Add eval artifacts (test)

eval/classification_report_test.csv

@@ -1,7 +1,7 @@
 ,precision,recall,f1-score,support
-LABEL_0,0.9577922077922078,0.9833333333333333,0.9703947368421053,300.0
-LABEL_1,0.9735099337748344,0.9767441860465116,0.9751243781094527,301.0
-LABEL_2,0.9863013698630136,0.9568106312292359,0.9713322091062394,301.0
+normal,0.9577922077922078,0.9833333333333333,0.9703947368421053,300.0
+promo,0.9735099337748344,0.9767441860465116,0.9751243781094527,301.0
+smish,0.9863013698630136,0.9568106312292359,0.9713322091062394,301.0
 accuracy,0.9722838137472284,0.9722838137472284,0.9722838137472284,0.9722838137472284
 macro avg,0.9725345038100186,0.972296050203027,0.9722837746859324,902.0
 weighted avg,0.9725508478189075,0.9722838137472284,0.9722858689629212,902.0

eval/confusion_matrix_test.csv

@@ -1,4 +1,4 @@
-,P_LABEL_0,P_LABEL_1,P_LABEL_2
-T_LABEL_0,295,4,1
-T_LABEL_1,4,294,3
-T_LABEL_2,9,4,288
+,P_normal,P_promo,P_smish
+T_normal,295,4,1
+T_promo,4,294,3
+T_smish,9,4,288

eval/metrics_test.json

@@ -1,6 +1,6 @@
 {
   "split": "test",
-  "timestamp": "2025-09-09 20:25:34",
+  "timestamp": "2025-09-09 20:46:58",
   "num_examples": 902,
   "accuracy": 0.9722838137472284,
   "precision_weighted": 0.9725508478189075,