minhy112 commited on
Commit
355437e
·
verified ·
1 Parent(s): 68882a2

Archive verified research artifact: summary.json

Browse files
artifacts/experiments/runs/test-mira-mmqa_ti-s17/summary.json ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": "experiments/runs/gate-search-mira-mmqa_ti-transfer_p2_r1-s17/epoch2.pt",
3
+ "dataset": "mmqa_ti",
4
+ "method": "mira",
5
+ "scores": [
6
+ {
7
+ "complete": true,
8
+ "dataset": "mmqa_ti",
9
+ "evaluator_sha256": "1b95323ffdf36a0e8a8afef28c598e6fd560f6e63aa8d48709e51607e84d6100",
10
+ "groups": {
11
+ "modality:image": {
12
+ "em": 26.582278481012658,
13
+ "f1": 31.20253164556962,
14
+ "n": 79
15
+ },
16
+ "modality:image+text": {
17
+ "em": 4.3478260869565215,
18
+ "f1": 4.3478260869565215,
19
+ "n": 23
20
+ },
21
+ "modality:text": {
22
+ "em": 28.658536585365855,
23
+ "f1": 35.43292682926829,
24
+ "n": 164
25
+ },
26
+ "type:Compare(Compose(TableQ,ImageQ),Compose(TableQ,TextQ))": {
27
+ "em": 0.0,
28
+ "f1": 0.0,
29
+ "n": 4
30
+ },
31
+ "type:Compose(ImageQ,TextQ)": {
32
+ "em": 14.285714285714286,
33
+ "f1": 14.285714285714286,
34
+ "n": 7
35
+ },
36
+ "type:Compose(TextQ,ImageListQ)": {
37
+ "em": 0.0,
38
+ "f1": 0.0,
39
+ "n": 12
40
+ },
41
+ "type:ImageListQ": {
42
+ "em": 0.0,
43
+ "f1": 9.3125,
44
+ "n": 32
45
+ },
46
+ "type:ImageQ": {
47
+ "em": 44.680851063829785,
48
+ "f1": 46.1063829787234,
49
+ "n": 47
50
+ },
51
+ "type:TextQ": {
52
+ "em": 28.658536585365855,
53
+ "f1": 35.43292682926829,
54
+ "n": 164
55
+ }
56
+ },
57
+ "input_sha256": "098c9d84dca21b07da69903ff247c5af18232b5c8a66ff157cef35006e1e9da1",
58
+ "metric_scale": "0-100",
59
+ "n_expected": 266,
60
+ "n_missing": 0,
61
+ "n_predictions": 266,
62
+ "parse_success_count": 266,
63
+ "prediction_sha256": "5a2141b3cfc56a404497ec147a7a9b73fe4cc91d5ec17366a174d44cf9f52a2e",
64
+ "protocol_sha256": "ca6b0397b6532489a12ad40a3a1d18408c0f52c45a20c15fae188d3a59eae462",
65
+ "scores": {
66
+ "em": 25.93984962406015,
67
+ "f1": 31.48872180451128
68
+ },
69
+ "split": "val_calibration",
70
+ "status_counts": {
71
+ "ok": 266
72
+ },
73
+ "target_sha256": "fe1557c98813401676c470515634238527d397080aee9f96f09c30a47e5fe907"
74
+ },
75
+ {
76
+ "complete": true,
77
+ "dataset": "mmqa_ti",
78
+ "evaluator_sha256": "1b95323ffdf36a0e8a8afef28c598e6fd560f6e63aa8d48709e51607e84d6100",
79
+ "groups": {
80
+ "modality:image": {
81
+ "em": 23.918575063613233,
82
+ "f1": 26.1323155216285,
83
+ "n": 393
84
+ },
85
+ "modality:image+text": {
86
+ "em": 20.338983050847457,
87
+ "f1": 21.889830508474578,
88
+ "n": 118
89
+ },
90
+ "modality:text": {
91
+ "em": 26.829268292682926,
92
+ "f1": 31.882926829268293,
93
+ "n": 820
94
+ },
95
+ "type:Compare(Compose(TableQ,ImageQ),Compose(TableQ,TextQ))": {
96
+ "em": 4.545454545454546,
97
+ "f1": 4.545454545454546,
98
+ "n": 22
99
+ },
100
+ "type:Compose(ImageQ,TextQ)": {
101
+ "em": 45.45454545454545,
102
+ "f1": 45.45454545454545,
103
+ "n": 33
104
+ },
105
+ "type:Compose(TextQ,ImageListQ)": {
106
+ "em": 11.475409836065573,
107
+ "f1": 14.475409836065573,
108
+ "n": 61
109
+ },
110
+ "type:ImageListQ": {
111
+ "em": 3.125,
112
+ "f1": 7.83125,
113
+ "n": 160
114
+ },
115
+ "type:ImageQ": {
116
+ "em": 38.197424892703864,
117
+ "f1": 38.69957081545064,
118
+ "n": 233
119
+ },
120
+ "type:Intersect(ImageListQ,TextQ)": {
121
+ "em": 50.0,
122
+ "f1": 50.0,
123
+ "n": 2
124
+ },
125
+ "type:TextQ": {
126
+ "em": 26.829268292682926,
127
+ "f1": 31.882926829268293,
128
+ "n": 820
129
+ }
130
+ },
131
+ "input_sha256": "bb92409d451ad352f22a819754ff924c2c92bbfec79df241aa382897095badad",
132
+ "metric_scale": "0-100",
133
+ "n_expected": 1331,
134
+ "n_missing": 0,
135
+ "n_predictions": 1331,
136
+ "parse_success_count": 1323,
137
+ "prediction_sha256": "002910799a30c04efdd241f3808e14fc16c6b0bd26b60cf26d1c44485f9e5da3",
138
+ "protocol_sha256": "ca6b0397b6532489a12ad40a3a1d18408c0f52c45a20c15fae188d3a59eae462",
139
+ "scores": {
140
+ "em": 25.39444027047333,
141
+ "f1": 29.29902329075883
142
+ },
143
+ "split": "test",
144
+ "status_counts": {
145
+ "ok": 1331
146
+ },
147
+ "target_sha256": "f9ac83d22210f0300be901ec863f8233e4438dcd99660a7ae628d7d9a290c5a8"
148
+ }
149
+ ],
150
+ "seed": 17,
151
+ "test_feedback_for_training": false,
152
+ "test_used": true,
153
+ "validation_run": "experiments/runs/gate-search-mira-mmqa_ti-transfer_p2_r1-s17"
154
+ }