----- report -----

       precision    recall  f1-score   support

           0       0.98      0.99      0.98  26181947
           .       0.91      0.92      0.92   1382956
           ,       0.82      0.72      0.77   1347569
           ?       0.76      0.75      0.75    116022
           -       0.82      0.60      0.69    199077
           :       0.81      0.73      0.76    151017

    accuracy                           0.97  29378588
   macro avg       0.85      0.78      0.81  29378588
weighted avg       0.97      0.97      0.97  29378588


----- confusion matrix -----

     t/p      0     .     ,     ?     -     :
        0   1.0   0.0   0.0   0.0   0.0   0.0
        .   0.1   0.9   0.0   0.0   0.0   0.0
        ,   0.3   0.0   0.7   0.0   0.0   0.0
        ?   0.2   0.1   0.0   0.7   0.0   0.0
        -   0.3   0.0   0.1   0.0   0.6   0.0
        :   0.2   0.0   0.0   0.0   0.0   0.7

{'eval_loss': 0.09778209030628204, 'eval_f1': 0.8127300278497737, 'eval_precision': 0.8505971964347974, 'eval_recall': 0.7825453583600391, 'eval_accuracy': 0.9672826685884292, 'eval_runtime': 228.8368, 'eval_samples_per_second': 251.734, 'eval_steps_per_second': 31.468, 'epoch': 2.0}
{'train_runtime': 9212.1556, 'train_samples_per_second': 62.224, 'train_steps_per_second': 7.778, 'train_loss': 0.10716916574063842, 'epoch': 2.0}