iFaz/whisper-SER-base-v3

first run with
iFaz/Whisper_Compatible_SER_benchmark
eavl: skip_special_token = True

rerun with this dataset
iFaz/Whisper_Compatible_SER_benchmark
eavl: skip_special_token = False             <<<<<---------------

V3 (rerun of base-v1)
################################################

Dataset: test_CREMA_D Dataset

Number of Samples: 2000
Overall WA: 0.5490
Overall UA: 0.5604
Micro F1 Score: 0.5490
Macro F1 Score: 0.4853

Classification Report:

              precision    recall  f1-score   support

           0       0.38      0.93      0.54       336
           1       1.00      0.04      0.08       346
           2       0.85      0.80      0.82       344
           3       0.65      0.90      0.76       288
           4       0.83      0.09      0.16       344
           5       0.52      0.61      0.56       342

    accuracy                           0.55      2000
   macro avg       0.71      0.56      0.49      2000
weighted avg       0.71      0.55      0.48      2000

Confusion Matrix:

[[312   0   8  12   4   0]
 [208  14  24  30   0  70]
 [ 56   0 276  10   0   2]
 [ 22   0   2 258   0   6]
 [174   0  10  16  30 114]
 [ 58   0   6  68   2 208]]

################################################

Dataset: test_ravdess Dataset

Number of Samples: 1200
Overall WA: 0.6333
Overall UA: 0.6617
Micro F1 Score: 0.6333
Macro F1 Score: 0.6463

Classification Report:

              precision    recall  f1-score   support

           0       0.32      0.97      0.48       160
           1       1.00      0.28      0.43       160
           2       1.00      0.68      0.81       160
           3       0.96      0.52      0.68       400
           4       0.90      0.88      0.89       160
           5       0.55      0.65      0.60       160

    accuracy                           0.63      1200
   macro avg       0.79      0.66      0.65      1200
weighted avg       0.82      0.63      0.65      1200

Confusion Matrix:

[[156   0   0   4   0   0]
 [116  44   0   0   0   0]
 [ 48   0 108   4   0   0]
 [108   0   0 208   0  84]
 [ 20   0   0   0 140   0]
 [ 40   0   0   0  16 104]]

################################################

Dataset: test_IEMOCAP Dataset

Number of Samples: 1000
Overall WA: 0.4790
Overall UA: 0.4360
Micro F1 Score: 0.4790
Macro F1 Score: 0.4421

Classification Report:

              precision    recall  f1-score   support

           0       0.53      0.53      0.53       370
           2       1.00      0.16      0.27       167
           3       0.37      0.70      0.48       269
           5       0.75      0.37      0.49       194

    accuracy                           0.48      1000
   macro avg       0.66      0.44      0.44      1000
weighted avg       0.61      0.48      0.46      1000

Confusion Matrix:

[[195   0 166   9]
 [ 56  26  82   3]
 [ 70   0 187  12]
 [ 50   0  73  71]]

################################################