选项平衡后的第一次试跑，约70%正确率

2025-06-02 17:18:30 +08:00
parent 7a725bc003
commit 3984ec002e
12 changed files with 16742 additions and 0 deletions
--- a/results/20250602_1706/claude-sonnet-4-20250514_metrics.json
+++ b/results/20250602_1706/claude-sonnet-4-20250514_metrics.json
@@ -0,0 +1,12 @@
+{
+  "timestamp": "2025-06-02T17:15:23.726253",
+  "metrics": {
+    "accuracy": 0.700507614213198,
+    "precision_micro": 0.6934673366834171,
+    "recall_micro": 0.700507614213198,
+    "f1_micro": 0.696969696969697,
+    "precision_macro": 0.7072180484244438,
+    "recall_macro": 0.7009183673469388,
+    "f1_macro": 0.69833034513671
+  }
+}