context-loop-engineering / audit /audit_results_v2.json
thanatchon36's picture
Publish data for v0.1.0
c8b99a6 verified
Raw History Blame Contribute Delete
2.96 kB
{
"round1_v2_in_sample": {
"layer": {
"kappa": 0.86,
"agreement": 0.9,
"n": 100,
"labels": [
"context",
"evaluation",
"harness",
"loop"
],
"confusion_rule_rows_manual_cols": [
[
25,
1,
0,
0
],
[
0,
11,
1,
1
],
[
0,
0,
35,
4
],
[
0,
0,
3,
19
]
]
},
"contribution": {
"kappa": 0.781,
"agreement": 0.84,
"n": 100,
"labels": [
"benchmark",
"empirical",
"method",
"practice_guide",
"survey",
"tool"
],
"confusion_rule_rows_manual_cols": [
[
4,
0,
0,
3,
0,
0
],
[
0,
2,
1,
0,
0,
0
],
[
3,
2,
30,
0,
3,
0
],
[
0,
0,
0,
14,
0,
2
],
[
0,
0,
0,
0,
2,
1
],
[
1,
0,
0,
0,
0,
32
]
]
}
},
"round2_v2_held_out": {
"layer": {
"kappa": 0.856,
"agreement": 0.9,
"n": 60,
"labels": [
"context",
"evaluation",
"harness",
"loop"
],
"confusion_rule_rows_manual_cols": [
[
14,
0,
0,
1
],
[
0,
6,
1,
0
],
[
0,
1,
24,
2
],
[
0,
0,
1,
10
]
]
},
"contribution": {
"kappa": 0.764,
"agreement": 0.833,
"n": 60,
"labels": [
"benchmark",
"empirical",
"method",
"practice_guide",
"survey",
"tool"
],
"confusion_rule_rows_manual_cols": [
[
1,
0,
0,
0,
0,
0
],
[
0,
1,
0,
0,
0,
0
],
[
3,
1,
15,
0,
1,
0
],
[
0,
0,
0,
7,
0,
2
],
[
0,
1,
0,
0,
2,
0
],
[
2,
0,
0,
0,
0,
24
]
]
}
},
"pooled_v2_layer": {
"kappa": 0.859,
"agreement": 0.9,
"n": 160
}
}