Upload benchmark_results.json with huggingface_hub
Browse files- benchmark_results.json +62 -0
benchmark_results.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"MLP-v0.4": {
|
| 3 |
+
"tier_f1_weighted": 0.8427272727272727,
|
| 4 |
+
"tier_f1_per_class": {
|
| 5 |
+
"low": 0.9318181818181818,
|
| 6 |
+
"medium": 0.0,
|
| 7 |
+
"high": 0.75,
|
| 8 |
+
"critical": 0.0
|
| 9 |
+
},
|
| 10 |
+
"tier_accuracy": 0.88,
|
| 11 |
+
"score_mae": 0.08412909507751465,
|
| 12 |
+
"score_mse": 0.03132753446698189,
|
| 13 |
+
"ece": 0.08948144972324372,
|
| 14 |
+
"critical_to_low": 3,
|
| 15 |
+
"n_samples": 50
|
| 16 |
+
},
|
| 17 |
+
"XGBoost": {
|
| 18 |
+
"tier_f1_weighted": 0.8517241379310344,
|
| 19 |
+
"tier_f1_per_class": {
|
| 20 |
+
"low": 0.9425287356321839,
|
| 21 |
+
"medium": 0.0,
|
| 22 |
+
"high": 0.75,
|
| 23 |
+
"critical": 0.0
|
| 24 |
+
},
|
| 25 |
+
"tier_accuracy": 0.88,
|
| 26 |
+
"score_mae": 0.08173864334821701,
|
| 27 |
+
"score_mse": 0.02230011485517025,
|
| 28 |
+
"ece": 0.07109888374805451,
|
| 29 |
+
"critical_to_low": 3,
|
| 30 |
+
"n_samples": 50
|
| 31 |
+
},
|
| 32 |
+
"qwen2.5:7b": {
|
| 33 |
+
"tier_f1_weighted": 0.08290196078431372,
|
| 34 |
+
"tier_f1_per_class": {
|
| 35 |
+
"low": 0.08888888888888889,
|
| 36 |
+
"medium": 0.058823529411764705,
|
| 37 |
+
"high": 0.0,
|
| 38 |
+
"critical": 0.11764705882352941
|
| 39 |
+
},
|
| 40 |
+
"tier_accuracy": 0.08,
|
| 41 |
+
"score_mae": 0.6142643094062805,
|
| 42 |
+
"score_mse": 0.42907100915908813,
|
| 43 |
+
"ece": 0.7952000546455383,
|
| 44 |
+
"critical_to_low": 1,
|
| 45 |
+
"n_samples": 50
|
| 46 |
+
},
|
| 47 |
+
"llama3-german-instruct": {
|
| 48 |
+
"tier_f1_weighted": 0.1360783752860412,
|
| 49 |
+
"tier_f1_per_class": {
|
| 50 |
+
"low": 0.13043478260869565,
|
| 51 |
+
"medium": 0.0625,
|
| 52 |
+
"high": 0.3157894736842105,
|
| 53 |
+
"critical": 0.0
|
| 54 |
+
},
|
| 55 |
+
"tier_accuracy": 0.14,
|
| 56 |
+
"score_mae": 0.5682483315467834,
|
| 57 |
+
"score_mse": 0.35942625999450684,
|
| 58 |
+
"ece": 0.7409999561309815,
|
| 59 |
+
"critical_to_low": 1,
|
| 60 |
+
"n_samples": 50
|
| 61 |
+
}
|
| 62 |
+
}
|