Update README.md

README.md

@@ -34,6 +34,47 @@ More information needed
 
 More information needed
 
+# Test results
+Binary classification accuracy (threshold at label 3) ≈ 73.45%
+
+Test Report:
+```
+              precision    recall  f1-score   support
+
+           0       0.73      0.44      0.55       100
+           1       0.25      0.30      0.28       100
+           2       0.22      0.47      0.30       100
+           3       0.26      0.33      0.29       100
+           4       0.40      0.14      0.21       100
+           5       1.00      0.02      0.04        50
+
+    accuracy                           0.31       550
+   macro avg       0.48      0.28      0.28       550
+weighted avg       0.43      0.31      0.30       550
+```
+
+Confusion Matrix:
+```
+[[44 32 22  2  0  0]
+ [13 30 47 10  0  0]
+ [ 3 33 47 17  0  0]
+ [ 0 14 44 33  9  0]
+ [ 0  5 36 45 14  0]
+ [ 0  4 14 19 12  1]]
+```
+Test metrics
+```
+  epoch                   =       20.0
+  eval_accuracy           =     0.3073
+  eval_f1_macro           =     0.2779
+  eval_loss               =      1.559
+  eval_precision          =     0.4789
+  eval_recall             =     0.2833
+  eval_runtime            = 0:00:04.86
+  eval_samples_per_second =    113.008
+  eval_steps_per_second   =      3.698
+```
+
 ## Training and evaluation data
 
 More information needed