Upload eval/ablation_results.json with huggingface_hub
Browse files- eval/ablation_results.json +32 -0
eval/ablation_results.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"full_v10": {
|
| 3 |
+
"success": 0.718,
|
| 4 |
+
"avg_cost": 0.016932746258400005,
|
| 5 |
+
"costRed": 94.65337977316072
|
| 6 |
+
},
|
| 7 |
+
"no_feedback": {
|
| 8 |
+
"success": 0.632,
|
| 9 |
+
"avg_cost": 0.014239461958399993,
|
| 10 |
+
"costRed": 95.50380108670666
|
| 11 |
+
},
|
| 12 |
+
"no_cascade": {
|
| 13 |
+
"success": 0.632,
|
| 14 |
+
"avg_cost": 0.014239461958399993,
|
| 15 |
+
"costRed": 95.50380108670666
|
| 16 |
+
},
|
| 17 |
+
"heuristic": {
|
| 18 |
+
"success": 0.57,
|
| 19 |
+
"avg_cost": 0.04103160790880004,
|
| 20 |
+
"costRed": 87.04401392207134
|
| 21 |
+
},
|
| 22 |
+
"always_frontier": {
|
| 23 |
+
"success": 0.782,
|
| 24 |
+
"avg_cost": 0.3166872804999999,
|
| 25 |
+
"costRed": 0.004016261446193603
|
| 26 |
+
},
|
| 27 |
+
"always_cheap": {
|
| 28 |
+
"success": 0.632,
|
| 29 |
+
"avg_cost": 0.014239461958399993,
|
| 30 |
+
"costRed": 95.50380108670666
|
| 31 |
+
}
|
| 32 |
+
}
|