ "has_observation_embedding_path": true,
113
  "has_observation_ref": true,
114
  "has_scene_id": true,
115
  "split": "test",
 
118
  {
119
  "chart_id": "ms-4b3a296acc66a14476c3",
120
  "has_instruction": true,
121
+ "has_observation_embedding_path": true,
122
  "has_observation_ref": true,
123
  "has_scene_id": true,
124
  "split": "test",
 
127
  {
128
  "chart_id": "ms-4b3a296acc66a14476c3",
129
  "has_instruction": true,
130
+ "has_observation_embedding_path": true,
131
  "has_observation_ref": true,
132
  "has_scene_id": true,
133
  "split": "test",
 
136
  {
137
  "chart_id": "ms-4b3a296acc66a14476c3",
138
  "has_instruction": true,
139
+ "has_observation_embedding_path": true,
140
  "has_observation_ref": true,
141
  "has_scene_id": true,
142
  "split": "test",
 
149
  "charts": 2044,
150
  "feature_dims": {
151
  "base": 112,
152
+ "base_context": 130,
153
+ "base_context_obs": 162
154
  },
155
  "include_outcomes": true,
156
  "index": "data/cil_charts_rgb_refs/train/index.json",
157
  "instruction_present": 32704,
158
  "instruction_rate": 1.0,
159
+ "observation_embedding_path_present": 32704,
160
+ "observation_embedding_path_rate": 1.0,
161
  "observation_ref_present": 32704,
162
  "observation_ref_rate": 1.0,
163
  "retrieval_index_allowed": true,
 
179
  "charts": 419,
180
  "feature_dims": {
181
  "base": 112,
182
+ "base_context": 130,
183
+ "base_context_obs": 162
184
  },
185
  "include_outcomes": true,
186
  "index": "data/cil_charts_rgb_refs/val/index.json",
187
  "instruction_present": 6704,
188
  "instruction_rate": 1.0,
189
+ "observation_embedding_path_present": 6704,
190
+ "observation_embedding_path_rate": 1.0,
191
  "observation_ref_present": 6704,
192
  "observation_ref_rate": 1.0,
193
  "retrieval_index_allowed": false,
 
209
  "charts": 410,
210
  "feature_dims": {
211
  "base": 112,
212
+ "base_context": 130,
213
+ "base_context_obs": 162
214
  },
215
  "include_outcomes": true,
216
  "index": "data/cil_charts_rgb_refs/test/index.json",
217
  "instruction_present": 6560,
218
  "instruction_rate": 1.0,
219
+ "observation_embedding_path_present": 6560,
220
+ "observation_embedding_path_rate": 1.0,
221
  "observation_ref_present": 6560,
222
  "observation_ref_rate": 1.0,
223
  "retrieval_index_allowed": false,
workspace/runs/chart_feature_audit_rgb_refs/report.md CHANGED
@@ -1,9 +1,9 @@
1
  # Chart Feature Source Audit
2
 
3
- Current chart exports do not contain observation embeddings or raw observation refs; visual/object-centric chart tokens require a new export or embedding pass.
4
 
5
  | Split | Rows | Obs embedding path | Obs ref | Scene id | Instruction | Feature dims |
6
  | --- | ---: | ---: | ---: | ---: | ---: | --- |
7
- | train | 32704 | 0 (0.00%) | 32704 (100.00%) | 32704 (100.00%) | 32704 (100.00%) | {"base": 112, "base_context": 130} |
8
- | val | 6704 | 0 (0.00%) | 6704 (100.00%) | 6704 (100.00%) | 6704 (100.00%) | {"base": 112, "base_context": 130} |
9
- | test | 6560 | 0 (0.00%) | 6560 (100.00%) | 6560 (100.00%) | 6560 (100.00%) | {"base": 112, "base_context": 130} |
 
1
  # Chart Feature Source Audit
2
 
3
+ Current chart exports expose observation embeddings in every split; visual chart features can be evaluated with leakage-audited indexes.
4
 
5
  | Split | Rows | Obs embedding path | Obs ref | Scene id | Instruction | Feature dims |
6
  | --- | ---: | ---: | ---: | ---: | ---: | --- |
7
+ | train | 32704 | 32704 (100.00%) | 32704 (100.00%) | 32704 (100.00%) | 32704 (100.00%) | {"base": 112, "base_context": 130, "base_context_obs": 162} |
8
+ | val | 6704 | 6704 (100.00%) | 6704 (100.00%) | 6704 (100.00%) | 6704 (100.00%) | {"base": 112, "base_context": 130, "base_context_obs": 162} |
9
+ | test | 6560 | 6560 (100.00%) | 6560 (100.00%) | 6560 (100.00%) | 6560 (100.00%) | {"base": 112, "base_context": 130, "base_context_obs": 162} |
workspace/runs/chart_feature_audit_rgb_refs/split_hash.txt CHANGED
@@ -1,14 +1,14 @@
1
  {
2
  "data/cil_charts_rgb_refs/test/index.json": {
3
- "content_hash": "a249a4d627b8113988660be3aecd867ba1b7b8b8d00f66eb0e3070439a395554",
4
  "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
5
  },
6
  "data/cil_charts_rgb_refs/train/index.json": {
7
- "content_hash": "2cad4dbf330d10ecaeb4ed4e86ceb162850223d2d9618170b6800b1c73e3d531",
8
  "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
9
  },
10
  "data/cil_charts_rgb_refs/val/index.json": {
11
- "content_hash": "bad9b7241350f81c31e394e231d066a14138a61bb8e23a68e13e15eb5ac65d64",
12
  "split_hash": "9d1dc1c10868f76fa080895df0f2730c173f12b6db14932599d55c2b60859381"
13
  }
14
  }
 
1
  {
2
  "data/cil_charts_rgb_refs/test/index.json": {
3
+ "content_hash": "d258ed27d587e6f41ef6ffd520832087fe7626b5e5cba2bd6352f50a1f45125a",
4
  "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
5
  },
6
  "data/cil_charts_rgb_refs/train/index.json": {
7
+ "content_hash": "1d15143588697e89f7fc2f6375b2745e4d479c726c9a947ff7eb7b2705280e1a",
8
  "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
9
  },
10
  "data/cil_charts_rgb_refs/val/index.json": {
11
+ "content_hash": "577761809f93f760fc7de6558a25d8085b2a0e51ca5d969680e99bcfa8586850",
12
  "split_hash": "9d1dc1c10868f76fa080895df0f2730c173f12b6db14932599d55c2b60859381"
13
  }
14
  }
workspace/runs/chart_feature_audit_rgb_refs/table.tex CHANGED
@@ -3,8 +3,8 @@
3
  \toprule
4
  Split & Rows & ObsEmbed & ObsRef & Instruction \\
5
  \midrule
6
- train & 32704 & 0 & 32704 & 32704 \\
7
- val & 6704 & 0 & 6704 & 6704 \\
8
- test & 6560 & 0 & 6560 & 6560 \\
9
  \bottomrule
10
  \end{tabular}
 
3
  \toprule
4
  Split & Rows & ObsEmbed & ObsRef & Instruction \\
5
  \midrule
6
+ train & 32704 & 32704 & 32704 & 32704 \\
7
+ val & 6704 & 6704 & 6704 & 6704 \\
8
+ test & 6560 & 6560 & 6560 & 6560 \\
9
  \bottomrule
10
  \end{tabular}